(Dan M. Frangopol (Ed.), Richard Sause (Ed.), Chad PDF

BRIDGE MAINTENANCE, SAFETY, MANAGEMENT AND LIFE-CYCLE OPTIMIZATION
PROCEEDINGS OF THE FIFTH INTERNATIONAL CONFERENCE ON BRIDGE MAINTENANCE,

SAFETY AND MANAGEMENT, PHILADELPHIA, PENNSYLVANIA, USA, 1115 JULY 2010
Bridge Maintenance, Safety,

Management and Life-Cycle
Optimization
Edited by
Dan M. Frangopol
ATLSS Engineering Research Center, Lehigh University, Bethlehem, PA, USA
Richard Sause
Chad S. Kusko
CRC Press/Balkema is an imprint of the Taylor & Francis Group, an informa business
2010 Taylor & Francis Group, London, UK
Typeset by MPS Ltd. (A Macmillan Company), Chennai, India

Printed and bound in USA by Edwards Brothers, Inc, Lillington, NC
All rights reserved. No part of this publication or the information contained herein may be reproduced,
stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, by
photocopying, recording or otherwise, without written prior permission from the publisher.
Although all care is taken to ensure integrity and the quality of this publication and the information herein,
no responsibility is assumed by the publishers nor the author for any damage to the property or
persons as a result of operation or use of this publication and/or the information contained herein.
Published by: CRC Press/Balkema

P.O. Box 447, 2300 AK Leiden, The Netherlands
e-mail: [email protected]
www.crcpress.com www.taylorandfrancis.co.uk www.balkema.nl
ISBN: 978-0-415-87786-2
Bridge Maintenance, Safety, Management and Life-Cycle Optimization Frangopol, Sause & Kusko (eds)
2010 Taylor & Francis Group, London, ISBN 978-0-415-87786-2
Table of Contents
Preface XXXIII
Conference organization XXXV
T.Y. Lin Lecture
Bridge maintenance and safety: A practitioners view 3

C. Seim
Keynote Lectures
Fundamentals of suspension bridge retrofit 11

P.G. Buckland
Modern bridges in China 17
M. Feng
Cable supported bridges design, maintenance, rehabilitation, and management 18
J.S. Jensen
Lifetime design of cable-supported super-long-span bridges 26
H.M. Koh, H.J. Kim, J.H. Lim, S.C. Kang & J.F. Choo
Managing old bridges 43
P.G. Malerba
The evolving science of terrorist threat mitigation for bridges 53
J.C. Ray
Preserving the asset for its intended use 57
R.A.P. Sweeney
On longevity and monitoring technologies of bridges a survey study by Japanese Society of
Steel Construction 64
E. Watanabe, H. Furuta, T. Yamaguchi & M. Kano
Wireless sensing technologies for bridge monitoring and assessment 72
C.-B. Yun, H. Sohn, H.J. Jung, B.F. Spencer & T. Nagayama
Mini-Symposia
MS1: Futuristic bridge maintenance technologies
Organizers: C.-B. Yun & B.F. Spencer Jr.
Passive sensors for monitoring corrosion in concrete bridges 81
A. Abu Yousef, S.L. Wood, P. Pasupathy & D.P. Neikirk
Coupled resonant coil sensors with increased interrogation distance 82
S. Bhadra, E. Thompson, B. Kordi, G.E. Bridges & D.J. Thomson
Concentration-dependent piezoelectricity and strain sensitivity of ZnO
nanoparticle-polymeric thin films 83
D. Chang & K.J. Loh
V
Prediction of displacement response of a suspension bridge using FBG strain sensors 84
S.J. Chang, N.S. Kim & H.K. Kim
Structural health monitoring of a cable-stayed bridge using acceleration data via
wireless smart sensor network 85
S. Cho, J. Park, H.-J. Jung, C.-B. Yun, S. Jang, H. Jo, B.F. Spencer Jr., T. Nagayama & J.W. Seo
Smart wireless tension force monitoring system for stay cables 86
S. Cho, C.-B. Yun & J.P. Lynch
An active sensor placement optimization strategy using data-driven Bayesian experimental design 87
E.B. Flynn & M.D. Todd
Investigation of the control performance of the smart passive system based on MR damper
using hybrid simulation 88
D.D. Jang, I.H. Kim, H.J. Jung & J.H. Koo
Autonomous structural health monitoring using wireless smart sensors on a cable-stayed bridge 89
S. Jang, H. Jo, K. Mechitov, S.-H. Sim, B.F. Spencer Jr., G. Agha, S. Cho,
H.-J. Jung, C.-B. Yun & J.A. Rice
Evaluation of ultimate performance of the reinforced concrete T-girder bridge
using optical fiber sensors 90
D.-S. Jung, C.-Y. Kim, S.-M. Cho, S.-A. Choi, B.-Y. Seo & K.-W. Lee
Statistical damage assessment based on the extreme value distribution using vibration responses 91
J. Kang & H.W. Park
Prestress-force monitoring using impedance-based smart sensor nodes in PSC girder bridges 92
J.-T. Kim, D.-S. Hong, S.-Y. Lee & J.-H. Park
Tension monitoring of a prestressing strand for concrete bridge using in-tendon FBG sensors 93
J.-M. Kim, H.-W. Kim, Y.-S. Kim & Y.-H. Park
Evaluation of the characteristics of local erosion of fine-grained soils in the west coast area of Korea 94
J. Lee, K. Kwak, J. Park & M. Chung
Impedance-based structural health monitoring using neural networks for autonomous
frequency range selection 95
J. Min, C.-B. Yun & S. Park
Output only modal identification and damage detection of bridge-structures using time
frequency and wavelet techniques 96
S. Nagarajaiah, B. Basu & V.S.M. Vemuru
Safety network system integration for bridge structures in Korea 97
K.T. Park, Y.J. Yoo, B.C. Joo, J. Lee & Y.K. Hwang
Damping properties identified from wind-induced vibration measurements of a suspension bridge 98
D.U. Park, N.S. Kim & H.K. Kim
Debonding condition monitoring of a CFRP laminated concrete beam using piezoelectric
impedance sensor nodes 99
S. Park, S.-K. Park, J.-W. Kim & H.-J. Chang
Development of a wireless power and data transmission system using laser and optoelectronic
devices for guided wave-based structural health monitoring 100
H.J. Park, H. Sohn, C.-B. Yun, J. Chung & I.B. Kwon
Development of structural health monitoring systems for railroad bridge testbeds 101
H.J. Park, C.-B. Yun, M.H. Shin, J.J. Lee & S.Y. Park
Development of a benchmark laboratory structure for finite-element model updating 102
S. Shang, G.J. Yun, S.G. Lee, J. Caicedo & S. Narasimhan
Variation of eigen-properties of a PSC bridge due to prestressing force 103
S. Shin, M.-S. Koo, H.-K. Lee & S.-J. Kwon
VI
Imaging ultrasonic waves in complex structures using a scanning laser Doppler vibrometer 104
H. Sohn, J.Y. Yang, D. Dutta, M.P. DeSimio, S.E. Olson & E.D. Swenson
Integrated wireless powering and data interrogation for civil infrastructure monitoring 105
S.G. Taylor, E.B. Flynn, D.L. Mascarenas, M.D. Todd, D. Dondi, T. Rosing, S. Kpotufe,
S. Dasgupta, K. Lin, R. Gupta, K.M. Farinholt, G. Park & C.R. Farrar
Multi-scale wireless sensor node for impedance-based SHM and long-term civil
infrastructure monitoring 106
S.G. Taylor, K.M. Farinholt, G. Park, C.R. Farrar & M.D. Todd
Energy harvesting and wireless energy transmission for powering SHM sensor nodes 107
S.G. Taylor, N.A. Miller, K.M. Farinholt, G. Park & C.R. Farrar
Feasibility investigation for identifying bridges fundamental frequencies from vehicle vibrations 108
T. Toshinami, M. Kawatani & C.W. Kim
Analysis and prediction for bridge maintenance costs based on life-cycle and Markov approach 109
Y.Q. Xiang, C. Zhou & D.M. Guo
Transmissibility-function-based structural damage detection with tetherless mobile sensors 110
X. Yi, D. Zhu, Y. Wang, J. Guo & K.-M. Lee
Long-term structural health monitoring for Tamar Suspension Bridge 111
K.Y. Koo, J.M.W. Brownjohn, P. Carden, D.I. List, R. Cole & T. Wood
International collaborative research and education on smart sensors and monitoring technologies 112
C.-B. Yun, H. Sohn, M.L. Wang, B.F. Spencer & Y. Fujino
Piezo paint-based smart tape sensor for bridge diagnosis 113
Y. Zhang, C. Zhou, Z. Li, C.C. Fu & C. Wang
MS2: Monitoring & assessment of bridges using novel techniques

Organizers: A. Strauss & D.M. Frangopol
Analysis of the structural response to a moving load using empirical mode decomposition 117
M. Bradley, A. Gonzlez & D. Hester
Measurement tool kit for corrosion and defect monitoring of bridge tendons 118
H. Budelmann, A. Holst & H.-J. Wichmann
Field application of innovative scour monitoring techniques for bridges 119
K.C. Chang, C.C. Chen, Y.B. Lin, L.S. Lee & S.C. Wong
Performance evaluation of bridge seismic bearings based on in-situ quick-release tests 120
S.S. Chen & I.-S. Ahn
Experimental assessment of prestressing force in concrete bridges 121
J. Ciesla
Assessment and strengthening of prestressed damaged beams 122
L. Della Sala, R. Cerone & M. Franzese
Development of rapid, reliable, and economical methods for inspection and
monitoring of highway bridges 123
J.D. Fasl, T.A. Helwig, S.L. Wood, V.A. Samaras, A.A. Yousef, K.H. Frank,
D.L. Potter & R.E. Lindenberg
Wireless sensor performance monitoring of an innovative bridge design in New York State 124
M.V. Gangone, M.J. Whelan, K.D. Janoyan, L. Minnetyan & T. Qiu
The benefit of monitoring for bridge maintenance 125
R. Geier
Monitoring method for curved concrete bridge girders using long-gauge deformation sensors 126
B. Gliic, D. Inaudi & D. Posenato
Multiple sensor subsurface condition assessment of reinforced concrete bridge decks 127
D.R. Huston, J. Cui, D. Burns, D. Hurley & R. Arndt
VII
Output-only substructural identification for local damage detection 128
C.G. Koh & T.N. Thanh
Steel beam fatigue life prediction using acoustic emission amplitude histograms and
backpropagation neural networks 129
A. Korcak, J. Suleman, F.F. Barsoum & E.v.K. Hill
Health monitoring of concrete bridges: model simulations of pre-stressed beams under
static environmental loading based on experimental data 130
F. Lanata
Jointless bridge: Determination of fracture mechanical parameters values for nonlinear analysis 131
D. Lehk, Z. Kerner & D. Novk
Coupled field monitoring and structural analysis to assess scour conditions 132
J. McConnell & M. Cann
Proof load testing supported by acoustic emission. An example of application 133
P. Olaszek, G. Swit & J.R. Casas
Jointless bridge: Reliability assessment 134
J. Podrouek, D. Novk & A. Strauss
An innovative approach for dynamic damage detection in bridge girders 135
H.A. Rasheed & Sh. Nayyeri Amiri
Definition of structural parameters span structures of bridges by results of their
tests by mobile loading 136
V. Redchenko
Repair monitoring and experimental work associated with Ferrycarrig Bridge 137
P.C. Ryan, A.J. OConnor, L. Duffy, A. Daly & G. Jones
Analysis of in-service data collected during biennial inspections on typical bridges 138
H.W. Shenton III, K. Connor, M.J. Chajes, M. Rakowski & B. Brookes
Determination of concrete bridge ageing by structural health monitoring 139
S. Soyoz & M.Q. Feng
Monitoring based performance checking of the jointless Marktwasser Bridge S33.24 140
A. Strauss, R. Wendner, K. Bergmeister & D.M. Frangopol
Monitoring based verification of the soil structure interaction of the Markwasser Bridge S33.24 141
R. Wendner, A. Strauss & K. Bergmeister
Experimental trials on the detection of reinforcement breaks with the magnetic flux leakage method 142
T. Wolf & T. Vogel
Innovative structural health monitoring for Tamar Suspension Bridge by automated
total positioning system 143
K.Y. Koo, J.M.W. Brownjohn, D. List, R. Cole & T. Wood
MS3: Present & future of bridge inspection & evaluation

Organizers: S. Alampalli, A.K. Agrawal & M. Ettouney
Inspection needs of deteriorating bridge components 147
A.K. Agrawal & S. Alampalli
Periodic NDE in support of structural health monitoring of bridges 148
R.W. Arndt, F. Jalinoos, J. Cui & D. Huston
To see is to know: visualization in bridge inspection and management 149
J. Bien, M. Kuzawa & B. Bien
Knowledge-based expert tools in bridge management 150
J. Bien, A. Banakiewicz, M. Gadysz, T. Kaminski, P. Rawa & B. Bien
VIII
Enhanced bridge management via integrated remote sensing 151
S. Chen, E. Hauser, K. Dai, W. Liu, B. Ribarsky, S. Lee, B. Tolone & C. Boyle
Full field mapping of bridge deformation using digital speckle photography 152
F.P. Chiang & J. Yu
Design for inspection one way towards durable infrastructures 153
M. Drissi-Habti
Non-homogeneous Markov Chain for bridge deterioration modeling 154
G. Fu & D. Devaraj
New tools for inspection and evaluation of steel truss bridge gusset plates 155
C. Higgins, Q.D. Nguyen & O.T. Turan
The computerized cables stay inspection of the Cooper River Bridge (Arthur Ravanel, Jr) 156
B. Kroely, G. Hovhanessian & J. Stieb
Load ratings of a concrete-encased steel 3-hinge arch bridge 157
L.-Y. Lai, W.-S. Chang & M. Majd
Rope access bridge inspections 158
B. Leshko
California bridge management 159
B. Newton
Post-earthquake bridge inspection guidelines for New York state 160
J.S. OConnor & S. Alampalli
A framework for evaluating the impact of structural health monitoring on bridge management 161
M. Pozzi, D. Zonta, W. Wang & G. Chen
Bridge management and inspection for the County of Baltimore, Maryland 162
J.K. Shaffer & M.C. Schellhase
Overview of remote structural health monitoring: Focus on end user demands 163
T. Spuler, G. Moor & R. Berger
Methods of quality control and quality assurance for highway bridge inspection in the United States 164
G.A. Washer
Studies on the use of KEEL software for intelligent analyzing of bridge load-bearing capacity 165
P. Wjtowicz & D. Krl
MS4: Research & applications in bridge health monitoring

Organizers: F.N. Catbas, J.R. Casas, H. Furuta & D.M. Frangopol
Design and implementation of load cell bearings to measure dead and live load effects
in an aged long span bridge 169
N. Dubbs, F. Moon & A.E. Aktan
Maintenance monitoring system design of a movable bridge 170
F.N. Catbas, H.B. Gokce, T. Dumlupinar, R. Zaurin, M. Gul, T. Terrell, D. Harper & D.M. Frangopol
Exploring indirect vehicle-bridge interaction for bridge SHM 171
F. Cerda, J. Garrett, J. Bielak, R. Bhagavatula & J. Kovacevic
Novelty detection based on symbolic data analysis applied to structural health monitoring 172
A. Cury & C. Cremona
A temporal multi-scale analysis based bridge health divide-and-conquer monitoring
and diagnosis strategy 173
D. Dan, M. Zhihua & S. Limin
Validation of an SHM procedure for concrete bridges based on static strain records 174
A. Del Grosso, F. Lanata, D. Posenato & A. Mercalli
IX
Uncertainty and reliability analysis using monitoring data and artificial neural
network (ANN) calibration 175
T. Dumlupinar, H.B. Gokce, K.R. Mackie, F.N. Catbas & D.M. Frangopol
Effects of environmental changes on the dynamic characteristic of reinforced concrete beams 176
M.R. Esfahani & G. Behnam
Damage detection using a novel time series methodology: Application to the Z24 Bridge data 177
M. Gul & F.N. Catbas
Development of a bridge health monitoring approach using train-bridge interaction
analysis and GA optimization 178
X. He, T. Hayashikawa, T. Matsumoto, M. Kawatani & H. Furuta
Assessment of remaining fatigue life of aging orthotropic steel deck bridges 179
K. Kashefi, M. Zeinoddini & A.P. Zandi
Identifying bending stiffness change of a beam under a moving vehicle 180
C.W. Kim, M. Kawatani & T. Fujimoto
Low-cost wireless sensor node for vibration monitoring of infrastructures 181
C.W. Kim, M. Kawatani, R. Ozaki, N. Makihata & M. Kano
Bridge retrofit design optimization for fatigue based on monitoring and FE analysis 182
K. Kwon & D.M. Frangopol
Updating the finite element model of bridge structures by an improved Taguchi updating method 183
Y. Liu, H. Li, Z. Duan & Y. Yao
High-order local vibration properties of RC Viaduct under the passing high speed train 184
K. Matsuoka, K. Kaito, T. Watanabe & M. Sogabe
Condition assessment of bridge deck truss using in-service monitoring data of strain 185
Y.Q. Ni, H.W. Xia, J.M. Ko & K.Y. Wong
Simultaneous monitoring of the coupled vibration between a bridge and moving trains 186
Y. Oshima, K. Yamamoto, K. Sugiura, A. Tanaka & M. Hori
Challenges and uncertainty mitigation in structural identification of long span bridges 187
J.B. Prader, J. Zhang, F.L. Moon & A.E. Aktan
Bridge instrumentation for long term structural health monitoring 188
M. Sanayei, J.D. Sipple, J.E. Phelps, E. Santini-Bell, P.J. Lefebvre & B.R. Brenner
Modeling and instrumentation of the Tobin Memorial Bridge 189
M. Sanayei, E.J. Pheifer, B.R. Brenner, E. Santini-Bell & W.L. Durack
Instrumentation for reinforced concrete durability monitoring of Qingdao Bay Bridge 190
X.P. Shao, B.L. Guo & Y.Q. Ni
Structural monitoring of Lezria Bridge since its construction 191
H. Sousa, J. Figueiras & J. Bento
Reliability assessment oriented monitoring system design for Shanghai Yangtze River Bridge 192
Z. Sun, H.-S. Wu, Q.-M. Wang, Y.-C. Lu & Z.-F. Zhou
Bridge Sensor Mart: A flexible and scalable data storage and analysis framework for
structural health monitoring 193
N.C. Tas, C. Raileanu, M. Dejori & C. Neubauer
Full-scale measurements on buffeting response of Sutong Bridge under typhoon Fung-Wong 194
H. Wang, A.Q. Li, T.Y. Zhu & R.M. Hu
Structural health monitoring for damage detection based on integration of computer imaging
and sensor data 195
R. Zaurin & F.N. Catbas
X
MS5: New procedures for bridge rehabilitation
Organizer: V. Popa
Sustainable and cost effective solutions to life extension of bridges 199
C.P. Atkins, P. Lambert, R. Brueckner, R. Merola & A.R. Foster
Electrochemical chloride extraction and electrochemical re-alkalization Foreva
Regebeton PA process 200
R.A. Bedn & D. Deschamps
Bridge crossings raised to provide more overhead clearance 201
R.A. Daniel, B. van Sinten & H.T. van Manen
Choosing the appropriate sustainable polymer concrete material for bridge
preservation and maintenance 202
A.M. Dinitz & M.S. Stenko
Shear resistances and strengthening of aged prestressed concrete bridges considering
deterioration and fatigue effects 203
P. Mark, M. Bender, M. Strack, Ch. Fust & V. Birtel
Fatigue strength and repair methods of corroded bridge wires 204
S. Nakamura & K. Suzumura
MS6: Measurement systems for bridge weigh-in-motion (B-WIM)

Organizers: B. Bakht, A. Znidaric & D.K. McNeill
Bridge capacity assessment by combined proof-loading and WIM data 207
J.R. Casas & J.D. Gmez
Using weigh-in-motion data for modeling maximum live load effects on highway bridges 208
M. Ghosn & B. Sivakumar
Bridge weigh-in-motion on steel orthotropic decks Millau viaduct and Autreville bridge 209
B. Jacob, M. Hannachi & S.S. Ieng
Winnipeg weigh in motion 210
A. Mufti, H. Vogel, E. Murison & B. Bakht
The influence of correlation on the extreme traffic loading of bridges 211
E.J. OBrien, B. Enright & T. Dempsey
A filtered measured influence line approach to bridge weigh-in-motion 212
E.J. OBrien, A. Gonzlez & J. Dowling
Evaluating remaining lifetime of bridges by means of BWIM 213
M. Petschacher
WIM data to assess consequences of new traffic regulations 214
F. Schmidt & B. Jacob
Vehicle loading and effect on the Tsing Ma Bridge using WIM data 215
Y.L. Xu, Z.W. Chen & K.Y. Wong
Latest practical developments in the bridge WIM technology 216
A. nidaric, I. Lavric & J. Kalin
Applications of B-WIM technology to bridge assessment 217
A. nidaric & I. Lavric
MS7: Bridge management tools & research

Organizer: L. Klatter
A methodology to estimate risk related to road links, due to latent processes 221
B.T. Adey, J.D. Birdsall & R. Hajdin
XI
A framework for comprehensive estimation of user costs for bridge management: A synopsis
of existing practices and discussion of new considerations 222
Q. Bai, S. Labi, G.P. Ong, A. Bhargava, K.C. Sinha & P.D. Thompson
Bridge management technique: Implementation of a deterioration model to different
highway networks 223
P. Crespi, D. Oggionni & G. Pasqualato
Structural monitoring with wireless sensor networks: Lessons learned from field deployments 224
G. Feltrin, R. Bischoff, J. Meyer & O. Saukh
An improved model for predicting NBI condition ratings 225
M.J. Fraher, S. Hwang & J. Fu
Bridge maintenance and practical bridge management systems in Japan 226
H. Furuta & E. Watanabe
Contribution of non-destructive defect detection to bridge management 227
R. Helmerich & B. Milmann
Bridge inspections: Are we getting it right? 228
B. Kamya
Maximizing return on investment utilizing a bridge depreciation model 229
H.S. Kleywegt
Trends in development of bridge management systems 230
J.S. Kong, J.H. Kim, B.T. Adey & L. Klatter
The bridge management system of the NYCDOT 231
B. Kroely, K. McAnulty & J. Daza
Integrated bridge inspection and management software for New Jersey Turnpike Authority 232
J.H. Laird, J.A. Paul & J.K. Shaffer
Integration of bridge management systems (BMS) and pavement management systems (PMS) 233
R.M. Ellis, A. Cheetham, K. Hong & P.D. Thompson
Development of a bridge network life cycle cost model 234
T. Reale, A. OConnor, L. Duffy & A. Daly
Implementation of a multi-period bridge investment optimization approach utilizing Pontis
results and additional constraints in three agencies 235
W. Robert, D. Gurenich & R. Thompson
BrIM based bridge operations and management 236
A.M. Shirole, S.S. Chen, Q. Gao & H. Hu
Updating bridge deterioration models with irregular inspection intervals 237
P.D. Thompson, R.M. Ellis & K. Hong
Estimation of enhanced Pontis deterioration models in Florida 238
P.D. Thompson & J.O. Sobanjo
Advanced dynamic testing and bridge management system 239
J. Zwolski, J. Bien & P. Rawa
MS8: Uncertainty in bridge damageability modelling

Organizer: M. Shinozuka
Failure estimation of highway bridges under combined effects of scouring and earthquake 243
A. Alipour, B. Shafei & M. Shinozuka
Deterministic and probabilistic evaluation of time to corrosion initiation for bridges located
in coastal areas 244
XII
Uncertainty in bridge fragility curves and its effect on the seismic risk evaluation of
a highway transportation network 245
S. Banerjee & M. Shinozuka
Sequential bridge fragility updating using long-term monitoring data 246
M. Shinozuka, M. Torbol & S. Banerjee
MS9: Performance-based asset & risk management of the highway

infrastructure system
Organizer: A.E. Aktan
Mitigating infrastructure performance failures through risk-based asset management 249
A.E. Aktan & F.L. Moon
One Team bridge maintenance strategy MACs in the UK 250
A.G. Ardrey & B. Stanforth
TEAMS: Floridas Turnpike Enterprises answer to asset management 251
D.L. Blake
Optimal resource allocation for seismic retrofitting of bridges in transportation networks 252
F. Carturan, C. Pellegrino, C. Modena, R. Rossi & M. Gastaldi
Inspection process of Bulgarian Bridges 253
A. Chaperon, V. Morisseau & B. Kroely
Current state of highway transportation asset management in Germany 254
P. Haardt & J. Krieger
Network level deterioration modeling: A case study on masonry arch bridges 255
M. Imran Rafiq, S. Sathananthan & M.K. Chryssanthopoulos
Road infrastructure asset management a holistic approach to road infrastructure supply 256
J.M. Jansen
A holistic approach to asset management in The Netherlands 257
L. Klatter, J. van der Velde & J. Bakker
Towards sustainable life cycle design of highway bridges 258
Z. Lounis & L. Daigle
Green stormwater infrastructure: A prototypical multi-domain, complex infrastructure system 259
F.A. Montalto & A. Waldman
Managing catastrophic risk to Istanbuls housing infrastructure 260
R.W. Muir, N. Sunger, R. Wittke & P.L. Gurian
System-identification of multi-domain (human, natural, and engineered) infrastructure systems 261
A.R. Pradhan, P.L. Gurian, F.A. Montalto, F. Moon & A.E. Aktan
A systems approach to assessing the sustainability of the Grand Canal of China 262
N. Tsung, R. Corotis, P. Chinowsky & B. Amadei
Strategies of investment in the management of urban bridges 263
G. Vacheyroux & R.B. Corotis
Eco-audit of seven green infrastructure practices 264
U. Wegst, C. Barr & F. Montalto
Determination of a performance baseline for lifecycle consideration of bridges 265
H. Wenzel & R. Veit-Egerer
Study on evacuation simulation system for urban expressways using multi-agent system 266
M. Yamawaki, W. Shiraki, H. Inomo, Y. Kawakami & K. Yasuda
MS10: Bridging the data gaps for effective management

Organizer: H. Ghasemi
Defining and implementing data collection goals for the Long Term Bridge Performance program 269
M.C. Brown
XIII
Bridging the data gaps for effective management, session 3: performance LTBP data
infrastructure and data integration 270
M. Dejori, R. Ciprian & C. Neubauer
LTBP pilot study methodology and preliminary results 271
A.J. Foden & V.R. Sibley
Enhancing bridge performance defining/measuring/improving bridge performance 272
H. Ghasemi & J.M. Hooks
NDE condition assessment of reinforced concrete bridge decks within the LTBP program 273
N. Gucunski, F. Romero, H. Parvardeh, F.A. Fetrat, S. Kruschwitz & R. Feldmann
Performance indicators for highway bridges and their integration in bridge management 274
N.M. Okasha & D.M. Frangopol
Parameter identification of a reinforced concrete T-beam bridge 275
Y. Zhou, J. Weidner, J. Prader, N. Dubbs, F. Moon & E. Aktan
MS12: Computational prediction & in field validation of bridge performance

Organizers: F. Biondini, F. Bontempi & P.G. Malerba
Probabilistic performance prediction modeling for bridges considering maintenance effects
within a combined computation, visualization and programming environment 279
F. Akgl
Structural maintenance and rehabilitation of a 1900s iron bridge 280
P. Bernardi, E. Michelini & I. Iori
Credibility indicator for bridge service life prediction 281
F. Biondini, E. Garavaglia & D.M. Frangopol
Structural lifetime and elapsed time between first failure and collapse: Application to an arch bridge 282
F. Biondini & D.M. Frangopol
Role of uncertainties on time dependent behaviour of prestressed and cable stayed concrete bridges 283
G. Camossi, P.G. Malerba & L. Sgambi
Variability of measured modal frequencies of a suspension bridge under actual environmental effects 284
Y.L. Ding, A.Q. Li & F.F. Geng
Fuzzy reliability assessment of bridge piers in presence of scouring 285
S. Dordoni, P.G. Malerba, L. Sgambi & S. Manenti
Non linear finite element analysis of a 50 years old reinforced concrete bridge: Comparison with
experimental data and sensitivity analysis 286
S. Epaillard, B. Richard, C. Cremona & L. Elfgren
Field test and analysis of the dynamic factors for bridge in urban railway transportation system 287
X.H. He, P. Li & A. Scanlon
A parametric study on the rocking behavior of bridge columns with spread footing foundation 288
H.H. Hung, K.Y. Liu & K.C. Chang
Tests to collapse of masonry arch bridges simulated by means of FEM 289
T. Kaminski
Dynamic evaluation of bridge mounted sign structures 290
D. Kim & A. Shroff
A special kind of analysis: CFD modeling for design and assessment of bridge
passive control devices 291
S. Manenti, F. Bontempi & P.G. Malerba
Residual service life of existing railway bridges 292
A. Pipinato, C. Pellegrino & C. Modena
Curved-tendon and fiber beam-column element for analysis of cable-stayed PSC bridges 293
D.J. Seong, H.M. Shin, J.R. Cho, K.T. Park & B.S. Kim
Remaining structural capacity of an early post-tensioned bridge verified after removal 294
T. Vogel
XIV
An analysis of long time deflection of long span prestressed concrete bridges 295
L. Vrablik & V. Kristek
Reliability-based assessment of RC bridges in a marine environment considering spatial
and temporal variability of deterioration processes 296
J.S. Zhu & D.M. Frangopol
MS13: Management & maintenance of long span bridges

Organizers: A. Chen & D.M. Frangopol
Anticorrosive coating and maintenance of suspension bridges in China 299
X. Gongyi, H. Yanqing & Z. Junhui
Integrity management of concrete bridges using Spatial Information Systems 300
J. Qin & M.H. Faber
Research on structural performance evolution of a long span concrete bridge 301
H. Tian & A. Chen
Example of bridge maintenance program applied by toll bridges agencies in
Europe: The Rion-Antirion Bridge 302
V. Stathopoulos, G. Hovhanessian & B. Kroely
Construction stages simulation for cable-stayed bridge on geometrical nonlinear analysis 303
J. Wu, Q.-L. Zhang, X.-Q. Luo & L. Kang
MS14: SmartEN Marie Curie ITN Smart management for

sustainable human environment
Organizers: T. Onoufriou & R. Helmerich
Use of sensors for efficient design of innovative seismic protection techniques for monuments 307
C.Z. Chrysostomou, R. Votsis, N. Kyriakides, Th. Demetriou & A. Stassis
In-node data processing in structural monitoring with wireless sensor networks 308
G. Feltrin, J. Meyer, R. Bischoff & O. Saukh
Capabilities of non-destructive testing of RC structures quality assurance for crack repair
using ultrasonic echo 309
M. Friese, U. Effner, F. Mielentz & R. Helmerich
Reference-free health monitoring system using chaos theory 310
H. Furuta, Y. Nomura, K. Koichiro, H. Hattori & Y. Teramae
SHM in integrity management of deterioration prone concrete structures 311
M. Imran Rafiq
Wireless sensor technology for continuous health monitoring of structures 312
A. Kounoudes, A. Kalis, T. Onoufriou & A.G. Constantinides
Role of monitoring in life-cycle assessment, prediction and management of deteriorating structures 313
Predicting the life of reinforced concrete structures in severe marine environments 314
R.E. Melchers & C.Q. Li
Research and training challenges within SmartEN Marie Curie ITN 315
T. Onoufriou, N. Mousoulides, E.A. Tantele, C. Chrysostomou, D. Hadjimitsis,
A. Kounoudes, A.G. Constantinides & A. Kalis
SmartEN A research framework on smart management for sustainable human environment 316
T. Onoufriou, A.G. Constantinides & A. Kounoudes
Role of structural health monitoring in Pareto optimization of bridge management strategies 317
A.D. Orcesi & D.M. Frangopol
Optimization of life-cycle preventative maintenance strategies using genetic algorithm and
Bayesian Updating 318
E.A. Tantele & T. Onoufriou
XV
MS15: European approach on integrated infrastructure risk management (IRIS)
Organizer: H. Wenzel
Health monitoring of a smart base isolated benchmark cable-stayed bridge using
symptom approach 321
G.P. Cimellaro, A. De Stefano, M. Domaneschi & L. Martinelli
Reliability of SHM procedures and decision support in infrastructure management 322
A. Del Grosso
Crystal clear data fusion in subspace system identification and damage detection 323
M. Dhler, B.X. Lam & L. Mevel
Experiments for damage detection by subspace identification on a tied arch bridge 324
C. Ebert & A. Lenzen
Web based monitoring and assessment of bridges and structures 325
M. Fritz, M. Stger & R. Berger
Practical implementation of SHM with a special focus on the end users needs 326
P. Furtner & H. Wenzel
Recording and simulating environmental effects upon Tamar Suspension Bridge 327
R.J. Westgate, K. Koo & J.M.W. Brownjohn
A statistical method for masonry arch bridges damage detection 328
G. Ruocci & A. De Stefano
Proposal of a workers and scaffolds monitoring and risk mitigation system for building sites 329
A. Sassi, L. Gioanola & P. Civera
Observed dynamic characteristics of an overpass bridge during destructive testing 330
D.M. Siringoringo, T. Nagayama, Y. Fujino, D. Su & C. Tandian
Case based reasoning systems for comparative assessment 331
R. Stumptner, B. Freudenthaler, J. Kng & M. Aburahma
European approach on integrated infrastructure risk management 332
H. Wenzel & H. Allmer
MS16: Safety & management of bridges in Mexico

Organizer: D. De Leon
Basis for risk management of bridges exposed to seismic loading 335
D. De Leon
Seismic vulnerability of free toll bridges in Michoacan state, Mexico 336
H. Hernandez & D. De Leon
New technologies to rehabilitate an old Mexican pier in a harsh environment 337
A.A. Torres-Acosta, P. Castro-Borges & M. Len-Irola
Corrosion damage evaluation and diagnosis of bridges in the Mexican highway network 338
A.A. Torres-Acosta, M. Martnez-Madrid, A. del Valle-Moreno, J.T. Prez-Quiroz &
M.A. Backhoff-Pohls
MS17: COWI Group mini-symposium Cable supported bridges

Organizer: J.S. Jensen
Fatigue monitoring systems Great Belt Bridge 341
N. Bitsch, J.L. Jensen & J.E. Andersen
Dehumidification of suspension bridge main cables 342
M.L. Bloomstine & O. Srensen
Imagination, good engineering and saving money 343
K.F. Kirkwood, K. Deng & P.G. Buckland
XVI
When major bridges need to carry more and heavier loads 344
K.F. Kirkwood & P.G. Buckland
Structural health monitoring for Bosphorus Bridge 345
D. Mackenzie, N. McFadyen & S. de Neumann
Bridge hanger deterioration 346
D. Mackenzie, N. Adamson & M. Bloomstine
Bridge barriers and parapets 347
D. MacKenzie, J. Tarrell & O. Srensen
Surfacing for orthotropic bridge decks 348
N. McFadyen & J. Blumensen
Bridge bearings and joints 349
O. Srensen, D. MacKenzie & E. Lauersen
Replacement of suspension bridge suspenders and main cables 350
O. Srensen, M.L. Bloomstine & P.G. Buckland
MS18: Lifetime design of super long span bridges

Organizer: H.-M. Koh
The prediction of long term operation and maintenance costs of long span bridges 353
N. Bitsch, E.S. Larsen & J.V. Thomsen
Deterministic and probabilistic durability design methods and their application to super
long span bridges 354
S.-W. Cha, K.-H. Kim & J.-Y. Cho
Analysis and comparison of recent bridge failures in China and the rest of the world 355
Y.-J. Ge & Z. Sun
Terminology for treating disproportionate collapse 356
M. Haberland & U. Starossek
Identification of aeroelastic parameters for cable-supported bridges using measured accelerations 357
Y.H. Hong, K. Jung, H.K. Kim & H.S. Lee
Reliability-based durability design and quality control of long span bridge of
Busan-Geoje Fixed Link project in Korea 358
S. Kim, S. Jeong, J. Kim & I. Koo
Field surveying and FE model updating of a suspension bridge in service 359
H.-K. Kim, H.-S. Chung & N.-S. Kim
Forth replacement crossing design for safe maintenance and management 360
S. Kite, M. Carter & N. Hussain
R&BD programs on the lifetime design technology for super long span bridges in Korea 361
H.-M. Koh & W. Park
Multi-scale system reliability analysis of bridge structures using dominant failure modes
identified by selective searching technique 362
N. Kurtz, J. Song, D.-S. Kim & S.-Y. Ok
Elasto-plastic behaviors and ultimate strength of continuous super long-span suspension bridge 363
K. Nogami, A. Okubo, Y. Morizono & M. Nagai
Probabilistic kernel principal component analysis for monitoring a suspension bridge
under environmental variations 364
C.K. Oh & D. Bae
Integrated geometry control system for cable stayed bridge: Application to Incheon bridge 365
H.Y. Shin, D.K. Im, J.K. Yoo & C.H. Kim
XVII
Redundancy analysis for a cable-stayed bridge using fibre model 366
K. Sugioka, H. Kanaji, S. Matsumoto, K. Magoshi & M. Nagai
Analytical study on detection of structural damages of a long-span suspension bridge by
wind-induced response 367
N.D. Thang, H. Katsuchi, H. Yamada & E. Sasaki
New high performance steels for long-span bridges 368
T.-Y. Yoon, J.-K. Lee, K.-S. Kim & J.-Y. Cho
Special Sessions
SS1: Use of health monitoring for life-cycle cost analysis & optimization
Organizers: D.M. Frangopol & A. Orcesi
Optimization of the amount of structural health monitoring via probabilistic system assessment 371
C. Albrecht, D. Hosser, C. Klinzmann & R. Schnetgke
The potential link between bridge management systems, structural health monitoring and bridge
weigh-in-motion progress and challenges 372
R.M. Ellis & P.D. Thompson
Probabilistic optimal bridge monitoring planning 373
S. Kim & D.M. Frangopol
Optimization of bridge maintenance actions considering combination of sources of information:
Inspections and expert judgment 374
L.C. Neves & D.M. Frangopol
Structural health monitoring: from data acquisition to optimum life-cycle management
under uncertainty 375
A.D. Orcesi & D.M. Frangopol
The role of structural identification in asset management 376
J. Weidner, J. Prader, N. Dubbs, F. Moon & E. Aktan
SS2: Safety monitoring & maintenance strategy for long span bridges
Organizer: A. Chen
Safety monitoring of the cable stayed bridge in the Commercial Harbor of Venice, Italy 379
B. Briseghella, E. Siviero, C. Lan, E. Mazzarolo & T. Zordan
Iterative buckling analysis for steel cable-stayed bridges 380
D.H. Choi, H.S. Na, S.G. Gwon, D.H. Yoo & H. Yoo
Rehabilitation planning and stay cable replacement design for the Hale Boggs Bridge in
Luling, Louisiana 381
A.T. Ciolko, S.T. Wyatt & C.A. Ligozio
Reliability assessment of Yonghe Bridge based on structural health monitoring 382
S. Li, H. Li & J. Ou
Risk management of newly-built bridge vibration effects on surrounding buildings 383
X. Shi, S. Geng, X. Ruan, Z. Wang & Z. Qu
Bridge safety assessment based on field test data with SORM method 384
X.M. Wang, X.F. Shi, X. Ruan & T.Y. Ying
SS3: Bridge condition assessment

Organizers: A. Miyamoto & F. Tondolo
Use of vibration parameters for evaluating structural degradation in R.C. elements and structures 387
R. Ceravolo, G. Mancini & F. Tondolo
Structural condition assessment of long-span suspension bridges using long-term monitoring data 388
Y. Deng, Y.L. Ding & A.Q. Li
XVIII
Time-dependent reliability of PSC bridge box-girders exposed to aggressive environments 389
T. Guo, R. Sause, D.M. Frangopol & A.Q. Li
Condition improvement of deteriorating bridges by using high performance ceramic materials 390
H. Nakajima, T. Yamagami, S. Tsukamoto, H. Sugiyama, T. Ura & H. Koide
Rehabilitation of Haynes Avenue Bridge 391
A.P. Ranasinghe & V. Ganji
Influence of corrosion on prestressed concrete beams: An experimental survey 392
Z. Rinaldi, S. Imperatore, C. Valente & L. Pardi
Development of a compound inspection method to detect fatigue damages on orthotropic steel deck 393
H. Sugiyama, A. Tabata, H. Nakajima, T. Yamagami, S. Tsukamoto, M. Uchima, M. Okuno & Y. Kohno
Application of Bayesian Network for concrete bridge deck condition rating 394
A. Tarighat & A. Miyamoto
Vulnerability analysis of a reinforced concrete structure by visual inspection 395
M.V. Piras, L. Deias & F. Mistretta
Remaining lifetime and resistance of corrosion attacked bridge deck 396
J. Vican & M. Skora
The condition assessment and strengthening for the deteriorating prestressed concrete
trussed combination arch bridge 397
G.W. Yao & S. Bai
Development of a bridge condition assessment system by using city bus 398
A. Yabe & A. Miyamoto
SS4: Bridges for high speed railways

Organizer: R. Calcada
Resonant effects on a bowstring railway bridge with orthotropic deck 401
C.M.C. Albuquerque, R.A.B. Calada & R.M. Delgado
Long-term monitoring of Sesia high speed railway bridge 402
G. Chellini, F.V. Lippi, L. Nardini & W. Salvatore
Fatigue assessment of Sesia high speed railway viaduct 403
G. Chellini, F.V. Lippi & W. Salvatore
New considerations on track-structure interaction in railway bridges 404
P. Gonzlez Requejo, M. Cuadrado Sanguino & D. Losa Baur
Comfort evaluation of railway bridge vibration using bridge-train transfer function 405
B.G. Jeon, N.S. Kim & S.I. Kim
Bridges E-2 and E-3 in the new railway to the Northwest of Spain 406
C. Jurado
New bridge in pergola for the new railway of high speed trains to the Northwest of Spain 407
C. Jurado
Fatigue damage assessment of steel bridge members using paint cracking 408
Y. Okumura, M. Sakano, Y. Horie & H. Kobayasi
Movable Scaffolding Systems for 72 m span in one casting stage 409
A.A. Pvoas
SS5: Industrial smart material applications for civil infrastructure (ISMA)

Organizer: T.B. Messervey
Streicker Bridge: Initial evaluation of life-cycle cost benefits of various structural
health monitoring approaches 413
B. Gliic & S. Adriaenssens
XIX
Optimal design of bridge SHM systems based on risk and opportunity analysis 414
D. Inaudi
Smart textiles and their application in bridge engineering 415
T.B. Messervey, D. Zangani & J.L. Withiam
Conical couplers for non-pigtailed, free-space optical coupling to fiber optic sensors for
bridge monitoring 416
L. Qiu, K. Goossen, D. Heider, D.J. OBrien & E.D. Wetzel
SS6: Advances in structural robustness: dependability framework

Organizer: F. Bontempi
Dependability of complex bridge structural systems 419
F. Bontempi & S. Arangio
Multilevel structural analysis for robustness assessment of a steel truss bridge 420
F. Brando, R.B. Testa & F. Bontempi
A framework for robustness assessment in the context of corroded RC structures 421
E.S. Cavaco, L.A.C. Neves, J.R. Casas & A.H. Huespe
Buckling of steel gusset plates 422
C. Crosti & D. Duthinh
Cable-loss analyses and collapse behavior of cable-stayed bridges 423
M. Wolff & U. Starossek
SS8: Nondeterministic schemes for structural safety & reliability of bridges

Organizer: S. Arangio
Degradation history simulation a tool for assessment of structural lifetime 427
M.A. Ahrens, P. Mark & Ch. Fust
Non deterministic approaches in the current structural codes for assessing the safety and
reliability of bridges 428
S. Arangio & M. Ghosn
Basic aspects for the uncertainty in the design and analysis of bridges 429
F. Bontempi & L. Giuliani
Block shear failure of steel gusset plates 430
C. Crosti & D. Duthinh
Generation of modified earthquake time-histories using Hilbert-Huang transform 431
S.-H. Ni, W.-C. Xie & M.D. Pandey
Influence of model parameter uncertainties on the seismic vulnerability analysis of
continuous steel-concrete composite bridges exhibiting dual-load paths 432
E. Tubaldi, M. Barbato & A. DallAsta
Reliability evaluation of complex bridges under multiple limit states 433
J.S. Zhu & D.M. Frangopol
SS10: Life cycle bridge engineering in Korea

Organizers: H.-N. Cho & J.-S. Kong
Study on bridge asset management in Korea based on infrastructure asset management methodology 437
K.-H. Park, J.-W. Sun, Y.-K. Hwang, C. Park & M.-J. Lee
Consideration of safety in the revision of the bridge management system in Korea 438
H. Park, J.-S. Kong, J.-H. Kim, K.-H. Park & S.-Y. Oh
Development of life-cycle analysis based new bridge management system in Korea 439
J.-W. Sun, K.-H. Park, S.-Y. Lee, J.-J. Song, J.-S. Kong & H.-Yong
XX
SS11: ARCHES: Assessment & rehabilitation of Central European highway structures
Organizers: T. Wierzbicki & J.R. Casas
Smart cathodic protection systems 443
R.B. Polder, J. Leggedoor, G. Schuten, S. Sajna & A. Kranjc
Assessment and monitoring of existing bridges to avoid unnecessary
strengthening or replacement 444
J.R. Casas
Composite UHPFRC-concrete construction for rehabilitation most recent advances
and applications 445
A. ajna, J.S. uput, E. Denari, E. Brhwiler, G. Habert, P. Rossi,
L. Recic & T. Wierzbicki
Corrosion resistant steels as reinforcement in concrete 447
I. Stipanovic Oslakovic, D. Bjegovic, M. Serdar, A. Legat, A. ajna,
T. Kosec, V. Kuhar & N. Gartner
SS13: Current advancements in bridge technology

Organizer: A.H. Malik
Planning and design of 1-pylon suspension bridge, Dandeung Bridge 451
H.S. Choi, J.G. Yoon & J.H. Kim
Recent development of wireless bridge monitoring system 452
C.C. Fu
Technical advancements in bridge technology 453
A.H. Malik
Can nanotechnology address todays civil engineering challenges? 454
D.M. Mulenga & P.C. Robery
Impact of longitudinal tendons on long-term deflections of long-span concrete cantilever bridges 455
Z. Pan, C.C. Fu & Z. L
Assessment of effective flexure rigidity and composite capacity of corrugated steel-concrete
composite deck with I-beam welded 457
J.J. Shim, S.N. Hong, J.M. Park, J.W. Choi, S.K. Park & K.B. Han
SS14: Using technology to manage, preserve, & renew landmark signature bridges
Organizers: D.S. Lowdermilk & F.L. Moon
Load capacity estimation for the Burlington Bristol Bridge 461
Design and implementation of load cell bearings to measure dead and live load effects
in an aged long span bridge 462
Structural health monitoring of signature bridges: An engineers perspective 463
D. Lowdermilk, L. Egan & J. Prader
Health monitoring of the Tacony-Palmyra bridge bascule span 464
J. Weidner
SS15: Modeling of bridge seismic response

Organizer: M. Fischinger
Application and evaluation of a performance-based methodology for the seismic assessment
of multi-span simply supported deck bridges 467
D. Cardone, G. Perrone, S. Sofia & L. Pardi
XXI
Probabilistic seismic response and retrofit assessment of aging bridges 468
J. Ghosh & J.E. Padgett
Seismic response of repaired bridges by pseudodynamic tests 469
D. Lavorato & C. Nuti
Response of curved steel bridges to seismic loading 470
D.G. Linzell, J. Seo & V.P. Nadakuditi
The design of r.c. bridge deck subjected to horizontal actions by strut-and-tie models 471
M. Mezzina, F. Palmisano & D. Raffaele
Hybrid seismic isolation design of Sakarya-II Viaduct in the proximity of the North Anatolian Fault 472
C. Ozkaya, N. Celebi, F. Tulumtas & M. Dicleli
Seismic abutment-deck interaction of a four-span R/C bridge model 473
M. Saiidi, A. Ebrahimpour & N.B. Johnson
Influence of shear key modelling on the performance of bridges under simulated seismic loads 474
P.F. Silva & T.L.T. Nguyen
A case study of analysis techniques for precast segmental bridges subjected to vertical
seismic excitation 475
M.J. Veletzos & J.I. Restrepo
Compressive stress-strain model for high-strength concrete confined with spirals 476
J.C. West & R.A. Hindi
SS16: Recent challenging bridge structures

Organizer: I.S. Darwish
Assessing and extending the service life of bridges 479
C.A. Ligozio, S.T. Wyatt & A.T. Ciolko
The role of finite element analysis in bridge assessment and design 480
T. Cakebread
Challenges in design, rehabilitation and construction of bridge structures 481
M. Kasi, H. Abdou & D. Morrill
Innovative design approach to a GFRP pedestrian bridge: Structural aspects, engineering
optimization and maintenance 482
S. Russo & A. Adilardi
SS17: Life-cycle design of structural systems

Organizers: F. Biondini & D.M. Frangopol
Structural health monitoring role in bridge life cycle analysis 485
S. Alampalli & M. Ettouney
Lifetime seismic performance of concrete bridges 486
F. Biondini, E. Camnasio & A. Palermo
Structural robustness and redundancy of deteriorating concrete bridges 487
F. Biondini & D.M. Frangopol
Life-cycle analysis of a new composite material for bridge pavement waterproofing 488
M. Crispino, F. Giustozzi & G. Flintsch
The fatigue limit states of the AASHTO LRFD Bridge Design Specifications 489
D.R. Mertz
Reliability assessment of reinforced concrete beams rehabilitated with CFRP sheets 490
C.M. Paliga, M.V. Real, A. Campos Filho & S.M.C. Diniz
XXII
Time varying risk modeling of deteriorating bridge infrastructure for sustainable
infrastructure design 491
A.S. Rao, M.D. Lepech, A.S. Kiremidjian & X.Y. Sun
Study of remaining fatigue life of Brazilian concrete bridges 492

J.F.S. Rodrigues, P.A.O. Almeida & J.R. Casas
SS18: Optical monitoring techniques for bridge maintenance & safety

Organizers: S. Sumitro & H. Matsuda
Real-life applications and experiences with fiber optic bridge monitoring installations 495
K. Chandler, T.W. Graver & A. Mndez
Base study on measure of transformation of concrete structures used by digital image

correlation method 496
H. Goda, M. Hibino, M. Uchino, K. Yamaguchi & H. Matsuda
Development of approximation process of existing action stress of pre-stressed concrete

by stress relief technique 497
K. Hida, Y. Ito, M. Uchino, T. Okamoto, P. Sumitoro & H. Matsuda
Development of on-board image measurement system for actual running and application
to wall surface surveys of structures 498
H. Kurashige, S. Nishimura, K. Hara, J. Kato, T. Omae & H. Yamasaki
Development of a remote laser scanning system for continuous monitoring of

cable-stayed bridges 499
T. Miyashita, N. Miyamoto, S. Umemoto & Y. Fujino
3D profile measurement and buckling simulation on thin-walled cylindrical shells under

compression by utilizing 3D-digital image correlation method 500
C. Morita, C. Zhao, A. Demizu, K. Makino, H. Matsuda & K. Ichimiya
Strain analysis method using multi-rosette analysis by digital image correlation method 501
M. Uchino, T. Okamoto, K. Hida, Y. Ito, P. Sumitoro & H. Matsuda
Deflection measurement for bridges with frequency-shifted feedback laser 502

S. Umemoto, M. Fujii, N. Miyamoto, T. Okamoto, T. Hara, H. Ito & Y. Fujino
SS19: Implementation of bridge management administration in Japan

Organizers: H. Furuta & E. Watanabe
Asset management of bridge structures in Osaka prefectural government 505
H. Ito, R. Tanaka & H. Furuta
Evaluation of maintenance cost estimation and feedback to the BMS 506

H. Kawaragi, M. Kaneuji & E. Watanabe
Risk based inspection strategy considering structural redundancy of long span bridges 507
N. Mashima, K. Sugioka, H. Kobayashi, H. Kanaji, H. Oishi & K. Kaito
Approach for bridge management using BMS in West Nippon Expressway Company Limited 508
T. Matsui, Y. Wada, S. Sakai & T. Yasuzato
Preventive maintenance and technical development on long-span bridges 509

M. Okuda, I. Yamada & Y. Hasegawa
Risk evaluation and financial analysis for road maintenance on urban expressway
based on H-BMS 510
Y. Sakai, M. Jido, H. Furuta & K. Kobayashi
Study for bridge renewal and repair by Osaka municipal government 511
E. Watanabe, H. Furuta, T. Yokota, Y. Komatsu & S. Nagahashi
XXIII
SS20: Challenges for enhancing bridge security
Organizer: S.R. Duwadi
Recognizing and reducing vulnerabilities of transportation infrastructure 515
S.R. Duwadi
Bridge design relevance and efficiency of protective measures for bridge structures
under severe loading 516
M. Noeldgen, W. Caspari & J. Krieger
Mitigation of terrorist threats to structural components 517

J. Ray, V. Chiarito, B. Walker & J. Minor
SS21: Structural monitoring of bridges: Hong Kongs experience

Organizers:Y.-L. Xu & M.C.H. Hui
Structural health monitoring and safety evaluation of Stonecutters Bridge under the
in-service condition 521
K.Y. Wong
Structural health monitoring of Stonecutters Bridge under the construction stage 522
K.Y. Wong, D.M.S. Yau & M.C.H. Hui
Thermal behaviors of Tsing Ma Suspension Bridge 523

Y. Xia, B. Chen, Y.Q. Bao & Y.L. Xu
SHMS-based bridge rating method for long span cable-supported bridges 524
Y.L. Xu, Y. Zheng, Z.W. Chen, Y. Xia, Q. Li, A.X. Guo & K.Y. Wong
A method for stress concentration factor determination of welded steel bridge T-joints
under moving load 525
X.W. Ye, Y.Q. Ni, J.M. Ko & K.Y. Wong
Long-term performance of structural health monitoring system in bridges 526

S. Zhu, W.A. Shen & Y.L. Xu
SS23: Performance-based design for steel structures

Organizers: S.-H. Kim & J.-S. Kong
Determination of load actions for performance-based design 529
H. Furuta, N. Sato & M. Kawatani
A study on vibration performance estimation of footbridge using human body model 530
S.-H. Kim, K.-I. Cho & M.-S. Choi
I-girder with discrete torsional bracings: Lateral-torsional buckling and torsional vibration 531
C.T. Nguyen & H.E. Lee
Reliability evaluation of RC piers of highway bridge using time series multi-state

system approach 532
W. Shiraki, H. Inomo, M. Matsushima, S. Yabe & M. Dogaki
Reliability based design optimization of steel box girder bridge under corrosion attack 533
N.V. Son, M.C. Jeong, S.H. Kim & J.S. Kong
Performance-based durability design of weathering steel bridge 534

E. Yamaguchi
SS24: Steel bridge rehabilitation

Organizer: M. Sakano
Rehabilitative design/build of a railroad bridge 537
W.J. Castle
XXIV
Strengthening design for two-span steel-concrete composite bridges strengthened by
external tendons 538
D.H. Choi, H.S. Na & S.G. Gwon
Assessment and corrosion performance enhancement of weathering steel highway structures 539
N. Damgaard, S. Walbridge, C. Hansson & J. Yeung
Analytical study of fatigue repair of coped beams using carbon fibre reinforced polymers 540
T.A. Holden, J.J.R. Cheng, C.C. Lam & M.C.H. Yam
CFRP repair of fatigue cracks and bonding behavior subjected to cyclic load during curing 541
F. Lin, H. Nakamura, K. Maeda, H. Suzuki & T. Irube
An analytical study on the fatigue cracking of the top flange in steel railway through
truss stringers 542
K. Matsumoto, Y. Niwa, T. Nakayama & M. Sakano
A study of longer service life of a multispan simple steel railway bridge 543
Y. Niwa, T. Nakayama, K. Matsumoto, M. Sakano & A. Koshiba
Development of XFEM analysis code for simulation of fatigue crack propagation in
steel structure 544
K. Shibanuma, H. Aoi, T. Utsunomiya, M. Sakano & Y. Natsuaki
Development of health monitoring indicators for steel bridges and its implementation
to sensing systems 545
K. Sugisaki, Mas. Abe, Mak. Abe & M. Shimamura
Retrofit and advanced investigation on fatigue cracks penetrating orthotropic steel deck plates 546
H. Sugiyama, K. Sugioka, A. Tabata, S. Tsukamoto & K. Utsunomiya
Study on improvement of fatigue strength of out-of-plane gusset welded joints
by attaching GFRP 547
H. Suzuki, N. Inaba, Y. Tomita & K. Shito
Study on improvement of the fatigue durability by filling of mortar in U-shaped rib of
orthotropic steel deck 548
A. Tabata, Y. Aoki, Y. Takada, H. Onishi & S. Matsui
Study of fatigue assessment of orthotropic steel decks in consideration of variations in
locus distribution of wheels 549
Y. Takada, A. Tabata & T. Yamaguchi
Experimental study on the root fatigue life improvement for the reasonable design of steel
pier seismic reinforcement 550
T. Tominaga, K. Matsuoka, C. Miki & Y. Hirabayashi
Effect of reinforcing method against fatigue cracking of orthotropic steel deck with bulb ribs 551
D. Yamaoka, M. Sakano, Y. Natsuaki, S. Nonaka, Y. Nakagawa & K. Nakamura
SS25: New developments in bridge design codes

Organizer: A.S. Nowak
Development and calibration of new reliability-based bridge design code in Korea 555
E.-S. Hwang, I.R. Paik & H.-M. Koh
A study on ductility of high strength steel bridge girder 556
H.S. Joo, S.H. Cha, H.E. Lee & B.H. Choi
Development of the AASHTO guide specifications for bridges vulnerable to coastal storms 557
J.M. Kulicki
Approach for developing calibrated service limit states for the AASHTO LRFD bridge
design specifications: A progress report 558
J.M. Kulicki, W.G. Wassef, A.S. Nowak & D.R. Mertz
XXV
Effect of tension stiffening in composite bridges in the light of Eurocodes 559
G. agoda, W. Radomski & M. agoda
SS26: WIM-based live loads for bridges

Extreme truck load effect prediction for bridge structural reliability 563
G. Fu & J. You
US bridge formula (FBF-B) and implications of its possible application in Europe 564
B.A. Jacob, M. Bereni, M. Ghosn, B. Sivakumar & T. Kearney
Live load models based on weigh-in-motion data 565
M. Kozikowski & A.S. Nowak
Study of multiple presence probabilities for trucks using recent weigh-in-motion data 566
B. Sivakumar
MENSUSMONITOR Algorithm implementation for detecting live load events and
assessment of structural effects on bridges 567
H. Sousa, A. Henriques, J. Figueiras & B. Costa
SS27: High performance concrete lessons of past decades

Organizer: M.L. Ralls
High performance concrete in Washington state 571
J. Kapur
HPC lessons learned and future directions 572
S.N. Lane
HPC implementation in Virginia with lessons learned 573
C. Ozyildirim
Status of Texas DOT HPC implementation 574
K.R. Pruski
High performance concretes evolution in NH 575
D.L. Scott
SS28: Construction, architecture & testing of soil-steel bridges

Organizers: Z. Manko & D. Beben
Dynamic analysis of soil-steel arch road bridges 579
D. Beben & Z. Zee Manko
Dynamic testing of a soil-steel composite railway bridge 580
R. Karoumi, E. Bayoglu Flener & H. Sundquist
Rehabilitation of old arch bridges using corrugated shell structures 581
J. Kurzynski, Z. Zee Manko & D. Beben
SS29: Lessons learned from instrumented bridges

Organizer: M.Q. Feng
Complementing long term health monitoring of bridges using numerical models 585
P.J. Fanning, M.Q. Feng & H. Gomez
Long-term bridge performance monitoring program in California 586
M.Q. Feng & L.-H. Sheng
Performance assessment of Jiangyin Bridge using long-term structural health monitoring data 587
H.F. Zhou, Y.Q. Ni, Y.L. Guo & J.M. Ko
XXVI
SS30: Chinese bridges
Organizer: M.-C. Tang
Major steel bridges for high speed railway in China 591
Z. Gao
Major bridges in Shanghai 592
Y.-P. Lin
Bridges in Chongqing the bridge capital of China 593
D. Yin, W. Xu & X. Duan
Cross-river bridges in Jiangsu 594
Q. You, L. Ji & X. Dong
General Sessions
Organizers: F. Biondini & S. Pakzad
Experimental study on repair welding under static and cyclic loads 597
Y. Agano, S. Fujihira, S.H. Lee & Y.C. Kim
Long-term performance prediction of RC bridge slabs in a marine environment 598
M. Akiyama, D.M. Frangopol & I. Yoshida
Arrigoni Bridge inspection and testing program 599
B. Bantimba & A. Siddiqui
On the applicability of random field theory to transportation network analysis 600
P. Bocchini & D.M. Frangopol
Design check of bridge over Tejo River in Carregado, Portugal 601
P. Cachim & P. Fidalgo
Using microsimulation to estimate highway bridge traffic load 602
C.C. Caprani
Pseudodynamic and cyclic tests on reduced-scale pier-deck sub-systems 603
D. Cardone, G. Perrone, S. Sofia, M. Dolce & L. Pardi
The case study and application of the substructure replacement technique for bridges
having serious scouring of foundation 604
D.-W. Chang, J.-L. Wang, C.-P. Chang & M.-Y. Yu
Bridge maintenance prioritization through visual inspection results 605
M.H. Chen, H.C. Tsai, C.Y. Wang & M.C. Chen
Time-dependent reliability analysis of systems with repairable or non-repairable components 606
T. Cheng & M.D. Pandey
An empirical life-cycle assessment of the relative cost-effectiveness of alternative
materials for reinforcing bridge decks 607
A. Cope, S. Labi & C. Fang
A procedure to derive probabilistic fatigue strength data for riveted joints 608
J.A.F.O. Correia, A.M.P. de Jesus, A.L.L. da Silva & J.F.N. da Silva
Uncertainty evaluation of reinforced concrete structures behavior 609
P.J.S. Cruz, I. Valente, J.C. Matos & J. Batista
Toward a more rational serviceability considerations for high performance steel bridges 610
S. Darjani, M.A. Saadeghvaziri, S. Saigal, N. Aboobaker & A. Khan
Refined load rating analysis of in-service bridges in North Carolina, USA 611
S. Das
Designing the rehabilitation of the Henley Street Bridge in Knoxville, Tennessee, USA 612
S. Das
XXVII
Stress intensity factors evaluation for riveted beams applying FEA with VCCT 613
A.M.P. de Jesus & J.A.F.O. Correia
Deterioration and maintenance of RC bridge decks under uncertainty: Condition and
reliability indicators 614
A. Dec & D.M. Frangopol
Multi directional hysteretic damper with adaptive post-elastic stiffness for seismic
protection of bridges in near fault zones 615
M. Dicleli & A. Salem-Milani
Distribution of live load effects in integral bridge abutments and piles 616
M. Dicleli & S. Erhan
Static and dynamic model validation and damage detection using wireless sensor network 617
S. Dorvash, S. Pakzad, R. Yao & K. Okaly
Modern permanent and portable restraint systems for bridges 618
T. Edl & A. Barnas
Effect of soil bridge interaction on the distribution of live load effects among integral
bridge components 619
S. Erhan & M. Dicleli
Live load distribution in integral bridge girders 620
Comparison of fatigue behavior under concentrated loads of orthotropic decks and crane runways 621
M. Euler & U. Kuhlmann
The strain development in concrete under cyclic loading 622
M. Foglar
Cracking of reinforced concrete deck on skewed bridges 623
G. Fu, J. Feng, J. Dimaria & Y. Zhuang
The Filetto Bridge on the Santerno River near Bologna (Italy): Seismic retrofit and
reinforcement design 624
G. Gasparini, S. Silvestri, T. Trombetti & C. Ceccoli
Masonry arch bridges towards a hierarchical assessment framework 625
N. Gibbons & P.J. Fanning
Target proof load factors for highway bridge assessment in Central and Eastern
European Countries (CEEC) 626
J.D. Gmez & J.R. Casas
Synthetic fiber ropes to replace steel wire in pedestrian suspension bridges 627
S. Gupte, R. Betti & T. Zoli
Seismic evaluation of Sogutlucesme Viaduct 628
A.F. Gurdil, A. Yagcioglu & A.G. Ozcebe
Sustainable design for steel-concrete bridges 629
G. Hauf, P. Maier & U. Kuhlmann
Estimating natural frequencies, damping ratios and FEM models of suspension bridges
from wind response measurement 630
A.L. Hong, R. Betti & F. Ubertini
Using orthogonal pairs of rollers on concave beds (OPRCB) in seismic design and retrofit
of highway bridges 631
M. Hosseini, A. Soroor & S. Khavari
Two cable-stayed bridges designed for easy access and low maintenance 632
R. Hsu
Development of a cooperative management framework for bridge maintenance using IFC data model 633
M.G. Huang, H.-J. Kim, K.-H. Kim & S.-H. Lee
XXVIII
Development of live load model using weigh-in-motion data 634
E.-S. Hwang
A review of metallic bridge failure statistics 635
B.M. Imam & M.K. Chryssanthopoulos
Shear transfer mechanism in slab-on-girder bridges 636
A. Jamali & M.A. Saadeghvaziri
Estimation of optimum isolator parameters for effective mitigation of seismic risk for bridges 637
M. Karalar & M. Dicleli
Development of a new cycle counting method for cyclic thermal strains in integral bridge piles 638
Properties of self-consolidating light-weight concrete in massive structures 639
M. Kaszynska
Seismic vulnerability assessment and comparing various retrofit designs for an
existing highway bridge 640
S.R. Khavari & M. Hosseini
Experimental evaluation of spliced PSC-steel-PSC hybrid girder connected with advanced
joints consisting of perfobond ribs 641
S.-H. Kim, J.-H. Yoon, S.-J. Park, J.-H. Won, C.G. Lee & S.-J. Kim
Construction of the hybrid cable-stayed bridge 642
D.G. Kim, H.C. Kwon & K.J. Lee
Low cycle fatigue strength of cruciform welded joints considering plate thickness 643
K. Kinoshita & K. Ueda
An analysis on the torsion of hybrid bridges with corrugated steel webs considering tensile
strength of concrete 644
H.J. Ko, H.E. Lee & J. Moon
Application of a new metal spraying system for steel bridges. Part 5. Analysis of the
corrosion prevention mechanism of the system 645
T. Kondo, S. Okuno, H. Matsuno & M. Ohshiba
Assessment of existing steel structures Recommendations for estimation of the remaining
fatigue life 646
B. Khn
Computational methodologies for the blast vulnerability assessment of steel bridge girders 647
L.D. Leininger & A.M. Coughlin
Design live load for long span bridges 648
M. Lutomirska & A.S. Nowak
Management of interstate 95 in Pennsylvania 649
T.P. Macioce, H.M. Berman & D. Constable
Monitoring of the Manhattan Bridge and interferometric radar systems 650
L. Mayer, B. Yanev, L.D. Olson & A. Smyth
A rational partial composite bridge beam transition 651
P.S. McCarten
Structural health monitoring for bridge management 652
P.S. McCarten
Destructive test of a steel slab-on-girder bridge 653
J. McConnell, M. Chajes, H. Shenton III, K. Michaud, C. Russo & J. Ross
Bridge inspection and repair technologies on the expressways in Japan 654
H. Miyauchi
XXIX
Improving transparency of bridge condition for decision-making and analysis in
bridge management 655
J.T. Newlin, W.L. Meyers, G.P. Patil & S.W. Joshi
Countermeasures for rain- and wind-induced vibrations on the Meiko Nishi Bridge 656
A. Noguchi & H. Miyauchi
Proposal for modification of load and resistance factors in the AASHTO LRFD bridge design code 657
A.S. Nowak & P. Paczkowski
Recommendations for dynamic allowance in bridge assessment 658
E.J. OBrien, A. Gonzlez & A. nidaric
Influence of steel reinforcement corrosion on the stiffness of simply supported concrete beams 659
F.J. OFlaherty, P.S. Mangat, P. Lambert & E.H. Browne
Influence of shear reinforcement corrosion on the performance of under-reinforced concrete beams 660
F.J. OFlaherty, P.S. Mangat, P. Lambert & E.H. Browne
Fatigue evaluation of steel finger type expansion joints for highway bridges 661
S. Ono, S. Sakai & T. Imamura
Design of elastic foundation layers for buildings new design and calculation
methods design examples 662
M. sterreicher & W. Unterberger
An attempt of rationalization for maintenance of railway structures using management
supporting system 663
M. Oyado, M. Miyashita, S. Ueda & A. Sakairi
Sensitivity of bridge reliability to parameter variation in systems susceptible to spatially
distributed soil liquefaction 664
J.E. Padgett, J. Ghosh & L. Dueas-Osorio
An enhanced XML Schema matching technique for checking of the missing items in
construction document delivery 665
S.I. Park, B.-G. Kim, H.J. An & S.-H. Lee
Structural analysis and fatigue life assessment of the Paderno Arch Bridge 666
A. Pipinato
New high endurance sliding material for bridge bearings 667
V. Quaglini, C. Tavecchio, P. Dubini, D. Cuminetti & D. Ferroni
What is bridge durability official regulations and reality 668
W. Radomski
Effective management of concrete assets concrete durability: Achievement and enhancement 669
P.C. Robery
Condition inspection, analysis and maintenance of three prominent railroad bridges 670
C. Rolwood & R. Drew
Full scale laboratory testing of replacement orthotropic deck for Verrazano Narrows Bridge 671
S. Roy, J.W. Fisher & R.S.D. Alapati
Structural hardening for cable elements of cable supported bridges 672
N.D. Sauer
Peformance assessment of bridges under progressive damage and abnormal actions 673
D. Saydam & D.M. Frangopol
Detection of fatigue crack of steel deck plate by ultrasonic nondestructive testing 674
H. Shirahata, K. Akasaka & T. Iizuka
Fiber-reinforced polymer composites: An effective solution to our nations crumbling infrastructure 675
M. Stafford
XXX
Seismic assessment on the existing highway bridges in Taiwan 676
Y.-C. Sung, K.-C. Chang, D.-W. Chang, C.-L. Chen, H.-M. Wu & K.-Y. Liu
Prediction of restrained shrinkage stresses and premature deck cracking in composite bridges 677
A. Szwed & M.M. Szerszen
Evaluation of corrosion deterioration of weathering steel bridge under the
environmental corrosiveness 678
Y. Tamaki, T. Shimozato, Y. Arizumi, T. Yabuki & S. Ono
Study on the cause of brittle fracture during earthquakes in steel bridge bent
focusing on stress triaxiality 679
H. Tamura, E. Sasaki, H. Yamada & H. Katsuchi
A simplified approach to calculate the secondary moments of continuous prestressed
concrete bridges 680
V. Tiruchirapalli & R.A. Hindi
Fatigue testing of stay cables at resonant frequency 681
W. Traeger & J. Kollegger
Durable connections between precast bridge components: Fundamental approach 682
A.E. Ulku, U.B. Attanayake & H.M. Aktan
Bridge strengthening by structural change: from continuous beam to network arch 683
M. Valenzuela & J.R. Casas
Innovations in fabrication of the self-anchored suspension span of the
San Francisco-Oakland Bay Bridge 684
M. Wahbeh, P. Siegenthaler, T. Nilsson, M. Nader & A. Cavendish-Tribe
Fatigue life evaluation of existing highway reinforced concrete bridges 685
C. Wang, G. Li, X. Dong, L. Hao & J. Wang
Durability rehabilitation of a reinforced concrete bridge damaged by corrosion and overload 686
H.L. Wang, X.Y. Sun & Z.J. Zheng
Where are the sustainable bridges in the United States? 687
D. Whittemore
A heuristic approach for optimizing bridge inspection route 688
C.-I. Yen, H.-K. Liao, J.-J. Chai & N.-J. Yau
Assessment of the operation level of a bridge network in postearthquake scenarios 689
Y.C. Yue, D. Zonta, F. Bortot & R. Zandonini
A distributed application for infrastructure risk prevention 690
M.H. Zaharia, F. Leon, G.M. Atanasiu, G. Pagu & E. Czeran
The relevance of whole life costs for infrastructure buildings 691
T. Zinke & T. Ummenhofer
Author Index 693
XXXI
Preface
The number of deteriorating bridges is increasing worldwide. Costs of maintenance, repair and rehabilitation of
these bridges far exceed available budgets. Maintaining the safety and serviceability of existing bridges by making
better use of available resources is a major concern for bridge management. Internationally, the bridge engineering
profession continues to take positive steps towards developing more comprehensive bridge management systems.
Consequently, it continues to be appropriate to bring together all of the very best work that has been done in the
field of bridge maintenance, safety, management and life-cycle optimization at the Fifth International Conference
on Bridge Maintenance, Safety and Management (IABMAS2010), held in Philadelphia, Pennsylvania, USA,
from July 11 through 15, 2010. The First (IABMAS02), Second (IABMAS04), Third (IABMAS06), and
Fourth (IABMAS08) International Conference on Bridge Maintenance, Safety and Management were held in
Barcelona, Spain, July 1417, 2002, Kyoto, Japan, October 1822, 2004, Porto, Portugal, July 1619, 2006, and
Seoul, Korea, July 1317, 2008, respectively.
The International Association for Bridge Maintenance and Safety (IABMAS, www.iabmas.org), which serves
as the organizing association of IABMAS2010 in conjunction with Lehigh Universitys Advanced Technology for
Large Structural Systems (ATLSS) Engineering Research Center, encompasses all aspects of bridge maintenance,
safety and management. Specifically, it deals with: health monitoring and inspection of bridges; bridge repair
and rehabilitation issues; bridge management systems; needs of bridge owners, financial planning, whole life
costing and investment for the future; bridge related safety and risk issues and economic and other implications.
The objective of IABMAS is to promote international cooperation in the fields of bridge maintenance, safety,
management, life-cycle performance and cost for the purpose of enhancing the welfare of society. The interest
of the international bridge community in all these fields has been confirmed by the high response to the call
for papers. In fact, 835 abstracts were received by the Conference Secretariat. Approximately 70% of them
were selected for final publication as full-papers and presentation at the Conference within mini-symposia,
special sessions, and general sessions. Compared to IABMAS08 the total of number of papers scheduled for
presentation has increased from 465 to 511.
IABMAS2010 covers all major aspects of bridge maintenance, safety, management and life-cycle optimization
including advanced materials, ageing of bridges, assessment and evaluation, bridge codes, bridge diagnostics,
bridge management systems, bridge security, composites, design for durability, deterioration modeling, emerging
technologies, fatigue, field testing, financial planning, health monitoring, high performance materials, innova-
tions, inspection, life-cycle performance, load capacity assessment, loads, maintenance strategies, new technical
and material concepts, non-destructive testing, optimization strategies, prediction of future traffic demands,
rehabilitation, reliability and risk management, repair, replacement, residual service life, safety and service-
ability, service life prediction, strengthening, sustainable materials for bridges, sustainable bridges, informatics,
whole-life costing, and multi-criteria optimization, among others.
Bridge Maintenance, Safety, Management and Life-Cycle Optimization contains the lectures and papers pre-
sented at IABMAS2010. It consists of a book of abstracts and a CD-ROM containing the full texts of the lectures
and papers presented at IABMAS2010, including the T.Y. Lin Lecture, nine Keynote Lectures and 501 technical
papers from 35 countries. This set provides both and up-to-date overview of the field of bridge engineering
and significant contributions to the process of making more rational decisions in bridge maintenance, safety,
management, life-cycle performance, and cost for the purpose of enhancing the welfare of society.
On behalf of IABMAS and Lehigh Universitys ATLSS Engineering Research Center, the chairs of the
Conference would like to take this opportunity to express their sincere thanks to the authors, organizers of
special sessions and mini-symposia, and participants for their contributions; to the Conference Honorary Chair,
Professor John W. Fisher; to the members of the Conference Scientific and Organizing Committees for their
dedicated work; and to the members of the Local Organizing Committee for the time and effort they have
devoted to making IABMAS2010 a successful event. Finally, we would like to register our sincere thanks to all
the sponsors of IABMAS2010.
Dan M. Frangopol and Richard Sause
Chairs, IABMAS2010
Bethlehem, Pennsylvania, USA, April 2010
XXXIII
Conference organization
ORGANIZING ASSOCIATION
The International Association for Bridge Maintenance and Safety (IABMAS) (http://www.iabmas.org/)
CONFERENCE CHAIRS
Dan M. Frangopol, Lehigh University, Bethlehem, PA, USA

Richard Sause, Lehigh University, Bethlehem, PA, USA
CONFERENCE HONORARY CHAIR
John W. Fisher, Lehigh University, Bethlehem, PA, USA
HOST ASSOCIATION
Advanced Technology for Large Structural Systems (ATLSS) Engineering Research Center, Lehigh University,
Bethlehem, PA, USA (http://www.atlss.lehigh.edu)
CONFERENCE INTERNATIONAL SCIENTIFIC COMMITTEE
Fabio Biondini (Chair) Politecnico di Milano, Milan, Italy

Andrzej Nowak (Chair) University of Nebraska, Lincoln, NE, USA
Chung-Bang Yun (Chair) Korea Advanced Institute of Science and Technology, Daejon, Korea
Shamim Pakzad (Secretary) Lehigh University, Bethlehem, PA, USA
A. Emin Atkan Drexel University, Philadelphia, PA, USA
Haluk Aktan Western Michigan University, Kalamazoo, MI, USA
Sreenivas Alampalli New York State Department of Transportation, Albany, NY, USA
Alfredo H. S. Ang University of California-Irvine, Irvine, CA, USA
Giuliano Augusti University of Rome La Sapienza, Rome, Italy
Konrad Bergmeister University of Natural Resources & Applied Life Sciences, Vienna, Austria
Jan Bien Wroclaw University of Technology, Wroclaw, Poland
Franco Bontempi University of Rome La Sapienza, Rome, Italy
Geoff Boully VIC Roads, Camberwell, Victoria, Australia
Fernando Branco Technical University of Lisbon, Lisbon, Portugal
James Brownjohn The University of Sheffield, Sheffield, UK
Eugen Bruehwiler EPFL, Lausanne, Switzerland
Michael Bruneau State University of New York-Buffalo, Buffalo, NY, USA
Christian Bucher Vienna University of Technology, Vienna, Austria
Harald Budelmann Technical University of Braunschweig, Braunschweig, Germany
Joan R. Casas Technical University of Catalonia, Barcelona, Spain
F. Necati Catbas University of Central Florida, Orlando, FL, USA
Chin-Chen Chang Hong Kong University of Science & Technology, Hong Kong, P.R.C.
Michael Chajes University of Delaware, Newark, DE, USA
Kuo-Chun Chang National Taiwan University, Taipei, Taiwan, R.O.C.
Sung-Pil Chang Seoul National University, Seoul, Korea
Steven Chase University of Virginia, Charlottesville, VA, USA
Airong Chen Tongji University, Shanghai, China
Moe Cheung The Hong Kong University of Science & Technology, Hong Kong, P.R.C.
XXXV
Hyo-Nam Cho Hanyang University, Ansan, Korea
Marios Chryssanthopoulos University of Surrey, Guilford, Surrey, UK
Young Soo Chung Chung-Ang University, Seoul, Korea
Marcello Ciampoli University of Rome La Sapienza, Rome, Italy
Joel Conte University of California, San Diego, CA, USA
Ross Corotis University of Colorado, Boulder, CO, USA
Christian Cremona LCPC, Paris, France
Paulo Cruz University of Minho, Guimaraes, Portugal
Andrea Del Grosso University of Genoa, Genoa, Italy
George Deodatis Columbia University, New York, NY, USA
Sophia Diniz Federal University of Minas Gerais, Minas Gerais, Brazil
Sheila Duwadi Federal Highway Administration, McLean, VA, USA
Lennart Elfgren Lule University of Technology, Lule, Sweden
Bruce Ellingwood Georgia Institute of Technology, Atlanta, GA, USA
Ib Enevoldsen RAMBLL, Copenhagen, Denmark
Allen Estes California Polytechnic State University, San Luis Obispo, CA, USA
Glauco Feltrin Swiss Federal Laboratories for Materials Testing & Research, Dbendorf,
Switzerland
Joo Almeida Fernandes National Civil Engineering Laboratory, Lisbon, Portugal
Joaquim Figuerias University of Porto, Porto, Portugal
John W. Fisher Lehigh University, Bethlehem, PA, USA
Michael Forde University of Edinburgh, Edinburgh, UK
Dan M. Frangopol Lehigh University, Bethlehem, PA, USA
Karl Frank University of Texas, Austin, TX, USA
Ian M. Friedland Federal Highway Administration, McLean, VA, USA
Gongkang Fu Wayne State University, Detroit, MI, USA
Yozo Fujino University of Tokyo, Tokyo, Japan
Hitoshi Furuta Kansai University, Osaka, Japan
James H. Garrett Carnegie Mellon University, Pittsburgh, PA USA
Yaojun Ge Tongji University, Shanghai, China
Hamid Ghasemi Federal Highway Administration, McLean, VA, USA
Michel Ghosn City University of New York, New York, NY, USA
Paul Grundy Monash University, Victoria, Australia
Rade Hajdin Infrastructure Management Consultants GmgH, Zrich, Switzerland
Geir Horrigmoe Norut Technology, Narvik, Norway
Michael C.H. Hui Government of Hong Kong Special Administrative, Hong, Kong, P.R.C.
Naeem Hussain ARUP, Hong Kong, P.R.C.
Daniele Inaudi SMARTEC SA, Manno, Switzerland
Jens Sandager Jensen COWI A/S, Lyngby, Denmark
Maarten-Jan Kallen HKV Consultants, Lelystad, The Netherlands
Ahsan Kareem University of Notre Dame, South Bend, IN, USA
Maria Kaszynska Szczecin Technical University, Szczecin, Poland
Malcolm Kerley Virginia Department of Transportation, Richmond, VA, USA
Sang-Hyo Kim Yonsei University, Seoul, Korea
Sungkon Kim Seoul National University of Technology, Seoul, Korea
Risto Kiviluoma WSP Finland Ltd., Helsinki, Finland
F. Wayne Klaiber Iowa State University, Ames, IA, USA
Leo Klatter Public Works & Water Management, Utrecht, The Netherlands
Jan-Ming Ko The Hong Kong Polytechnic University, Hong Kong, P.R.C.
C. G. Koh National University of Singapore, Singapore
Hyun-Moo Koh Seoul National University, Seoul, Korea
Jung Sik Kong Korea University, Seoul, Korea
Ulrike Kuhlmann University of Stuttgart, Stuttgart, Germany
John Lane Railway Safety and Standards Board, London, UK
Jrn Lauridsen Danish Road Directorate, Copenhagen, Denmark
Kincho Law Stanford University, Stanford, CA, USA
Hakeun Lee Korea University, Seoul, Korea
Jeeho Lee Dongguk University, Seoul, Korea
Chin-Hsiung Loh National Taiwan University, Taipei, Taiwan, R.O.C.
Jerome Lynch University of Michigan, Ann Arbor, MI, USA
Pier Giorgio Malerba Politecnico di Milano, Milan, Italy
XXXVI
Ayaz Malik Rensselaer Polytechnic Institute, Troy, NY, USA
Lance Manuel University of Texas at Austin, Austin, TX, USA
Ren Maquoi University of Lige, Lige, Belgium
Antonio Mari Universitat Politcnica de Catalunya
Barney Martin Modjeski & Masters, Poughkeepsie, NY, USA
Sami Masri University of Southern California, Los Angeles, CA, USA
Thomas Messervey DAppolonia S.p.A, Genoa, Italy
Ayaho Miyamoto Yamaguchi University, Ube, Japan
Aftab Mufti ISIS Canada Research Network, Winnipeg, Manitoba, Canada
Hani Nassif Rutgers, The State University of New Jersey, Piscataway, NJ, USA
Jinping Ou Dalian University of Technology, Dalian, China
Livia Pardi Autostrade per lItalia, Rome, Italy
Udo Peil Technical University of Braunschweig, Braunschweig, Germany
Stephen Pessiki Lehigh University, Bethlehem, PA, USA
Victor Popa Search Corporation, Bucharest, Romania
Mark Reno Quincy Engineering Inc., Sacramento, CA, USA
James Ricles Lehigh University, Bethlehem, PA, USA
Xin Ruan Tongji University, Shanghai, China
Salvatore Russo University of Iuav of Venice, Venice, Italy
Mehdi Saiidi University of Nevada, Reno, NV, USA
Richard Sause Lehigh University, Bethlehem, PA, USA
Richard Shepard City of Elk Grove, Public Works, Elk Grove, CA, USA
Masanobu Shinozuka University of California, Irvine, CA, USA
Dong-Ku Shin Myongji University, Gyeonggido, Korea
Soobong Shin Inha University, Incheon, Korea
Arunprakash M. Shirole Arora and Associates, P.C., MN, USA
Jongsung Sim Hanyang University, Ansan, Korea
Marja-Kaarina Sderqvist Finnish Road Administration, Helsinki, Finland
Mark Stewart University of Newcastle, Newcastle, Australia
Luc Taerwe University of Ghent, Ghent, Belgium
Man-Chung Tang T.Y. Lin International, San Francisco, CA, USA
Palle Thoft-Christensen Aalborg University, Aalborg, Denmark
Paul Thompson Castle Rock, CO, USA
Tomoaki Utsunomiya Kyoto University, Kyoto, Japan
Pedro Vellasco State University of Rio de Janeiro, Rio de Janeiro, Brazil
Thomas Vogel ETH Zrich, Zrich, Switzerland
Eiichi Watanabe Kyoto University, Kyoto, Japan
You-Lin Xu Hong Kong Polytechnic University, Hong Kong, P.R.C.
Zhishen Wu Ibaraki University, Hitachi, Japan
Bojidar Yanev New York City DOT, New York, NY, USA
Riccardo Zandonini University of Trento, Trento, Italy
Yunfeng Zhang University of Maryland, College Park, MD, USA
CONFERENCE INTERNATIONAL ORGANIZING COMMITTEE
A. Emin Aktan (Chair) Drexel University, Philadelphia, PA, USA

Jens Sandager Jensen (Chair) COWI A/S Lyngby, Denmark
Hyun-Moo Koh (Chair) Seoul National University, Seoul, Korea
Ex-Officio Members:
Chad Kusko Lehigh University, Bethlehem, PA, USA
Members:
Eugene Bruehwiler EPFL, Lausanne, Switzerland
Joan R. Casas Technical University of Catalonia, Barcelona, Spain
Airong Chen Tongji University, Shanghai, China
Sheng-Jin Chen National Taiwan University of Science & Technology, Taipei, Taiwan, R.O.C.
Christian Cremona LCPC, Paris, France
XXXVII
Paulo Cruz University of Minho, Guimaraes, Portugal
Yozo Fujino University of Tokyo, Tokyo, Japan
Hitoshi Furuta Kansai University, Osaka, Japan
Rudolph Kotze New Zealand Transport Agency, Hamilton, New Zealand
Pier Giorgio Malerba Politecnico di Milano, Milan, Italy
Chitoshi Miki Tokyo Institute of Technology, Tokyo, Japan
Aftab Mufti ISIS Canada, Research Network, Winnipeg, Manitoba, Canada
Young Suk Park Myongji University, Gyeonggido, Korea
Robert Sweeney Modjeski & Masters, Montreal, Canada
Man Chung Tang T.Y.Lin International, San Francisco, CA, USA
Jin-Guang Teng Hong Kong Polytechnic University, Hong Kong, P.R.C.
Eiichi Watanabe Kyoto University, Kyoto, Japan
Chung-Bang Yun Korea Advanced Institute of Science and Technology, Daejon, Korea
CONFERENCE NATIONAL ORGANIZING COMMITTEE
Mehdi Saiidi (Chair) University of Nevada, Reno, NV, USA
Ex-Officio Members:
Chad Kusko Lehigh University, Bethlehem, PA, USA
Members:
Sreenivas Alampalli NY State DOT, Albany, NY, USA
Raimondo Betti Columbia University, New York, NY, USA
Harry Capers Arora & Associates, P.C., Lawrenceville, NJ, USA
F. Necati Catbas University of Central Florida, Orlando, FL, USA
Michael Chajes University of Delaware, Newark, DE, USA
Steven Chase University of Virginia, Charlottesville, VA, USA
Karl Frank University of Texas, Austin, TX, USA
Ian M Friedland Federal Highway Administration, Washington, DC, USA
Gongkang Fu Wayne State University, Detroit, MI, USA
James H. Garrett Carnegie Mellon University, Pittsburgh, PA, USA
Hamid Ghasemi Federal Highway Administration, Washington, DC, USA
Michel Ghosn City University of New York, New York, NY, USA
Martin Kendall Edwards & Kelcey, Inc., New York, NY, USA
Myint Lwin Federal Highway Administration, McLean, VA, USA
Tom Macioce Pennsylvania Department of Transportation, Harrisburg, PA, USA
Ali Maher Rutgers, The State University of New Jersey, Piscataway, NJ, USA
Barney Martin Modjeski & Masters, Poughkeepsie, NY, USA
Dennis Mertz University of Delaware, Newark, DE, USA
Andrzej Nowak University of Nebraska, Lincoln, NE, USA
Mary Lou Ralls Ralls Newman, LLC., Austin, TX, USA
Mark Reno Quincy Engineering, Inc., Sacramento, CA, USA
Bala Sivakumar HNTB Corporation, New York, NY, USA
Bill Spencer Jr. University of Illinois, Urbana-Champaign, IL, USA
Richard Walther Wiss, Janney, Elstner Assoc., Inc., Northbrook, IL, USA
Bojidar Yanev New York City DOT, New York, NY, USA
CONFERENCE LOCAL ORGANIZING COMMITTEE
Chad Kusko (Chair) Lehigh University, Bethlehem, PA, USA

Mary Ann Cahalan Lehigh University, Bethlehem, PA, USA
Xiaohua (Hannah) Cheng New Jersey Department of Transportation, Trenton, NJ, USA
John W. Fisher Lehigh University, Bethlehem, PA, USA
Dan Hall Bridge Software Development International, Ltd, Coopersburg, PA, USA
Ian C. Hodgson Lehigh University, Bethlehem, PA, USA
Jian Jin Jacobs Engineering Group, West Chester, PA, USA
XXXVIII
Peter A. Mazzeo HNTB, Philadelphia, PA, USA
Ronnie Medlock High Steel Structures, Inc., Lancaster, PA, USA
Bill Michaleyra Lehigh University, Bethlehem, PA, USA
Franklin Moon Drexel University, Philadelphia, PA, USA
Clay Naito Lehigh University, Bethlehem, PA, USA
Shamim Pakzad Lehigh University, Bethlehem, PA, USA
Alan W. Pense Lehigh University, Bethlehem, PA, USA
Stephen P. Pessiki Lehigh University, Bethlehem, PA, USA
James M. Ricles Lehigh University, Bethlehem, PA, USA
Matt Sanfilippo Carnegie Mellon University, Pittsburgh, PA, USA
Harry W. Shenton III University of Delaware, Newark, DE, USA
Wagdy Wassef Modjeski & Masters, Inc., Mechanicsburg, PA, USA
Ben T. Yen Lehigh University, Bethlehem, PA, USA
Yunfeng Zhang University of Maryland, College Park, MD, USA
IABMAS2010 MINI-SYMPOSIA
MS1: Futuristic bridge maintenance technologies, organized by C.-B. Yun & B.F. Spencer Jr.
MS2: Monitoring & assessment of bridges using novel techniques, organized by A. Strauss & D.M. Frangopol
MS3: Present & future of bridge inspection & evaluation, organized by S. Alampalli, A.K. Agrawal &
M. Ettouney
MS4: Research & applications in bridge health monitoring, organized by F.N. Catbas, J.R. Casas,
H. Furuta & D.M. Frangopol
MS5: New procedures for bridge rehabilitation, organized by V. Popa
MS6: Measurement systems for bridge weigh-in-motion (B-WIM), organized by B. Bakht, A. Znidaric &
D.K. McNeill
MS7: Bridge management tools & research, organized by L. Klatter
MS8: Uncertainty in bridge damageability modelling, organized by M. Shinozuka
MS9: Performance-based asset & risk management of the highway infrastructure system, organized by
A.E. Aktan
MS10: Bridging the data gaps for effective management, organized by H. Ghasemi
MS12: Computational prediction & in field validation of bridge performance, organized by F. Biondini,
F. Bontempi & P.G. Malerba
MS13: Management & maintenance of long span bridges, organized by A. Chen & D.M. Frangopol
MS14: SmartEN Marie Curie ITN Smart management for sustainable human environment,
organized by T. Onoufriou & R. Helmerich
MS15: European approach on integrated infrastructure risk management (IRIS), organized by H. Wenzel
MS16: Safety & management of bridges in Mexico, organized by D. De Leon
MS17: COWI Group mini-symposium Cable supported bridges, organized by J.S. Jensen
MS18: Lifetime design of super long span bridges, organized by H.-M. Koh
IABMAS2010 SPECIAL SESSIONS
SS1: Use of health monitoring for life-cycle cost analysis & optimization, organized by
D.M. Frangopol & A. Orcesi
SS2: Safety monitoring & maintenance strategy for long span bridges, organized by A. Chen
SS3: Bridge condition assessment, organized by A. Miyamoto & F. Tondolo
SS4: Bridges for high speed railways, organized by R. Calcada
SS5: Industrial smart material applications for civil infrastructure (ISMA), organized by T.B. Messervey
SS6: Advances in structural robustness: dependability framework, organized by F. Bontempi
SS7: Bridge adaptation to the environmental & climate changes, organized by V. Popa
SS8: Nondeterministic schemes for structural safety & reliability of bridges, organized by S. Arangio
SS10: Life cycle bridge engineering in Korea, organized by H.-N. Cho & J.-S. Kong
SS11: ARCHES: Assessment & rehabilitation of Central European highway structures, organized by
T. Wierzbicki & J. R. Casas
SS13: Current advancements in bridge technology, organized by A.H. Malik
SS14: Using technology to manage, preserve, & renew landmark signature bridges, organized by
D.S. Lowdermilk & F.L. Moon
SS15: Modeling of bridge seismic response, organized by M. Fischinger
SS16: Recent challenging bridge structures, organized by I.S. Darwish
XXXIX
SS17: Life-cycle design of structural systems, organized by F. Biondini & D.M. Frangopol
SS18: Optical monitoring techniques for bridge maintenance & safety, organized by S. Sumitro & H. Matsuda
SS19: Implementation of bridge management administration in Japan, organized by H. Furuta & E. Watanabe
SS20: Challenges for enhancing bridge security, organized by S.R. Duwadi
SS21: Structural monitoring of bridges: Hong Kongs experience, organized by Y.-L. Xu & M.C.H. Hui
SS23: Performance-based design for steel structures, organized by S.-H. Kim & J.-S. Kong
SS24: Steel bridge rehabilitation, organized by M. Sakano
SS25: New developments in bridge design codes, organized by A.S. Nowak
SS26: WIM-based live loads for bridges, organized by A.S. Nowak
SS27: High performance concrete lessons of past decades, organized by M.L. Ralls
SS28: Construction, architecture & testing of soil-steel bridges, organized by Z. Manko & D. Beben
SS29: Lessons learned from instrumented bridges, organized by M. Feng
SS30: Chinese bridges, organized by M.-C. Tang
CONFERENCE SPONSORS (as of May 1, 2010)
FINANCIAL SPONSORS
ATLSS Engineering Research Center, USA

Digitexx Data Systems, Inc, USA
Federal Highway Administration, U.S. Department of Transportation, USA
Pennsylvania Infrastructure Technology Alliance (PITA), USA
T.Y. Lin International, USA
XL
NON-FINANCIAL SPONSORS
AASHTO, American Association of State Highway and Transportation Officials, USA

ACI, American Concrete Institute, USA
AICAP, Associazione Italiana Calcestruzzo Armato e Precompresso, Italian Association for Reinforced and
Prestressed Concrete, Italy
AIOM, Associazione di Ingegneria Offshore e Marina, Association of Offshore and Marine Engineering, Italy
ASCE-SEI, American Society of Civil Engineers-Structural Engineering Institute, USA
ASCP, The Portuguese Group of IABMAS, Portugal
CAU, Chung-Ang University, Korea
CERIC, Civil Engineering Research Information Center, Korea
COWI A/S, Denmark
CTE, Collegio dei Tecnici dellEdilizia, Italian Society of Building Engineers, Italy
COSEIK, Computational Structural Engineering Institute of Korea, Korea
CMM, Portuguese Steelwork Association, Portugal
CU, University of Colorado at Boulder, USA
DAPPOLONIA, S.p.A., Genoa, Italy
DGU, Dongguk University, Korea
ECCS, European Convention for Constructional Steelwork, Belgium
IALCCE, International Association for Life-Cycle Civil Engineering
IASSAR, International Association for Structural Safety and Reliability
ISISE, Institute for Sustainability and Innovation in Structural Engineering, Portugal
JSCE, Japan Society of Civil Engineers, Japan
JSSC, Japanese Society of Steel Construction, Japan
JSMS, Japanese Society of Materials Science, Japan
ISTEC, Smart Infra-Structure Technology Center, KAIST, Korea
KBRC, Korea Bridge Design & Engineering Research Center, Korea
KCI, Korean Concrete Institute, Korea
KEC, Korea Expressway Corporation, Korea
KICT, Korea Institute of Construction Technology, Korea
KISTEC, Korea Infrastructure Safety & Technology Corporation, Korea
KSSC, Korean Society of Steel Construction, Korea
KU, Kansai University, Osaka, Japan
KU-CERR, Kyoto University, Dept. of Civil and Earth Resources Engrg., Japan
MJU, Myongji University, Korea
PCI, Precast/Prestressed Concrete Institute, USA
Politecnico di Milano, Milan, Italy
RCEAS, P.C. Rossin College of Engineering and Applied Science, Lehigh University, USA
Tongji University, Shanghai, China
UM, University of Minho, Guimaraes, Portugal
UniS, University of Surrey, Guilford, UK
UPC, Technical University of Catalonia, Barcelona, Spain
Samsung C&T Corporation, Korea
SNU, Seoul National University, Korea
TRB, Transportation Research Board, USA
XLI
XLII
TECHNICAL EXHIBITORS
A-1 Engineers, Inc., Middletown, CT, USA

ADVITAM, Sterling, VA, USA
AP-Bridge Construction Systems, Lisboa, Portugal
ATLSS Engineering Research Center, Bethlehem, PA, USA
Bentley Systems, Incorporated, Exton, PA, USA
BGFMA, North Baltimore, OH, USA
CAD Rechenzentrum AG, Allschwil, Switzerland
Campbell Scientific, Logan, UT, USA
ChemCo Systems, Inc., Redwood City, CA, USA
Crafco, Inc., Chandler, AZ, USA
DBI Services (DeAngelo Brothers, Inc.), Hazelton, PA, USA
Digitexx Data Systems, Inc., Scottsdale, AZ, USA
Direct Measurement, Inc., West Chester, PA, USA
Dynamic Surface Applications, LTD, Pennsdale, PA, USA
Grace Composites/Design Plastics, Norristown, PA, USA
GWY, Inc., Greenfield, NH, USA
Halcrow, Inc., New York, NY, USA
IABMAS, International Association for Bridge Maintenance and Safety
InspectTech, Pittsburgh, PA, USA
Kinemetrics, Inc., Pasadena, CA, USA
LTBP (Long-Term Bridge Performance) Program, Rutgers CAIT, Piscataway, NJ, USA
Lusas Bridge Analysis Software, NYC, NY, USA
MAGEBA, Buelach/Kanton Zurich, Switzerland
McClain & Co., Inc., Culpeper, VA, USA
Olson Engineering, Wheat Ridge, CO, USA
R.J. Watson, Inc., Amherst, NY, USA
Roctest, Ltd., St-Lambert, Canada
Taylor & Francis Group, London, UK
The Castle Group, Hainesport, NJ, USA
The D.S. Brown Company, North Baltimore, OH, USA
TRANSPO Industries, New Rochelle, NY, USA
VCE Holding GmbH, Vienna, Austria
Vector Corrosion Technologies, Tampa, FL, USA
Watson Bowman Acme Corp., Amherst, NY, USA
XLIII
T.Y. Lin Lecture
Bridge maintenance and safety: A practitioners view
C. Seim
Consulting Bridge Engineer, El Cerrito, California, USA
ABSTRACT an over-height barge crane struck and buckled a major

compression-member of a cantilever-truss; bearings
Maintenance engineering must be practical. Safety is and pinned joints froze up; expansion joints wore out
always paramount in the maintenance operations of and needed replacement under traffic! Of course, all
a bridge. OHSA provides rules for the health and these issues were in addition to the required day-to-
safety of maintenance personnel, and is one of the day cleaning, painting, and inspection of all of the
most important bills ever passed by Congress. Todays nine structures to keep them in first-class condition,
bridge code is recognizing the need for designs that use as was required by the toll bond covenants backing the
durable materials that are long-lasting and can reduce construction of these Toll Bridges.
maintenance demands. Bridge owners and mainte- I had all the maintenance money needed to per-
nance engineers must ensure that maintenance money form these many functions, and I inherited a won-
is effectively utilized for good maintenance. Let the derfully trained team of maintenance workers and
bridge show you where it needs maintenance. a small staff of registered engineers that were kept
busy inspecting the bridges; designing repairs for
damaged members and maintenance access-facilities;
keeping the electrical systems fully functional; and
1 INTRODUCTION
responding to emergencies that always occur from time
to time.
If you ask me what I do, I would reply, I work on
During my tenure in maintenance of these major
bridges. I practice bridge engineering. I am a prac-
bridges, Professor T. Y. Lin, who taught several of my
titioner. Bridge Engineering covers many activities
classes while I was at Berkeley, offered me a position
such as Planning, Designing, Constructing, Maintain-
with his firm, T. Y. Lin International, which was and is
ing, and, finally, the Demolishing of Bridges.
a major consulting firm practicing bridge engineering
Ive done some designing, some construction, and
in the Western Hemisphere and in Asia.
some maintenance-engineering; I also have done some
In 2004, I retired from T.Y. Lin International and
teaching, researching, serving on committees, and
started my own Bridge Consulting Engineering Office.
writing papers. I have been very fortunate to have
I now consult on major bridges in China with Dr. Man-
worked in various capacities on many major long-
Chun Tang and on the new Self-Anchored Suspension
span bridge projects such as the Golden Gate and the
span of the San Francisco-Oakland Bay Bridge.
San Francisco-Oakland Bay Bridges, the Bridge of the
Based upon this long career in bridge engineer-
Americas, the Lake Maracaibo Bridge, the new Cooper
ing, I have some thoughts that I would like to share
River Bridge, and, during the past few years, on several
with you about the maintenance and the safety of
large bridge projects in China.
bridges.
I began my 50 year plus career with the State of
California Bridge Department (now called Caltrans)
on the design and construction of six major Toll
Bridges, that included the 1300-foot cable-stayed 2 MAINTENANCE MUST BE PRACTICAL
Southern Crossing of San Francisco Bay that, unfor-
tunately, was never constructed. Scientists and mathematicians say it must be beau-
I remember clearly that fortuitous day that I was tiful but maintenance engineers say it must be
assigned to the position of Chief Maintenance Engi- practical. Maintenance engineers must be practical
neer for all nine of the State owned Toll Bridges. I because they are responsible for drawing out the last
thought I was being side-tracked into a do-nothing bit of service life that a bridge has to offer. What they
position with no challenging work. do to preserve and extend the life of the bridge must
Was I ever wrong about this do nothing position! I work and work well!
never experienced a dull moment in the five years that Without regular, professional maintenance engi-
I worked in maintenance! Trucks collided with struc- neering, the factor of safety built into the structure
tural members; vehicles caught on fire, damaging the could be greatly reduced, the bridge posted for a load
bridge structure; ships collided with piers and fenders; limit, and it might look unsafe to drive or walk across.
3
Most importantly, if the bridge were to be taken out of padding a header above a sub-standard height opening
service, the highway system would lose an important or posting signs warning of low clearances.
investment of public money.
In emergency situations, maintenance engineers
must act quickly, decisively, and wisely to protect the 4 THE CHANGING MAINTENANCE SCENE
safety of the traveling public. Usually there is no time
to look at textbooks, and, if there were time, there From time to time new materials are introduced to
are few textbooks on the subject that will solve the improve the maintenance of bridges. About 1975,
immediate problem. some states and communities started to impose volatil-
A good example of this occurred during the Septem- ity limits on paint and solvents used in bridge
ber 2009 repair of a fractured eyebar on the San maintenance; sandblasting was curtailed, and full
Francisco-Oakland Bay Bridge. The installation was containment of removed material was required to
not actually a repair, but was a partial-strengthening avoid sweeping it into bays or rivers. The limit on
of the cracked eyebar: the strengthening failed seven paint volatility was an opportunity which resulted in
weeks after its installation. The bridge was closed for improved paints for structural steel. Todays paint sys-
six days to all traffic while a second temporary, but tems have three times the life that they had when I
improved, strengthening system was again installed. started in maintenance with the traditional red lead
The permanent repair was then installed in December paint system.
2009 over eight days by working at night with partial High-Performance Steel and Concrete require less
deck closures. material for construction, should last longer, reducing
the cost of maintenance. Protective coatings for steel
reinforcing bars in concrete decks and in splash-zone
of concrete piers also extend the service lives of these
3 MAINTENANCE AND SAFETY areas.
Todays bridge codes are recognizing the need for
Safety is always paramount in the maintenance oper- designs that use durable materials that are long-lasting
ations of a bridge. Good maintenance includes the and that can reduce maintenance demands. A new buzz
safe management of facilities such as traffic, roadway, word, Sustainable-Design is being used to denote the
aviation, and navigation lighting; keeping deck drains use of these improved materials.
open; repair of expansion joints that spring loose; as Bridge Diagnostic-Systems are being developed
well hundreds of other items. Safety also covers the that make detection of bridge elements with structural
safe management of accesses for maintenance person- problems easier to find, record, and maintain. Bridge
nel such as ladders, cat walks, and their personal gear Management Systems are also being developed that
such as coveralls, gloves, respirators, body harnesses make routine and special maintenance easier to track.
and lanyards. Bridge Security is a new, developing technology
Today there are a number of personnel hoists for for protecting important bridges that will require trial
above-deck and below-deck access that aid in the installations on bridges to determine how they may
safe placement of maintenance workers at the point affect bridge-maintenance functions.
of work. However, these facilities come at the price These few examples show that Bridge Maintenance
of one or two lane closures and the redirecting of can indeed be a changing, exciting profession!
traffic.
These personnel hoists are also used in the impor-
tant function of inspecting all elements of the bridge 5 THE ROLES OF THE BRIDGE OWNER
for close-up inspection. These hoists do have their
limitations, and sometimes inspectors need to climb All bridges that have been built and all the bridges to be
steel, install temporary rigging, or use mountain- built in the future will have owners. What should be
climbing equipment to get to all the inspection points. the role of the owners of bridges in regard to main-
Whichever methods are used, Safety First is the tenance and safety operations? Perhaps, for a new
primary mandate for inspecting a bridge. bridge, their role would be to conceive of a beautiful
Additionally, OHSA compliance for the health and bridge, thoughtfully designed, well constructed, and
safety of maintenance personnel was one of the most safely maintained. For existing bridges, the owners are
important directives for health and safety ever passed limited to safely maintaining their bridges to obtain the
by Congress. OHSA has saved many lives and pre- built-in service life.
vented many injuries over time. Although OHSA rules The design and construction of a bridge requires
may be costly to implement and to self-enforce, and money up front, whereas good maintenance func-
thus have drawn some criticism; this is money well tions require money after the bridge has been built;
spent. All maintenance operations performed on a sometimes that money, unfortunately, can be slow in
bridge must conform to the requirements of OSHA, coming, or worse yet, be cut-off. The owner must
wherever possible. Sometimes, on existing bridges, appreciate that good maintenance will prolong the life
OHSA rules cannot be fully implemented; in such of the bridge, will provide safe passage for the pub-
cases the maintenance engineer must do whatever lic, and will require reserves of money that must be
measures he/she can to increase worker safety, such as appropriated as needed.
4
6 THE ROLES OF THE BRIDGE DESIGNER Fortunately, this Association sponsors conferences,
such as this one, that promotes and advances the art,
Good maintenance starts with good design. If the practice, and development of Bridge Maintenance and
bridge designer does a worthy job of designing the Safety.
structure and a thorough job of selecting proper materi-
als; providing maintenance access when possible; and
providing enough clear space for inspecting, cleaning, 8 WHAT IS GOOD MAINTENANCE?
painting, and replacing good maintenance should
follow. I think that politicians do not always understand why
A bridge designer has many difficult roles to fill: money must be spent on Maintenance. They may think:
he/she needs to satisfy a large number of requirements, You built the bridge, it is carrying traffic, and now you
such as those contained in a four-inch thick design want to fix it. Didnt you do your work right the first
code: choosing the proper structural type, span lengths, time? When money is short, maintenance is usually
and foundation types; selecting the materials, bearings, the first item cut; politicians refer to this process by
and expansion joints; and writing specifications and the euphemism deferred maintenance.
estimating costs.Thus a designer may become too busy Although it is fair for the politicians to ask: Is
to think about the future maintenance of the bridge. our money being well spent or, can this money be
However, I believe the appropriate time to think spent more effectively? It is up to bridge owners to
about installing facilities for maintenance operations provide, and to maintenance engineers to ensure, that
is during the design phase. These facilities will lower maintenance money is effectively utilized for good and
the life-cycle cost of the bridge; but only if the bridge necessary maintenance.
designer is motivated to provide maintenance access Since the advent of the LRFD bridge design spec-
and space on the bridge plans, and if the bridge owner ifications, we have seen the probabilistic basis of this
will provide a little bit more money now to save more new approach to bridge design from publications dis-
money in the future. playing two bell-shaped curves, superimposed on a
diagram, with the curve on the left representing the
loads L (dead, truck, wind, others), and with the curve
7 ROLES OF THE MAINTENANCE ENGINEER on the right representing the resistances R (of the
bridge elements). A simplified interpretation of this
The roles of the maintenance engineer are too numer- diagram is that the distance between the peaks of
ous to mention here; suffice to say that the major the two curves is a measure of safety called Relia-
roles of the maintenance engineer are: maintaining bility Index. The acceptable value of the Reliability
the safety of personnel, the safety of the travelling Index used for the design of bridges, is set by a
public, and the safety of the structures. These roles specification-writing committee and is based upon
demand, among many other requirements, knowledge their judgment and experience, and by comparison to
of structures and materials; some knowledge of con- the performances of existing bridges.
struction practice, repair methods, and of arresting In the practical world, Politicians control L by their
corrosion; familiarity with OHSA, safety devices and votes on legal-load weights for highways, hopefully
access equipment; ability to communicate with and guided by the advice of bridge engineers. However,
to motivating personnel; and the ability to talk pleas- politicians are constantly being lobbied to increase
antly to people who are heatedly complaining about legal-load weights, which they often vote to do.
pot holes in the bridge deck. The resistance, R, of each of the myriad of bridge
Where can bridge owners find good maintenance elements that make up a bridge, is controlled by the
engineers? They usually are made the hard way from bridge designer, using the current, thick bridge design
working on the maintenance of bridges, and if lucky, specifications and his/her knowledge and experience.
under the mentorship of a seasoned maintenance engi- After the bridge is constructed and opened to service,
neer. We never see advertising: Enroll now, get your the resistance, R, is partially controlled in one way
degree in bridge maintenance engineering, and make of looking at it by the bridge maintenance engineer,
a fortune! ensuring that the service life designed into the structure
In my case, I was appointed, kicking and screaming, will be obtained through good maintenance.
into the maintenance engineering function. Luckily I The measure of good maintenance (and good
inherited a good staff that were well trained and knew design) is that the bridge will safely serve society,
what they were doing before I came; they educated me without reduction in load capacity, to the end of its
very quickly as to what I was to do and how I could design life.
best help them to do their job!
However, other maintenance engineers may not be
so lucky, and they may have to start almost from scratch 9 SERVICE LIFE OF A BRIDGE
in educating themselves and in training their own per-
sonnel; or worst, they have small staffs, little or no What is the lifetime of a bridge, or better yet, what is
maintenance money, and must keep their bridges open the service life of a bridge? Service life is the bet-
with bailing wire and sheer determination. ter description because it implies that the bridge will
5
safely carry the loads, without reduction, for which it Maintenance engineers had to stand by helplessly
was designed, over its specified lifetime. An old bridge because these failures were beyond their control to
can live on after its service life has passed but it may manage or prevent; all they could do was watch, patch,
require rehabilitation or extensive reconstruction. and repair. However, there are now several installa-
At the beginning of my career, most bridges were tions of orthotropic decks and wearing surfaces that
designed for a life time of 50 years. We used slide-rules are nearing, or have achieved, a 30 to 40 year service
and the now-obsolete allowable stress design(ASD); life. These successful installations are typified by the
the bridge design specifications were only about use of engineered and laboratory-tested materials, as
inches thick, and life was much simpler! That 50- all of our materials should be, before being used on
year service life was increased, a couple of decades bridges.
ago, to 75 years when the LFD method of design was
developed. I have recently worked on oversight of the
design and the construction of the Cooper River cable- 10 PRIMARY, SECONDARY, AND TERTIARY
stayed bridge which has a specified service life of 100 EFFECTS
years. I am presently working in the same capacity on
the bridge from Hong Kong to Macau, which has a One of the jobs of the bridge designer is to determine
specified life of 125 years. the effects of the primary stresses generated by loads
There is certainly a trend toward increasing the and structural action, and to use appropriate materi-
design service-life of our bridges in the United States: als in the proper amount, to provide the necessary
a hundred year life has been suggested. At what point resistance to meet the load demands. There are now
does a longer specified service life of the bridge trigger programs that have the ability to analyze secondary
an increase in the Reliability Index or load factors, or stresses, non-linear and inelastic structural actions,
in a reduction in fatigue stress? Do we know enough and even dynamic load-time-histories applications.
about the tertiary effects of aging on our materials? However, we still lack the ability to analyze for ter-
More importantly, what effects will longer service tiary effects of time, loading, and the environment on
lives have on the maintenance and safety functions material used in our bridges. For example, the break-
of our bridges? down of paint films under the aging effects of weather,
Rivets have been around for about two centuries; oxygen, and moisture; the migration of chloride ions
though not used very often now; reinforced concrete through concrete, that, when reaching the level of
began to be used in bridges a little over a century the steel reinforcing bars, starts corrosion; and the
ago; prestressed concrete and welded steel girders have fatigue-effects of out-of-plane bending of steel plates.
been used for a little over a half-century. Elastomeric To be sure, specialists can do these things, but usu-
and pot bearings and modular expansion joints have ally not typical bridge designers. We try to cover these
less than a half-century of use in bridges; high perfor- adverse effects whenever they are discovered by code
mance concrete and steel have an even shorter history requirements that are based upon experience. How-
of bridge-use about a decade; yet even shorter is ever, tertiary effects are the very effects that the bridge
the use of advanced composites, which utilize plastic maintenance engineer must inspect, monitor, and con-
resins and fiber reinforcement. trol, so as to provide a long service life for the bridge.
If the century, or the century and a quarter of bridge Although, at the time of this writing, no testing
service life is successfully to be achieved, perhaps results have been published on the eyebars of the San
we need a comprehensive test program for traditional Francisco-Oakland Bay Bridge, I believe the cause of
materials being used for longer-life applications as the crack in the troubled eyebar will be found to be a
well as for the new materials being used for longer tertiary effect that could not be calculated nor found
life in traditional applications. by inspecting at the time that the bridge was designed,
A disastrous example of not performing sufficient nor could the cause be found with todays technology.
development and testing of new materials and new These tertiary effects showed up very vividly in
structural forms was the introduction of orthotropic my experience on the Golden Gate Bridge, back in
steel decks to the United States, about forty years ago. the 1970s. The Golden Gate Bridge, Highway, and
The first-draft design specifications for orthotropic Transportation District (The District) is a completely
steel decks were based on strength-design, rather than separate organization from the California Department
on serviceability-design. In some of the early installa- of Transportation, (Caltrans), for whom I worked at
tions of this deck-type the steel began to fatigue-crack that time.
in high stress areas and had to be repaired in the field. The District had employed a consultant to evalu-
An even worse example was the choice of wearing ate the concrete deck; the consultant reported that the
surface material placed on the steel deck to provide deck reinforcing bars were fatiguing under wheel loads
skid resistance, a smooth ride, and to protect the steel and would begin fatigue-fracturing within a few years.
deck from corroding. Asphalt, or modified asphalt was The concrete deck could not be replaced under traf-
used with very little laboratory testing to prove its dura- fic. Therefore a new lower deck in the plane of the
bility. All of the original asphalt-based materials failed lower truss chords would need to be constructed, traf-
in just a few years and now need to be replaced every fic diverted to the new lower deck, the upper concrete
decade or so. deck removed, and a new deck cast in place, all under
6
full traffic. Obviously this would be a tremendously But this will never happen again on the Golden
expensive and traffic-disrupting plan. The District Gate Bridge. The corroded rivets were replaced with
asked Caltrans for a second opinion and I drew the high strength bolts and new lacing bars were installed.
assignment. About 1970, and over a twenty-year period, the
I read the report stating that corrosion had occurred maintenance crews blasted off the old red lead paint,
between and on the top flanges of the longitudinal steel applied an inorganic zinc primer, and protected the
stringers, lifting the concrete slab free of its support primer with an overcoat of durable paint. The Bridge
from the flanges of the stringers. The deck was now has not lost steel section since. It is now a model bridge
spanning over one or two longitudinal stringers, and for showing what good maintenance should be.
the extra-long spans were producing higher fatigue
stresses under truck wheel loading. 11 THE BRIDGE WILL SHOW YOU
I noted that the report used a beam analogy instead
of a plate or arch analogy. I requested that the Cal- The most important function for the maintenance and
trans Transportation Laboratory in Sacramento place safety operations of a bridge is inspection, either by
strain gauges on a few of the reinforcing bars and mon- eye or by instruments. If you look, the bridge will show
itor stresses during the morning commute. The strain you where it needs maintenance help.
gauge showed that the maximum stress recording was During the most famous bridge collapse of all-time,
about 2000 psi. Even with an impact factor of 100 per- the Qubec Bridge, under construction in 1906, was
cent, this low stress would not be significant in terms deflecting abnormally and some of the iron work-
of fatiguing the rebars. ers walked off the job. The now infamous I 35W
However, our inspection did find that reinforcing truss bridge had bucked gusset plates before it col-
bars near the soffit of the deck appeared to be cor- lapsed in 2007. Several of the lifting cables of the
roding. We then performed half-cell readings on the old Dumbarton Bridge lift-span in San Francisco Bay
deck and took two-inch diameter cores through the were vibrating excessively and had to be replaced in
6 1/2 inch thick deck, sliced the cores, and analyzed 1975. The cable stays of the Luling Bridge, near New
the slices for Chloride ions. The analysis showed quite Orleans, were galloping abnormally and developing
clearly that the chloride content in the lower third of cracks in the plastic tubing and in the cement grout;
concrete in the deck was above the threshold content the stays are now being replaced. And so on, as there
that sustains steel reinforcing bar corrosion. The deck are literally thousands of examples; however the most
was corroding, not fatiguing. important observation is what your bridge is telling
The chloride was being deposited on the soffit of you, if you look.
the deck from the salt laden fogs that continually roll Of course there are distress items that are hidden
through the Golden Gate. The Golden Gate Bridge from view. The flaw in the eyebar of the Silver Bridge
needed a new steel deck that could be constructed across the Ohio River that failed and precipitated the
under traffic. The new steel orthotropic deck was total collapse of the suspension bridge in 1969 was a
opened in 1985 and is still performing well but that tertiary effect hidden from view.
is another story. But there are now ongoing efforts to develop detec-
Both the corrosion of the top flange of the stringers tion instrumentation, data acquisition recorders, and
and the intrusion of the Chloride ions were all tertiary transmission methods to find hidden distressed areas
effects that could not be calculated or predicted by the the tertiary effects but these effects are still what the
designers at that time. We can do this today, but with bridge is telling you, but looking in a different and
the exception of the cable-stayed Cooper River Bridge, more effective way.
we just dont do it. The conditions on the Golden Gate We all can look forward to the development of new
Bridge were made worse by the chief engineer stip- technologies, and to the improvements of the ongo-
ulating that sandblasting was not to be used to clean ing technologies, which will assist us with inspection,
the steel for paint application, as each blasting cycle maintenance, and safety work on our bridges.
removes some steel, and during the multi-century life
of the structure, would remove too much of the steel
sections. 12 SUMMING UP
The irony of this requirement is that the bridge has
lost more steel section from corrosion than it ever I have had a wonderful career in bridge engineering;
would have from sandblasting. I remember many years but my stint in bridge maintenance stands out as the
ago talking to the paint superintendent of the San high-light. I use that experience all the time in my
Francisco-Oakland Bay Bridge, who was very criti- current bridge consulting work, when I ask myself,
cal of the maintenance of the Golden Gate Bridge. He How can THAT be maintained? I mentor younger
told me sandblasting was not allowed on that bridge, engineers to acquaint them with, and to think about the
and as a result many rivets have lost their heads from maintenance functions of a bridge: Remember, if you
corrosion, and that the lacing bars are sharp enough to cant access THAT for inspection and maintenance,
shave by. I thought he was exaggerating, but when I THAT will not last as long as it should!
was inspecting the deck, I did see rivets without heads, I have not listed any references, as these are my
and lacing bars sharp as a razor. thoughts alone, based on my own experiences.
7
Keynote Lectures
Fundamentals of suspension bridge retrofit
P.G. Buckland
Buckland & Taylor Ltd., North Vancouver, BC, Canada
ABSTRACT: The argument is made that when retrofitting a suspension bridge, the greatest value is obtained
when the engineer considers the fundamentals of engineering, rather than simply following the code. By fun-
damentals two aspects are considered. One is a thorough understanding of not only structural behavior, but also
how loads, resistances, and safety factors are derived. The other aspect covers the fundamentals of good design
(as opposed to good analysis), in preparing designs that can be constructed during short occupancies and are
tolerant of problems that might arise during the short construction period between deadlines. The design aspect
includes looking for opportunities to reduce maintenance and improve inspection access, and produce such other
benefits as may be desired.
1 INTRODUCTION increased, sometimes the bridge was designed for a

loading much less than specified today, often it has
This paper reflects the authors opinions on what simply deteriorated.
should be the fundamental considerations for the refur- 3. Improved access for inspection;
bishment or capacity increase of a suspension bridge. 4. Easier and less costly maintenance; and
The reason for going to the fundamentals is that very 5. Other benefits that may be desired, such as
often code requirements are inadequate to cover a par- improved seismic performance.
ticular situation, so it is important to understand how
code provisions are derived, and therefore how they
might be modified in a controlled way if modification 2.2 Costs
is appropriate. The four main cost items are:
There are also fundamental considerations that the
engineer should have in mind while he or she is design- 1. The physical cost of upgrading;
ing the retrofit of a suspension bridge. These include: 2. The (often hidden) cost to the public caused by traf-
fic disruption. To this can be added lost revenue to
What am I trying to accomplish? the operator where tolls are charged.
How can disruption to traffic be minimized? 3. Over-strengthening by being conservative. Design-
What will be easiest for the constructor to build? ing a new bridge for an increase in live load usually
What other benefits can be achieved, such as adds only a small percentage to the cost of the
reduced and easier maintenance, or better access, structure, but increasing the capacity of an existing
or improved seismic resistance? bridge can be extremely costly.This is why unneces-
sary conservatism can be horrendously expensive.
The larger the bridge, the more this is so.
2 KEY DRIVERS 4. Under-strengthening can be the most costly of all,
as it can lead to failure of the bridge and even loss
The key considerations driving the retrofit of a bridge of life.
are, as usual, benefits and costs. The purpose of this
paper is to see how these are best managed for a It can be seen that Costs 1 and 2 are related. The cheap-
suspension bridge. est retrofit will probably coincide with the maximum
traffic disruption. Since ultimately the user pays the
price of retrofit, and also suffers the loss caused by
2.1 Benefits disruption, an economist could in theory find a way to
balance the two. In the authors experience, this is not
The main benefits are: usually attempted, and it may be that good judgment
1. Improved traffic flow (more lanes, for example) and intuition will provide close to the optimum solu-
which can also lead to increased tolls; tion. By the time a bridge is being retrofitted the traffic
2. Increased safety, both in terms of how the traf- volume has usually built up to the capacity level of the
fic moves (perhaps wider lanes) and the ability bridge. Therefore only traffic disruption that occurs at
to take increased loads. Sometimes the loads have off-peak times will normally be tolerated.
11
The best balance between Cost Items 1 and 2 can be
found by careful and creative design, and this theme
will be examined later in the paper.
It can also be seen that Cost Items 3 and 4 are
related, in the sense that either cost will be over the opti-
mum unless the narrow band is found between too safe
and not safe enough. In other words: how safe is safe
enough? That is the starting point for this presentation.
3 SAFETY
One can think of safety in several ways. At its most

basic, safety is the margin between the capacity of the
bridge to carry its loads, and the loads imposed upon Figure 1. Traffic loading on the Blue Water International
it. We therefore need to know three things: Bridge.
1. the capacity of the bridge;
2. the loading to be applied; and can only be because something has happened to the
3. the required degree of separation between the two bridge that was not modeled because it was not known.
(i.e. the safety factor). For example, when this technique was first applied
to the Lions Gate Bridge in Vancouver, BC, it was
It is true that the bridge design codes provide guid- found that the centre of the bridge was about one meter
ance on these matters, but if there is uncertainty, design lower than expected, and the towers were leaning in
codes err on the side of conservatism, or should do. towards each other. One of the towers had settled a
And as mentioned earlier, conservatism may be a very few centimeters, and some dead load had been added
expensive luxury, which can make it poor engineering. to the bridge, but these effects only accounted for about
The above three aspects of safety will be considered half of the observed discrepancies. The only explana-
in turn. tion of the observed discrepancies was that the main
cables, which comprised spiral strands, had stretched
3.1 The capacity of the bridge over time. Once this was accepted as the only ratio-
nal explanation, these effects could be added to the
Modern analysis software and hardware should be able others in the computer, and the true stress state was
to provide a good idea of the capacity of a bridge, but determined.
experience has shown that this is not always the case. In this instance, a cable-bent (side tower) was found
Inspection of the bridge can detect any weak points, to be seriously overstressed, and remedial action was
such as corroded steel, broken rivets and the like, taken immediately.This deficiency could not have been
but even the best inspection does not provide any discovered by inspection, but it was seriously reducing
information about the stress state of the bridge. the capacity of the bridge.
The shape of a suspension bridge is quite sensitive to
any change in its condition.This characteristic is useful
3.2 Applied loading
for what we call the Analysis-and-Survey method of
determining the stress condition of the bridge. In this Dead load can be determined from a weight take-
method, a computer model is created of the bridge as it off, and this may be modified by the findings of the
was originally built, preferably from as-built drawings analysis-and-survey.
or from design drawings if as-builts are not available. The main unknown is the live load imposed by
The model is then altered to reflect any changes that traffic. Various studies have provided some guidance
have occurred since the bridge was constructed. These on the subject, such as Buckland et al 1980 which
changes would include any additional dead load, struc- provided the basis for the long span loading in both
tural changes, foundation settlement, and so on. Once AASHTO-LRFD and the Canadian Highway Bridge
these changes have been made, the geometry of the Design Code CAN/CSA-S6-06.
bridge will also have changed. An extreme case of heavy loads is shown in Figure 1.
The bridge is then surveyed, preferably in the early Recent research by the Authors company has re-
morning before sunrise, with overcast sky, no wind, examined the multi-lane reduction factors, which are
and no traffic on the bridge. The survey will include the probably the least known part of the live loading
profile of the deck from end to end, and the verticality for long spans. In one recent study it was confirmed
in both directions of the towers. Once the computer that the multi-lane reduction factors vary with loaded
model has been adjusted to account for the temperature length, being generally greater for long lengths than for
at which the bridge has been surveyed, the shape of the short lengths, as shown in Table 1. This could add con-
bridge in the model should be the same as that in the siderable complexity when trying to apply the design
computer. loading.
In our companys examination of more than 20 sus- Table 1 shows that at long loaded lengths of about
pension bridges, the two have never fully agreed! This 500 m (1,600 ft.) the reduction factor for 8 lanes loaded
12
Table 1. Multi-lane reduction factors, recent study.
Number of loaded lanes

Loaded length, m 1 2 3 4 8
6 1.00 0.70 0.60 0.50 0.35

16 1.00 0.65 0.60 0.50 0.35
480 1.00 0.90 0.85 0.65 0.60
Table 2. Multi-lane reduction factors, AASHTO.
Number of loaded lanes

AASHTO 1 2 3 4 8 Figure 2. A. Murray MacKay Bridge, Halifax, NS, Canada.
All lengths 1.00 1.00 0.90 0.75 0.75
is nearly double the measured value for 8 lanes of short

loaded lengths up to about 20 m (70 ft.).
For comparison, Table 2 gives the reduction factors
from AASHTO. (Interestingly, AASHTO-LRFD uses
different reductions even though the traffic is iden-
tical.) It can be seen that the AASHTO factors are
conservative, as they should be. But it can also be
seen that using the AASHTO factors for short loaded
lengths is so conservative that it could result in a large
amount of very expensive strengthening that is in fact
not needed.
The parts of a suspension bridge governed by long
loaded lengths are the cables, towers and anchor-
ages. These will be at a maximum when all lanes are Figure 3. Stiffening truss, MacKay Bridge.
loaded, so in the example given in Table 1 the multi-
lane loading would be taken as 0.6 for these major started at the end of December 2009, and will continue
components. for about a year. This should indicate the combined
For some short loaded lengths, however, such as effects of both single-lane loading and multi-lane
govern transverse floor beams or trusses spanning presence, without identifying them separately.
between cables, a multi-lane load factor of 0.6 would
produce a design loading of nearly double the observed
load, and a value of 0.35 would be more appropriate. 5 SAFETY FACTORS
This does increase the complexity of the analysis,
but if unnecessarily doubling the design load results Having looked at the capacity of the bridge and deter-
in strengthening of all the transverse floor beams or mined the loads, the question arises as to what load
trusses, the cost would be enormous. Compounding (safety) factors to use. LRFD design has a huge advan-
the problem, the extra strengthening may increase the tage for suspension bridge analysis. The principle of
weight of the suspended structure to the point that the LRFD (known as Limit States Design outside the US)
cables, towers and/or anchorages may need upgrading. is that safety factors are applied more rationally than
This example makes a very compelling case for they are with Working Stress Design. This is because
going back to the fundamentals! a safety factor is really an ignorance factor: it covers
what we do not know about a load or resistance. It
therefore follows that if we can lessen our ignorance,
4 INSTRUMENTATION we can lessen the safety factor that must be applied.
Thus, loads that we know well, such as dead load, com-
In a recent case, there was concern that the live load of mand a small load factor, and those loads that we know
the A. Murray MacKay Bridge in Halifax, NS, (Fig. 2) much less about, such as live loads or wind loads, are
was being over-estimated by the design code. given larger load factors.
The possibility of measuring traffic axle loads This concept is particularly useful for suspension
directly and doing a study as in the previous exam- bridges, which have different components governed by
ple was considered for the MacKay Bridge, but it was different types of loads. For example, the main cables
decided instead to instrument directly the members of are dominated by dead load, and only 10 to 30% of
most concern, which were the diagonals of the stiff- the total load that they carry is live load, whereas the
ening trusses as shown in Figure 3. Recording of data stiffening truss or box girder is essentially unstressed
13
Figure 6. Distribution of load effects.
Figure 4. Distribution of measured values for yield values

of A36 steel with a nominal yield of 248 MPa (36 ksi).
Figure 7. Distributions of load and resistance combined.
Figure 5. Stylized distribution of yield strengths.
under dead load only, so almost 100% of the applied

load is live load.
Towers are a mixture: the axial load in the tower
legs, coming from the cables, is almost all dead load,
but the displacement of the tower-top is caused by live
load. Thus if one thinks of the P- effect, P, the axial
load, is governed by dead load, and , the displacement,
is governed by live load. Figure 8. Distribution of (resistance minus load).
overloads occur, so conceptually we have a distribution

5.1 Loads and resistances
of load effects as shown in Figure 6.
Let us examine the concept of safety in slightly more Now the two distributions of load and resistance can
depth, particularly as it relates to load factors. be combined, as shown in Figure 7.
Think first about the resistance or strength of a piece The distance between the nominal load, L, and the
of steel or concrete. Figure 4 shows a distribution of nominal resistance, R, in Figure 7 is the safety fac-
measured values for the yield of ASTM-A36 steel, tor, being the combined effects of load and resistance
which has a nominal yield of 248 MPa. Clearly the factors.
actual yield strength of the material is not what is speci- As long as the resistance is always greater than the
fied. It is usually greater, but it could, on rare occasions, load effect, all will be well, but it can be seen that the
be less. two curves overlap, which means that it is possible for
Figure 5 shows a more stylized distribution of the load effect to exceed the resistance, which leads
nominal strengths of steel or concrete. to failure. It can also be seen that if the two curves
Now consider the applied live load. are further separated, the area of overlap is reduced
The bridge design code gives us a design live load, and the probability of failure is reduced; but it can not
such as an HS vehicle or a lane load. The design load be reduced to zero if there are no known limits to the
is usually not representative of any real vehicle, but curves, and a quantitative value of the probability of
it is intended to represent the effects of real vehicles. failure is not easy to derive. If, however, we produce a
But how accurately do we really know the effects of distribution of R-L (resistance minus load), the curve
real vehicles? And even if we do know them well, of Figure 8 results.
how confident are we that they will not change in In this case the measure of safety can be seen more
the future? The answer is that usually the design load clearly. The probability of failure is the area to the
gives a greater effect than real vehicles, but sometimes left of zero divided by the total area under the curve.
14
A measure of safety (as opposed to concentrating on
the probability of failure) is often represented by , the
safety index. is the number of standard deviations
that the mean of the distribution is from zero. The
greater is, the greater the safety and the lower the
probability of failure.
From this approach, once the target probability of
failure has been set (or the factor established) the
required separation between the nominal (design) load
and the nominal resistance can be defined. The sepa-
ration is the combination of the load factor and the
resistance factor, in general: the safety factor.
5.2 Variable safety Figure 9. Shims between new floor beam and existing
girder.
The issue of safety has been discussed in some detail
so that we can introduce the concept of varying the The consequences of having short and/or restricted
load factors for different members. occupancies for construction are:
Risk (perhaps the opposite of safety) can be 1. Any retrofit must be designed to be constructed dur-
defined as: ing the short occupancies permitted, and shall be
Risk = (Probability of an event occurring) (Con- safe for the public between construction periods;
sequences of such an event) 2. The design must be tolerant of problems that may
Therefore to have consistent risk (or safety), the be found during the construction occupancy. Gen-
greater the consequences of an event, the lower the erally these will be unexpected, or they would have
probability we should accept of the event occur- been planned for, and they can include finding that
ring. Conversely, if the consequences are reduced, the the structure is not as was expected, breakdown of
acceptable probability can be greater. the contractors equipment, errors in design or con-
Consider the tower of a suspension bridge. If it struction, and non-delivery to site of components
fails, the entire bridge collapses. Then consider a steel or consumables, such as weld rod.
stringer that supports a steel or concrete deck. It will 3. The designers responsibility is to provide a design
probably fail in bending, and the load will be shed to that is as tolerant of unexpected problems as pos-
adjacent stringers, with the result that there is a tem- sible. A simple example is providing shims in a
porary dip in the roadway until it is fixed. The two connection so that if the existing steel is misaligned,
consequences are completely different, and it is clearly the new steel can be leveled to the correct eleva-
desirable to have less probability of tower failure than tions, as shown in Figure 9. Here, the new transverse
of stringer failure. In other words, the safety (load floor beam (framing in from the left) was to sit
factors) for the two members should not be the same. on the old longitudinal girder. Instead of trying to
Properly considering these differing requirements make the connection directly, a gap was provided,
can result in large cost savings during a retrofit. The to be filled by steel shims. Thus if the elevations or
subject is well covered in Section 14 of the Cana- alignments were not as expected, the shims could
dian Highway Bridge Design Code (2006) and its be adjusted accordingly.
Commentary. 4. The contractors responsibility is to provide a higher
level of management than for a typical project. If
work is only to proceed at night, the contractor has
6 PRACTICAL CONSIDERATIONS two deadlines per day one to be ready to start work
at a certain time, and the other to have the bridge
The discussion so far has been rather theoretical, ready for traffic on time. This requires considerably
although I maintain that a thorough understanding of more attention to scheduling than the conventional
safety is fundamental to safe and economical retrofit project that must simply meet the one deadline of
of any major bridge. substantial completion.
There are also, however, a number of practical 5. An aspect that is too often overlooked is that most
matters to be considered. bridge failures occur during construction by a fac-
tor of about a hundred compared to finished bridges.
In the case of bridge retrofit, the public is using
6.1 Bridge occupancy and traffic disruption
the bridge during the reconstruction process, which
There have been a few cases of suspension bridges means that the public is placed at a much greater
being closed to traffic for months on end while ren- risk during reconstruction than it would normally
ovations are made. These are rare, however, and it is be. The answer to this is not to increase the safety
much more likely that construction operations will be factors; it is to increase the quality and quantity of
limited to only one or two lanes at a time, or only independent checking of the proposed construction
at night or weekends, or some combination of these. activities, to a higher level than normal.
15
Figure 12. Port Mann Bridge, BC, Canada.
Figure 10. Original cross-section, Lions Gate Bridge.
Figure 13. Partial cross-section of Port Mann Bridge

approach spans, original material shown grey, new widening
and inspection walkways in black.
6.4 The devil in the details

Details are typically connections, and field splices
have to be capable of fast connection because of the
limited time available, and they must be tolerant of
misalignment and other potential problems.
Unfortunately, details are also where problems can
be generated. Too often, details are designed in such
Figure 11. Renovated cross-section, Lions Gate Bridge. a manner that secondary strains are induced, and
fatigue cracking follows a few years later. It is fun-
6.2 Maintenance and access damentally important to consider not just forces in
a connection, but also the structural behavior. This
Although the primary purpose of retrofit is usually
is particularly true for orthotropic decks, which are
to restore or increase live load capacity, the opportu-
becoming increasingly, and in my view, unnecessa-
nity should be taken, wherever possible, of improving
rily expensive, without a commensurate increase in
access for inspection, and reducing the cost of main-
performance.
tenance. For example, when the Lions Gate Bridge
was widened by 47% (Figs. 10 and 11), the amount of
steel surface to be painted, instead of increasing, was REFERENCE
actually reduced by half.
When the Port Mann Bridge (Fig. 12) near Vancou- Buckland, P.G., Navin, F.P.D., Zidek, J.V. and McBryde, J.P.,
ver, BC, was widened, the opportunity was taken to add 1980, Proposed vehicle loading of long span bridges, Jour-
inspection walkways to the outer plate girders of the nal of the Structural Division, American Society of Civil
approach spans as shown in Figure 13. These have been Engineers, April, p 915.
invaluable in easing inspection of this ageing bridge
prior to its replacement in the next few years.
6.3 Other benefits

In addition to improved maintenance, other benefits
may sometimes be achieved with little extra cost, such
as enhanced seismic resistance.
16
Modern bridges in China
M. Feng
Ministry of Communications, Beijing, China
ABSTRACT
In the past 30 years, especially during recent ten

years, China has deployed the worlds largest scale
construction of road and bridges. Bridge construction
technology has been improved rapidly and greatly to
achieve more and more breakthrough in large spans.
Bridge construction has concentrated in the three main
regions, i.e. middle and lower reaches of the Yangtze
River, middle and lower reaches of the Pearl River as
well as the Yangtze River and the Pearl River Delta
Region. (Fig. 1).
From 1979 to 2008, 16,000 bridges (with total
length of 732 km) were built each year. By the end Figure 2. Xihoumen Bridge with 1650 m span.
of 2008, highway bridges built in China had reached
595,000 (with total length of 25247 km). 18 bridges with main span over 600 m had been
Bridges in China can be divided into three types, completed and 15 bridges with main span over
namely girder bridges, arch bridges and cable- 600 m are under construction.
supported bridges (including cable-stayed bridges and 10 bridges with main span over 800 m had been
suspension bridges). Girder bridges and arch bridges completed and 9 bridges with main span over 800 m
make up almost 70% and 30% respectively of all are under construction.
bridges built in China while cable-stayed bridges 5 bridges with main span over 1000 m had been
occupy less than 1% of total amount. completed and 5 bridges with main span over
With the focus of bridge construction shifting from 1000 m are under construction.
upper and middle reaches of rivers and lakes to lower
reaches, bay and strait, China will face challenges to The maximum main span of completed girder
apply advanced technology in bridge construction to bridges, arch bridges and cable-stayed bridge has
deal with more complicated meteorological, hydro- reached 330 m, 552 m and 1088 m, ranking the first
logic, navigational and geologic conditions and realize of same type of bridge in the world while that of the
a breakthrough in larger spans. suspension bridge has amounted to 1650 m, ranking
By the end of June, 2008 (statistics excludes Hong the second in the world.
Kong, Macao and Taiwan): This article introduces briefly the latest technical
achievements for the construction of highway bridges,
55 bridges with main span over 400 m had been the status quo of bridge maintenance and manage-
completed and 20 bridges with main span over ment as well as the planned construction tasks in the
400 m are under construction. mainland of China.
China has to face new technological challenge to
achieve new technological breakthrough of integrated
system, combined structure and compound materials
in light of bridge construction characterized by larger
span, sea-crossing long bridge, deep-water foundation
and extra-high bridge tower.
Heading to the golden age for bridge construc-
tion, the Chinese engineers are well prepared and will
cooperate with our colleagues from all over the world
to meet the challenge of the largest scale bridges
construction of the world in the new century with
new structure, new material, new technique and new
equipment as well as innovative management.
Figure 1. National expressway network.
17
Cable supported bridges design, maintenance, rehabilitation and

management
J. Sandager Jensen
COWI A/S, Kongens Lyngby, Denmark
ABSTRACT: This paper presents experience collected by engineers from COWI during involvement in design,
maintenance, rehabilitation and management of several major cable supported bridges all over the world. First,
the major problems encountered during the exploitation of major existing cable supported bridges are discussed.
Then, several aspects of maintenance and management of such bridges are presented. Finally, the main issues
related to the proper design and rehabilitation of major cable supported bridges are analyzed. At the end of the
paper several case studies related to the maintenance, management, rehabilitation and design of major cable
supported bridges are shown.
1 INTRODUCTION Hga Kusten Bridge, in Sweden. The third case study

is the replacement of bearings and expansion joints
Cable supported bridges are usually crucial elements in Faroe Bridges, in Denmark. The following is the
to the road and railway infrastructure. Very often replacement of expansion joints, surfacing and corro-
they constitute a part of critical links between highly sion protection at New Little Belt Bridge, in Denmark.
habited areas. As a consequence their closure or traf- Afterwards, the case study showing the assessment,
fic capacity reduction causes major inconveniences for monitoring and evaluation of hangers vibration in the
the users and result in significant losses to the economy. Great Belt Bridge in Denmark and the development
Furthermore, construction, maintenance and rehabil- of the management system for heavy transport at this
itation of major cable supported bridges are rather Bridge is included.
costly. Therefore, the design, maintenance, manage-
ment and rehabilitation of these structures are complex
and challenging tasks. Consequently, they have to be
2 SELECTED MAINTENACE PROBLEMS IN
performed using state-of-the-art technical solutions
MAJOR CABLE SUPPORTED BRIDGES
and the best practice gained over many years of expe-
rience. Also, they should account for the results of
2.1 Deterioration of main cables and hangers
Life Cycle Cost analysis and the interferences with
the bridge function. The main cables and hangers are the essential elements
This paper presents the experience collected by of a cable supported bridge. At the same time they are
engineers from COWI during involvement in design, also the most difficult element to inspect and maintain.
maintenance, rehabilitation and management of sev- Therefore appropriate attention should be paid to the
eral major cable supported bridges all over the main cables.
world. First, the major problems encountered dur- Deterioration of main cables on suspension bridges
ing the exploitation of major existing cable supported is in practice always associated with corrosion
bridges are discussed. Then, several aspects of mainte- (Srensen et al. 2006). Deterioration of hanger cables
nance and management of such bridges are presented. and stay cables are always associated with corrosion or
Finally, the main issues related to the proper design vibrations or a combination of these. The basic cause
and rehabilitation of major cable supported bridges of corrosion (see Fig. 1) is the presence of water or
are analysed. moisture.
At the end of the paper several case studies related The atmosphere at the bridge site often contains salt
to the maintenance, management and rehabilitation of spray from the sea or from deicing salt, as well as other
major cable supported bridges are shown. The first pollutants from vehicles or industry. These elements
case study is the rehabilitation of the Aquitaine sus- will eventually enter the cable in connection with rain
pended bridge, in France, involving replacement of or vapour, which can enter the cable through various
the main cables and widening of the bridge deck from openings. It causes corrosion attack on the wires.
4 to 6 lanes. The second case study is the installation of Worldwide there are many examples of serious cor-
the dehumidification system for the main cables at the rosion problems, even on relatively new bridges. For
18
Figure 1. Details of corroded strands.
newer bridges the problem can be even more seri- bridges loss of coating and corrosion is still a major
ous, as the design safety factor for the main cables cost problem.
has been progressively reduced over the years. It has
been demonstrated by many examples that a tradi- 2.3 Deterioration of bridge deck surfacing
tional system can slow down corrosion, but it can not
Modern cable supported bridges being part of road and
prevent it.
railway infrastructure are to a large extent designed
All these factors have led to the development of
using the orthotropic steel decks.
a new corrosion protection system for main cables,
In the case of cable supported railway bridges with
which is based on dehumidification (Bloomstine et al.
orthotropic steel decks the track support details are
2006). In contrast to a traditional system, a dehumidi-
typically consisting of a slab track solution or a direct
fication system eliminates the source of the corrosion
fastening system. Both solutions allow a controlled
problem water/moisture in the cables. Hence, a
load transfer to take place to the bridge superstructure
properly designed dehumidification system provides
with minimum fatigue problems. The main challenge
complete corrosion protection of the main cables, as
is to accommodate secondary loadings (non rail)
they are enclosed in an atmosphere with a sufficiently
and to ensure an efficient waterproofing at the inter-
low relative humidity, such that corrosion can not
face between the direct fastening/track slab and the
occur.
remaining part of the orthotropic bridge steel deck.
The roadway surfacing on modern orthotropic
2.2 Loss of paint and deterioration of the
bridge steel decks are generally composed of asphaltic
bridge deck
layers with a total thickness between 40 and 60 mm.
All steel elements exposed to the normal environmen- On the Great Belt East Bridge in Denmark the surfac-
tal conditions suffer from corrosion due to presence of ing under the traffic lanes consists of (from bottom to
oxygen and humidity. Steel elements of major bridges top):
are usually exposed to much more harsh conditions
4 mm mastic
of marine environments, where the presence of chlo-
25 mm intermediate layer of mastic asphalt
rides accelerates the corrosion process.Therefore, they
30 mm wearing layer of mastic asphalt
have to be well maintained and repainted every time
the coating losses its protective capabilities. This type of asphalt surfacing has been used on
Service life of coating in aggressive environments orthotropic steel bridge decks in Denmark since 1970
is often limited to about 25 years. This is a short period and has proven to be very durable (see Wegan and
compared with the service life of a major bridge, which Bloomstine 2004).
usually is designed for 120, 150 or even 200 years Road surfacing on movable bridges with orthotropic
of service life. Is it possible to improve this in order steel bridge decks with special requirements to weight
to avoid the costly repainting works which often also has been carried out using polymeric resin surfacing.
cause some traffic disturbances? The reduction in weight is a must from a functional
The problem of deterioration of steel elements point of view and it is expected to be the next step
has been partially solved in modern cable supported into optimization the design of orthotropic bridge
bridges with box girders, where dehumidification sys- steel decks for long span cable supported bridges.
tems for the interior of the box girder are installed. Some of the challenges to be overcome are fabrication
These mechanical systems keep humidity level inside tolerances and fatigue of steel.
the girder at such a level that corrosion of the steel does As regards practical experience of both mastic
not propagate. Also, in some modern bridges robots asphalt and epoxy asphalt roadway surfacing the five
for performing repainting works of the outside sur- U.K. bridges built during the 1960s and 1970s with
faces of the decks have been installed, which speed up orthotropic steel bridge decks provide valuable lessons
and simplifies the process. However, in many existing learned. (see McFadyen & Blumensen 2010).
19
Table 1. Expansion joint movements.
Additional mechanical Continuous Maximum movement Yearly accumulated

Bridge systems length at joint movement at joint
Great Belt Suspension Hydraulic system (buffers) 2694 m 1.0 m 100200 m

East Bridge Without hydraulic buffers 2694 m 1.0 m 50,000 m
Great Belt Steel Approach 2518 m 0.7 m 100 m
spans Halsskov Bridge
Standard steel beam bridge 50 m 0.02 m 5 m
2.4 Wear and malfunction of bearings, expansion 2.5 Vibrations of cable stays and hangers
joints and other movement devices
Vibrations of stay cables and hangers under the com-
Many years of experience of design and maintenance bination of rain, wind and frost are observed at many
of bridges show that bridge elements allowing for major cable supported bridges (Laursen et al. 2006).
movements, such as bearings, expansion joints, etc., This phenomenon is not yet well understood and due
are among the more expensive as far as maintenance to this fact it is quite difficult to control. Although it
costs are concerned. These mechanical devices require is rather impossible to eliminate completely the cable
regular maintenance and partial or full replacement vibrations, it is important to minimize their magnitude
several time during the bridge service life. Movement to such extend that the structural integrity is main-
elements generally represent a minor portion of the tained, and failure due to e.g. fatigue will not occur.
total budget for construction of a new bridge. During Sudden rupture of cable might put in risk safety of the
the service life this ratio is generally quite different whole structure and will cause disturbances to the traf-
and the cost for maintenance, repair and replacement fic when the replacement work will have to be carried
of movement elements is most often a major expense out. Consequently, the large vibrations of stay cables
seen in relation to other maintenance costs (Srensen and hangers have to be avoided.
et al. 2007). Several vibration mitigation alternatives may be
The movements of a major cable supported bridge used, from the most simple as the cable separators,
are naturally larger than on beam bridges with spiral ropes, wind ropes, to the more elaborated damp-
traditional spans. Suspension bridges in particular ing systems with hydraulic dampers or tuned liquid
may experience major longitudinal displacements and dampers. Although previous experience is helpful in
specifically a very large accumulated movement which choosing the appropriate vibration mitigation devices,
makes specific demands on durability. To illustrate in some situations trial and error application supported
this, the annual accumulated movement of an expan- by long term monitoring or frequent observation will
sion joint on different bridge types is presented in be necessary.
Table 1.
In Table 1 the maximum movement and yearly accu-
mulated movement at joint, for the case of the Great
3 OPERATION, MAINTENANCE AND
Belt Suspension Bridge are presented for the struc-
MANAGEMENT OF MAJOR CABLE
tural system with and without hydraulic buffers. The
SUPPORTED BRIDGES
buffers reduce bridge fast movements caused by traffic
or wind allowing in the same time for slow move-
3.1 General
ments due to temperature and static loads. Wear of
expansion joints is determined not only by accumu- Major cable supported bridges are normally planned
lated movements but also by other phenomena such as to be maintained for a very long period of time: 75, 100
frost, rain, traffic intensity, dynamic interaction, etc. or 120 years. Some bridges have been even assumed to
However, the above relative comparison indicates that be in service for more than 200 years. During all this
expansion joints and bearings for major cable sup- period, due to limited possibility of traffic diversion (or
ported bridges are more prone to wear and damages even lack of any), the traffic on major bridges has to be
than those applied in standard bridges. provided with the minimum possible interruption. Due
During several years of experience in mainte- to all above mentioned considerations and having in
nance of major cable supported bridges COWI had mind the fact that the rehabilitation or replacement of
to deal several times with rehabilitation and replace- such structures is extremely difficult and costly, special
ment of bearings and expansion joints, sometimes attention has to be paid to their operation, maintenance
only few years after bridge completion. The lesson and management.
learned from it is that bridges should be arranged Generally, the entire service life of a bridge con-
with as few movement elements as possible. Further- tains of two different subsequent stages (see Fig. 2).
more, all of them have to be maintained regularly The first stage is the normal stage of operation, where
and inspected much more often than other bridge just regular inspections are undertaken, some regular
elements. testing and monitoring is performed (if necessary) and
20
information regarding bridge condition, safety, func-
tionality, etc. All this information can be obtained
due to inspections, condition assessment, performance
assessment and testing and monitoring of bridges.
Some of the mentioned activities are carried out regu-
larly while others are performed just once or few times
along the whole life of the structure.
All the above mentioned activities (i.e. inspections,
condition assessment, performance assessment, test-
ing and monitoring) have very important economical
and functional impact on the operation, maintenance
and management of existing bridges due to the fact
that they may postpone the costly repair, retrofitting or
replacement of the structure and thus also reduce the
traffic interruption. COWI was recently involved in the
development of several Guidelines covering all these
issues for railway bridges (SB-ICA, 2007; SB-LRA,
2007; SB-MON, 2007; SB-STR, 2007).
3.2 Inspections
Effective maintenance management of major bridges
has to be based on their actual condition. Appropri-
ate scheme of carefully performed bridge inspections
allows to identify damages, malfunctions and wear of
several bridge elements and ancillary items (as has
been discussed in section 2) before any risk may arise
to the bridge users or to the structure itself.
Figure 2. Stages and activities during entire life of bridges. The inspections scheme for major bridges is usually
similar to the general scheme used for the inspection
of all bridges in the country and most often consist of
some minor maintenance actions are undertaken (e.g. three (eventually four) levels of inspections:
painting, lubricating bearings, etc.).
The second stage is a special stage which starts 1. Routine inspections every half/one year
when some concern regarding the safety or the func- 2. General inspection every 56 years
tionality of the bridge arises and the assessment of 3. Special inspections whenever required
the bridge have to be performed and/or some more During inspections of major bridges, much more
costly maintenance action have to be undertaken, as attention has to be paid to the ancillary items as
for example repair, rehabilitation or strengthening. In bearings, buffers, expansion joints, etc. than it is in
order to extend the service life of a bridge into the the case of ordinary structures. This is due to fact that
maximum, considering the functional and economi- their malfunction may cause major problems. Further-
cal constraints, several maintenance actions have to more, great attention has to be paid to the corrosion
take place within its lifetime. Depending of the type protection and vibration mitigation of main cables and
of maintenance action, they might: slow down the hangers.
rate of deterioration (routine maintenance), eliminate
the source of deterioration and/or restore capacity
3.3 Testing and monitoring
loss (repair or rehabilitation), upgrade the capacity
or the functionality of the bridge (strengthening or Several testing and monitoring techniques, developed
retrofitting). in last decades, have found application in broadly
The decision regarding what kind of maintenance meant operation and maintenance of some major
action to undertake and when to do it in order to mini- cable supported bridges. Testing and monitoring are
mize the maintenance costs and maximize the bridge very suitable to supplement visual inspections with
performance should be taken according to the main- some more subjective data. Moreover, continuous
tenance strategy established for the group of bridges monitoring techniques enable, to some extent, reduce
(e.g. along the whole railway line) or for the specific scope or the intervals between the visual inspections.
bridge (e.g. for some substandard bridges). Very important feature of the monitoring is also the
Bridge Management Systems (BMS) support the ability of the immediate warning whenever some min-
bridge owners and operators in their choice of opti- imum required conditions are not fulfilled and the
mum maintenance strategy that are consistent with the safety of the users or the bridge itself is in hazard.
owners policies, long-term objectives and financial Due to above mentioned, monitoring systems should
constraints. Obviously, any BMS, in order to sug- become the standard equipment of modern major
gest optimal decisions, requires adequate and reliable cable supported bridges, however selected testing
21
Figure 3. Basic elements of DANBRO + Bridge Management System (BMS) (Bjerrum et al. 2006).
methods should be used for the bridge inspections on age of the computers, a basic equipment of major
a routine basis. bridges. The systematic maintenance of major bridges
involves an extensive list of technical, traffic and
administrative activities. Therefore, an IT Bridge Man-
3.4 Maintenance activities agement System (BMS) has to be able to cover a
Major cable supported bridges require clear mainte- number of different disciplines such as maintenance,
nance strategy based on the: organization management, surveillance, inspections,
rehabilitation, traffic control, priority ranking, budget
1. Actual bridge/element condition identified based and quality control. The purpose of the manage-
on inspections combined with non-destructive test- ment system is to optimize and support the daily
ing and optional monitoring administration of bridges including systematization
2. Forecast of deterioration based on the previous and quality assurance of the management of the struc-
records, experience and theoretical models tures carried out by various engineers and technicians.
3. Identification of the importance (e.g bridge ele- A long cooperation of COWI with Danish Road
ment or ancillary item) to the global safety and Directorate during the development and application of
operability BMS in the maintenance management of large bridges
Maintenance and repair work in major bridges in has lead to the following conclusion. The effective
general should be carried out before deterioration BMS for major bridges has to be user friendly, pre-
develops to a major extent. Such a preventive mainte- senting data in an effective way, updating data and
nance policy reduces the risk of large scale repair work documents in an easy way and having all informa-
and herby minimizes the user inconveniences. From tion located in a single place. This is required due to
the experience of COWI in the maintenance manage- the number of disciplines, people and organizations
ment of several large bridges it can be concluded that involved. The system has to be capable of supporting
the rehabilitation of such structures is extremely costly several activities as presented in Fig. 3.
and that preventive maintenance policy is the most
appropriate. This of course does not have to be the
case for small bridges of low importance. 3.6 Managing heavy transports
In recent years heavy (overweight) transports have
increased significantly and thereby the need for effi-
3.5 Bridge management systems
cient overweight permit systems is more important
An effective management system that allows for an than ever to prevent bridges from overloading. In
easy access to updated information for bridge opera- Denmark the increase is first of all a result of the
tors, consultants and contractors is nowadays, in the quick-growing wind power industry.
22
Figure 4. Principles for application of heavy transport permission (Ravn et al. 2006).
The management of heavy transport on standard and restricted classes are taken into consideration as
bridges with spans up to 50 m is made usually by use well as vehicle classes on critical span lengths (see
of simple rating system, where all bridges and vehi- Fig. 4).
cles are rated according to the maximum capacity and Major cable supported bridges require refined rating
weight respectively (Ravn et al. 2006). The vehicle principles but the heavy transport management system
rating estimates the load effects which the transport can be similar as it is for the standard bridges. The
exposes to the bridge while the bridge rating esti- method used in Denmark for major bridges applies
mates the load capacity of the bridge. In Denmark the actual overweight vehicle for rating of critical
the standard rating classes are from 10 to 200 tones. structural elements. As the heavy transport is used
The classification of the bridge is determined by the directly in the evaluation for the particular bridge,
heaviest standard truck that crosses the bridge under the uncertainties introduced as consequence of the
specified conditions i.e.: vehicle rating are eliminated and the results are more
accurate.
1. Normal class Heavy transport and normal traffic,
The Danish approach for administrating overweight
whole bridge area may be used, no restrictions;
vehicles was developed keeping in mind that the rat-
2. Normal class: Heavy transport and normal traffic,
ing systems are used numerous times every day and
whole bridge area may be used, no restrictions;
hence shall be cost effective and at the same time
3. Restricted class 1 As normal class but only the
assure the best utilization of the load bearing capacity
roadway may be used;
of the bridges on the Danish road network. Practi-
4. Restricted class 2 As restricted class 1, but the
cal experience has proved that both goals have been
speed must not exceed 10 km/h;
achieved.
5. Restricted class 3 The heavy transport is the only
transport on the bridge, low speed of max. 10 km/h
and driving in the least critical position. 4 DESIGN, ASSESSMENT AND
REHABILITATION
The normal classes are visualized on maps on
the Internet, called the Heavy Load Grid. The road
4.1 Design of new bridges
stretches in the grid are given a classification equal to
the lowest Normal Class bridge present on the actual The design life requirement for new major bridges is
stretch. The maps are for daily use by police and road constantly increasing. It is often more than 100 years
administrators. Restricted classes are only used by and in some cases as high as 200 years, e.g. the Messina
bridge/road administrators, in cases where the vehi- Bridge. The design traffic load is also increasing with
cle class is higher than the bridge classes shown on regards to intensity and weight. Therefore it is neces-
the map. sary to especially focus on the durability of the bridge
The vehicle class is calculated for simply sup- elements that are directly affected by traffic.
ported bridge spans from 2200 m. The vehicle class Requirements concerning the operation period of
is determined by reference to predefined standard the bridge are also becoming more common as bridge
trucks labeled by their overall weight. The vehicle class owners are starting to realize that a low tender price
equals the standard truck that produce the same load does not usually give a low lifetime cost. Therefore,
effects as the actual heavy transport on the particular Life Cycle Cost (LCC) Analysis must be incorpo-
simply supported span. rated in the design to ensure a durable design with
Application for transport permission shall be for- the correct lifetime and the lowest possible Life Cycle
warded to the police and the police may issue per- Cost. The optimal solution with the lowest LCC must
mission if the maximum vehicle class for all spans is be determined for each bridge element, resulting in
lower than or equal to the bridge classes on the current the lowest LCC for the entire bridge. There is also
route. If not, bridge/road administrators are involved increasing focus on availability of the passageway,
23
Figure 5. Step-level procedure for reassessment of bridges (SB-LRA, 2007).
meaning that traffic disruptions are not allowed. The state of knowledge. Also, in the same manner, the re-
design must therefore allow inspection, maintenance qualification actions (strengthening and repairs) may
and repair without disturbing the traffic. Furthermore be evaluated, compared and selected. It should, how-
access facilities for all structures should be incorpo- ever, be noticed that economical considerations alone,
rated in the initial design, as this allows fulfillment may not be sufficient for re-qualification purposes as
of the aforementioned requirements with regards to explicit requirements to the safety of the bridge are
traffic and LCC. often dictated by legislation.
Key areas for future bridge design are therefore Fig. 5 shows the step-level procedure recommended
durability/low LCC, accessibility and availability. in the SB-LRA (2007) Guideline to be used in the pro-
cess of assessment of existing bridges. Considering
4.2 Assessment of existing bridges the above discussed topics, in the presented proce-
dure, the knowledge about the bridge is established
Assessment of an existing bridge with the purpose of
and refined in an adaptive manner according to the
re-qualifying the bridge for increased loading and/or
actual needs.
for prolonging the service life may be seen as an
As it can be seen in Fig. 5, an assessment of existing
adaptive, step-level process of refining the state of
bridges in the proposed procedure is divided into three
knowledge regarding the present and the future state of
levels, which in terms of refinement and detailing can
the bridge and its behavior. An assessment may involve
be characterized as follows:
a review of project documentation, inspection of the
structure, testing of materials, testing of structural per- 1. Purely heuristic experience or code based state-
formance, refined numerical analysis and planning of ments (initial assessment);
future inspections. 2. Application of deterministic and semi-probabilistic
The decision on whether or not to collect more safety formats (intermediate assessment);
information is always based on the existing informa- 3. Instrumentation, testing and/or probabilistic analy-
tion (prior information) and the expected reduction ses (enhanced assessment).
of the life cycle cost obtained on the basis of the
additional information. Depending on the actually Generally, an assessment can be carried out within
achieved knowledge (posterior information) it may or the framework of these three phases. However, the
may not turn out to be feasible to refine further the level of detail within each phase may vary. In this
24
way it is possible to tailor a reassessment for differ- variability in modeling loads and resistance proper-
ent purposes. The level of detail of the assessment is ties, and taking into account bridge redundancy and
recommended to be chosen considering the detailed system behavior.
assessment objectives for the particular bridge and its The sensitivity analysis, performed during the
characteristics. assessment, may help to identify where the refinement
According to the presented step-level procedure, the of the knowledge about the bridge may be the most
capacity of the bridge in cause is initially assessed beneficial for the assessment of the bridge in cause.
on the basis of simple calculation checks and readily As already discussed, such refinements may be based
accessible data (drawings, design calculations, earlier on detailing of the analysis methods and/or further
assessment calculations, inspection records, etc.). On collection of data.
this basis, the extent to which the bridge fails to comply
with the given requirements is evaluated.
4.3 Rehabilitation
In the intermediate level of assessment, the capacity
of the bridge (which fails the initial assessment) is eval- Rehabilitation of bridges should generally take into
uated using more advanced analysis (e.g. elastic but account the same requirements as mentioned above
giving better idealization, plastic, etc.) and more accu- for new bridges. Furthermore, before starting rehabili-
rate data (obtained from inspection an simple tests) tation design it is essential to carry out inspections
on the material properties, the loads, the current state and investigations to determine the cause of the dam-
and the behavior of the bridge (e.g. material properties ages/degradation, such that the rehabilitation design
obtained from simple measurement, loads defined by can prevent similar problems in the future. New and
measurements, etc.). durable details should be developed to replace the
Finally, in the enhanced level of assessment, the original design.
capacity of the bridge, which fails the intermediate During rehabilitation it is essential to focus on pre-
assessment and where repair or strengthening costs vention of traffic disruptions. Many bridge owners
are significant, can be evaluated using most advanced have begun requiring this to be a part of the LCC
assessment methods (e.g. reliability-based assessment Analysis by setting a price on traffic delays.
methods, system level assessment, etc.) and tools Many older bridges do not have safe and/or suf-
available (e.g. non-linear analysis, probabilistic anal- ficient access facilities. A rehabilitation project is
ysis, testing, monitoring, etc). Testing and monitoring an ideal opportunity to upgrade these facilities and
may provide relevant data regarding actual bridge an upgrade will decrease future maintenance costs.
loads, actual properties of material, and actual behav- Access facilities should therefore always be considered
ior of the bridge. However, probabilistic analysis and in connection with rehabilitation.
non-linear analysis allow for considering the actual
25
Lifetime design of cable-supported super-long-span bridges
H.M. Koh, H.J. Kim & J.H. Lim

Department of Civil and Environmental Engineering, Seoul National University, Seoul, Korea
S.C. Kang
Steel Structure Research Laboratory, Research Institute of Industrial Science & Technology, Gyeonggi-do, Korea
J.F. Choo
Department of Civil Engineering, Konkuk University, Seoul, Korea
ABSTRACT
However, design codes are not specifying explicitly
the concept of lifetime design.
The main issue when deciding upon a specific ser-
The AASHTO LRFD Bridge Design Specifications
vice life is to clarify the event, which will identify the
define service life as the period of time that the bridge
end of the service life. However, the lack of speci-
is expected to be in operation. The design life is defined
fication provisions related to the length of service life
as the period of time on which the statistical derivation
in the codes constitutes an obstacle to the realiza-
of transient loads is based. Though the specifications
tion of extended service life for bridges. This situation
prescribe transient loads based on a design life of 75
becomes more acute in the case of super-long-span
years, they are not giving any basis or explanation on
bridges because of their nature. In view of the cur-
the extent of the expected service life. In other words,
rent practice, service life of 100 to 150 years seems
the 75-year prescribed design life is an indicator of the
to be achievable and has already been implemented in
structural safety and reliability, while the service life
recent major bridges owing to special specifications by
is an indicator of the durability. However, the speci-
the owners of the bridge projects. This paper reviews
fications do not prescribe the expected duration for
current practices for service life design with some
service life nor give direct correlation between design
recent examples in Korea and addresses issues that
and service life (Bartholomew 2007, 2009).
will be faced in the lifetime design for cable-supported
It is noteworthy that lifetime design is a recent con-
super-long-span bridges. Limitations of the current
cept of which definition has not been clearly defined
practice are highlighted and, additional requirements
to date. The Roman builders did certainly not have
and research directions that will help designers in
imagined that many of their bridges would still remain
achieving extended lifetime of 200 years or more for
operational 2000 years after their construction. Roe-
the super-long-span bridges are suggested. Dedicated
bling, the designer of Brooklyn Bridge, also did not
research programs currently implemented in Korea are
apply the concept of design lifetime for his bridge,
also presented.
which is still one of the most important arterial road
linking New York and the district of Brooklyn even
after 126 years of service, while most of its con-
1 INTRODUCTION temporaries have collapsed or have been replaced
meanwhile.
1.1 Lifetime design, a definition On the other hand, recent literature sets a target
The purpose of lifetime design is to predict and opti- design lifetime for long-span bridges since the end
mize the lifetime quality of the designed structure. of 1990s. For example, the Great Belt East Bridge
Lifetime quality means the capability of the struc- (1994) in Denmark and the Confederation Bridge
ture to fulfill the requirements of its users, owners (1997) in Canada have been designed for a 100-year
and society over its entire planning and design life life expectancy, as well as the more recent Incheon
(Sarja 2010). Bridge (2009, Figure 1) and the Busan-Geoje Fixed
This sentence involves two major concepts that are Link actually under construction in Korea (Figure 2).
performance and design life of the structure, which are A useful project life of the Millau Viaduct (2004)
indeed closely related. In the case of bridge structures, crossing the Tarn Valley in France is 120 years. The
the required performances are given in the relevant Humber Bridge, the worlds longest single span sus-
design codes. These performances are currently spec- pension bridge has a design life of 120 years. And
ified in terms of allowable strength or limit-states. a record of 150 years design lifetime was set for the
26
longer than 1,000 m for cable-stayed bridges and
longer than 1,990 m for suspension bridges.
Differently from common bridges, long-span
bridges represent the unique access to remote places.
As a matter of fact, any disruption of such bridge for
repair or replacement will have tremendous effects on
the local economy and obstruct completely communi-
cations for relatively long period. In the case of such
super-long-span bridges, these effects will grow expo-
nentially since super-long-span bridges not only by
their length but also by their role as inter-continental
route will affect a wider area and a larger community.
In addition, the costs required for the construction and
maintenance of super-long-span bridges will also grow
Figure 1. View of Incheon Bridge linking Youngjong Island exponentially. Therefore, the lifetime design of super-
and the mainland (total length: 12.3 km, central span: 800 m, long-span bridges addresses issues of which feasibility
design service life: 100 years). and practicability should be discussed thoroughly.
The ability of the bridge to fulfill its intended func-
tions along its service life can be compromised due to
degradation. The ideal bridge any designer is dream-
ing of is a bridge that will stand for an infinite service
life with minimum maintenance or maintenance free.
However, such a bridge is not possible in practice as all
materials deteriorate gradually when exposed to envi-
ronment and traffic. Even though bridges are made
of different materials, have different structures, are
located in different locations and have different ages,
their common feature is their degradation with time.
Major causes of degradation are high transient loads
and severe environmental conditions.
A fatigue life of 200 years may be feasible through
Figure 2. View of one of the two cable-stayed bridges careful technical considerations involving high initial
(Geoga Bridge) of the Busan-Geoje fixed link under con- quality and adequate maintenance activities. Assum-
struction (total length: 8.2 km, central span: 470 m, design ing a fatigue life of the bridge of 200 years, the
service life: 100 years). probable range of the bridge life can be thought of in
the order of 100 to 400 years (Frangopol et al. 2000).
Akashi Bridge (1999) in Japan and the Oakland Bay Improving the design methods of bridge components
Bridge in San Francisco. is the basic step to achieve this target. Proper struc-
The question that arises now is the basis on which tural design addresses the effects of transient loads
such design lifetimes have been selected or computed. through adequate proportioning of the members and
A review of the relevant literature shows that the design details. In addition, well-coordinated design,
assessment of the design lifetime of these bridges relies construction and maintenance are essential to sustain
essentially on the lifespan of the involved materials. bridge structure with an extended service life. As men-
That is, if the durable life of concrete reaches 100 years tioned above, an important issue having determining
so does the structure. However, could we simply state influence on the service life is the material durability.
that durable lifespan of the materials is guaranteeing However, the design lifespan of the material consti-
equivalent lifespan of the whole structure? tutes a necessary but not sufficient condition in the
How can designers suggest reasonable service life extension of the service life of the structure.
for a bridge structure considering that the target or A typical example is the first Jindo Bridge, which is
expected service life relies on the owners desires or the first cable-stayed bridge in Korea and was also the
societal needs whereas the actual service life depends longest cable-stay span outside Europe at its comple-
on the quality of materials, design and construction, tion in 1984 (Figure 3). This bridge was designed with
the environmental conditions to which the structure is respect to a design live load DB-18, equivalent to the
exposed, and on the maintenance strategy? design load HS-18 in USA. In 1996, the close inspec-
tion of the bridge including loading and vibration tests
resulted in the closure of the bridge to heavy trucks
since the structure showed signs of fatigue and accel-
1.2 Lifetime design, the issues
erated degradation due to the increasing crossing of
In this paper, super long-span bridge refers to a bridge truck loads exceeding the design load. Thereafter, the
exhibiting span length equivalent or superior to the governmental authorities decided to improve the load
longest span lengths realized to date. That is span carrying capacity of the existing bridge while erecting
27
Accordingly, this paper reviews current practices
for service life design through recent examples in
Korea and addresses issues that will be faced in
the lifetime design for cable-supported super-long-
span bridges. Limitations of the current practice are
highlighted and, additional requirements and research
directions that are being discussed among Korean
experts to help designers in achieving extended life-
time of the structures are presented.
2 SERVICE LIFE AND DESIGN LIFE IN CODES

Figure 3. The First Jindo Bridge (right) and Second Jindo
Bridge (left). The bridge stock constitutes one of the most expensive
assets in the national infrastructure. Unfortunately,
bridge structures start to degrade since their construc-
a twin bridge design for vehicle load of DB-24 and tion. Therefore, reducing at the most efforts and costs
DL-24. This example shows clearly that, apart from to be invested in their future management so as to
material durability, the lifetime design of long-span extend their operational life must be dealt impera-
bridges should also consider future change of traffic tively since the design stage. In such aspect, defining
and loading in order for the bridge to provide extended or selecting a target service life is primordial to derive
useful service life. adequate specifications or regulations for durability
Another issue is the reliability of the service life design, fatigue design or lifecycle costs.
of the members predicted on the basis of the material In industrialized countries, the bridge stock has
strength. The project performance requirements (PPR) today reached severe degrees of degradation that are
of Incheon Bridge specified a design life of 100 years requiring immediate repair or replacement. Consider-
for the whole structure with a design life of 75 years ing that most of the bridges have been built in the sev-
for the major structural components and less than 75 enties and assuming that these bridges were designed
years for the secondary and replaceable components. to have service life of 50 years, a simple calculation
However, these values correspond to the target design implies that all these bridges should be replaced within
life which may be different from the actual service a short period of time, which means tremendous costs
life of the members. Any defect in a member of the induced not only by their reconstruction but also by
structure prior to the end of its target life will likely the indirect losses provoked by the subsequent closure
have effect on the neighboring members and endanger of traffic. Accordingly, extending the service life of
the integrity of the whole structural system. Need is bridge structures is a necessity not only in terms of the
thus to assess the eventual satisfaction of the target national economy but also in terms of safety.
design life for each of the members in order to achieve Accordingly, national codes are defining design life
the intended service life of the structure. and categories as shown in Table 1. Since service
Issue also is for the current design practice applied life involves consideration of many designs, mate-
for the safety under extreme events. A design life of rials, construction and environmental factors, these
200 years implies that the bridge will have larger prob- definitions of design life do not represent a basis for
ability to experience one or multiple extreme events service life. These specifications do not recommend
during its service life. For example, seismic design any specific period for service life.
considers a return period of 500 years or 1,000 years. On the other hand, some notable concrete bridges
Is the current practice still reasonable for a structure have been constructed recently using criteria devel-
that should stand operational for 200 years? If not, how oped to ensure service life of 100 to 150 years. Record
should we modify our design approach considering the service life criteria have been indicated by the 300-
fact that long-span bridges exhibiting longer lifespan year target service life of the Second Gateway Bridge
should also guarantee higher risk? in Australia. Special serviceability limit state specifi-
The main issue when deciding upon a specific ser- cations and concrete specifications were developed for
vice life is to clarify the event and feature, which will the Confederation Bridge to secure a minimum service
identify the end of the service life (Rostam 2005). life of 100 years. Concrete specifications were also
For typical structures, national codes and regulations developed to ensure a service life of 100 years for the
define implicitly service life requirements through Great Belt East Bridge between Sweden and Denmark.
the standards and codes. Complying strictly with the This paper identifies three major topics to be dealt
performance requirements stated in codes and stan- in order to achieve the intended service life of the
dards will only provide the minimum quality and bridge structure. These are material durability, traffic
performance being acceptable to society. For such live loads and extreme events. For each of these topics,
structures additional requirements would be required current practice is reviewed and corresponding limita-
if truly long-term performance and service life of the tions for the lifetime design are derived in the case of
structures are needed. super-long-span bridges.
28
Table 1. Design life in national codes. spalling of concrete. The end of the service life of
the structure occurs when the accumulated damage
Code Definition of design life in the bridge materials exceeds the tolerance limit.
However, the service life is typically extended by per-
ISO and Design working life (DWL): Assumed period for forming periodic repairs to restore the serviceability
Eurocode which a structure is to be used for its intended
of the structure.
purpose with anticipated maintenance but
without major repair being necessary. Accordingly, the durability design process for
achieving extended lifetime requires the analysis of
DWL Indicative the environmental conditions in which the structure
category DWL (years) Examples will be exposed, the adequate selection of the range of
1 10 Temporary structures materials involved and the understanding and identifi-
2 10 to 30 Replaceable structural cation of their rate of deterioration. The probabilistic
parts approach adopted by the DuraCrete method can be
3 15 to 25 Agricultural and similar cited as the best example achieving such durability
structures
design process for concrete structures.
4 50 Building structures and
other common structures
not listed elsewhere in 3.1 Durability design, current practice
this table Long-span bridges are expected to exhibit longer ser-
5 120 Monumental building
vice life than common bridges, but are exposed to more
structures, highway and
railway bridges, and severe conditions such as chloride ingress or wind load
other civil engineering due to their location in marine environment. There-
structures fore improved and specified design methods have been
adopted to achieve the expected design life.
BS 5400 Design life: Period of time on which the During the last several years, models have been
structure behaves safely without need of repair
developed to predict the service life of concrete bridges
and within acceptable probability (service life
design of 120 years since 1988) exposed to chlorides. Several service life prediction
models assume diffusion to be the most dominant
AASHTO Design life: Period of time on which the mode of transport for chloride ions. The time taken
statistical derivation of transient loads is based by chlorides to reach reinforcing steel and accumulate
(75 years for these Specifications) to a level exceeding the corrosion threshold is known
Service life: The period of time that the bridge
as time to initiation of corrosion (TIC). TIC depends
is expected to be in operation
on many factors like the diffusivity of concrete, con-
crete cover, temperature, and the degree of exposure.
The propagation time, from initiation of corrosion to
3 MATERIAL DURABILITY intolerable accumulation of damage, also depends on
many factors including environmental conditions and
As seen above, the current practice in assessing the corrosion protection strategies.
design lifetime of bridge structures relies essentially Therefore, design codes provide qualitative method
on the lifespan of the involved materials. to achieve the expected design life such as con-
Concrete and steel have been and are still the most crete quality, type of cement, unit cement content,
popular construction materials. Sarja (2010) reported water-cement ratio, concrete cover depth and cur-
several features identifying the end of the service life ing methods. However all these recommendations are
of bridge structures as follows. Degradation is the deem-to-satisfy requirements, which have no ratio-
dominant reason for maintenance and minor repair nal relationship to the service life and the service life
of concrete structures and represents 26% (steel) and as such is not defined in an operational manner.
27% (concrete) of the causes for bridge demolition; The following present several current practice
obsolescence load is responsible of 74% of the causes examples of measures taken to protect materials from
for bridge demolition. environmental ingress.
Environmental conditions that cause degradation
include carbonation, sulfate attack, alkali-silica reac- 3.1.1 Corrosion protections for reinforced concrete
tion, freeze-thaw cycles, and ingress of chlorides structures
and other harmful chemicals. Adverse environmen- Durability life of concrete constitutes the most essen-
tal conditions, if not properly addressed, typically tial factor in achieving the design life of long-span
cause chemicals to invade the concretes pore structure bridges in many cases. Deterioration of concrete is
and initiate physical and chemical reactions causing usually affected by salt damage, neutralization, free-
expansive by-products. The service life of the bridge thaw, alkali-aggregate reaction and sulfate attack.
structure in terms of materials depends on the deteri- In particular, main damage to RC structure is the
oration of concrete, corrosion of steel and interaction corrosion of the reinforcements caused by chloride
between these reactions. The most damaging conse- attack and carbonation process which interacts with
quence of these reactions is depassivation and eventual deterioration of concrete. These degradation mecha-
corrosion of reinforcing steel causing cracking and nisms are currently dealt with by two methods.
29
Figure 4. View of the 3-pylon cable-stayed bridge con-
struction site of Busan-Geoje Fixed Link.
The first method controls the material character-

istics through the concrete cover depth, strength of Figure 5. Required cover depth to achieve target design life
concrete, mix ratio and adoption of corrosion inhibitor. for HTC (KICT 2009).
Increased concrete strength with lowered W/C ratio
delays diffusion of the chloride ions, while increased
cover depth slows down penetration of chloride ions the ship collision protection. The CSR also requires
and subsequently lessens corrosion of reinforcement. additionally to execute maintenance every 4 years and
In case of the Busan-Geoje Fixed Link, an 8.2 km replacement every 60 years for the cathodic protection
motorway between Busan and the Geoje Island (Fig- systems itself (Incheon Bridge Corporation 2004).
ure 4), it is expected to last for 100 years on the basis Increase of concrete durability itself such as high
of durability life of concrete and reinforcements. Of performance concrete (HPC) is another approach to
this 8.2 km, four-lane fixed link, there will be 4 km of achieve design life by reducing the permeability. HPC
immersed tunnel and two 2 km cable-stayed bridges. with W/C ratio between 0.30 and 0.40 are usually more
Based on the DuraCrete method, speed of chloride pen- durable than ordinary concrete, not only because they
etration and mix ratio were experimentally identified are less porous, but also because their capillary and
and the cover depths of pylon, caisson foundation and pore networks are somewhat disconnected due to the
unit of immersed tunnel were accordingly determined development of self-desiccation (Aitcin 2003). In the
to be in a range of 78 cm to achieve sufficient dura- case of Millau Viaduct, researches into concrete mix
bility from chloride ion attack to reinforcement. It was formulation and greater control over its behavioral
applied in a similar way in the Incheon Bridge where mechanisms revealed that the lifetime of the bridge
blast furnace slag was used to increase water den- can be guaranteed over 100 years in terms of mate-
sity and the corresponding cover depths were required rial durability (Lafarge 2007). Increase of material
to be 13 cm for the splash zone and 6 cm for the durability of reinforcement can be a solution as well.
superstructure. Epoxy-coated reinforcements have been studied and
The second method is to protect corrosion through applied in many cases. Stainless steel reinforcements
cathodic protection (CP) and prevention. CP of steel have been also used for durability of bridge. Applica-
reinforcement in concrete aerial structures is obtained tion of stainless steel reinforcements to the outer most
by applying a direct current (D.C.) through the con- vertical layer of reinforcing bars and links can be found
crete from an anode system usually laid on the concrete in the application to the base pylon of Stonecutters
surface. The anode system is connected with the posi- Bridge (Mahmoud 2006).
tive terminal of a low voltage source and the negative It is interesting here to introduce a recently
terminal is connected with reinforcement acting as completed R&D project, called Bridge 200, per-
cathode. This method was later extended to the protec- formed by Korea Institute of Construction Technol-
tion of bridge slabs and piles, marine constructions, ogy (KICT). It intends to secure bridge technologies
industrial plants, garages and buildings exposed to extending the lifespan of concrete bridge to 200 years
chloride-induced corrosion (Pedeferri 1995). These and to develop ultra high-strength concrete exhibit-
methods are usually applied individually or in coming strength 5 times larger than current concretes
bination to the pylons or piers of long-span bridges through 5 key technologies related to the durability
exposed to severe salt attack. In real implementa- of concrete, development of high-durable structures,
tion, this method was adopted for Geogeum Bridge and reinforcement techniques of existing bridges.
in Korea, which is currently under construction and Ultra-high performance (HTC) featured by a water
scheduled to be completed late in 2011. The bridge is to binder ratio of 20% was developed and exhibited
a cable-stayed bridge with a central span of 1.1 km and increased compressive strength of 180 MPa, which is
its durability is expected to ensure a design life of 100 larger than that of existing high performance con-
years. Cathodic protection method was also included cretes. Chloride ingress tests showed that HTC secures
in design of Incheon Bridge project. Steel piles in longer lifespan than other concretes with identical
ship collision protections were required to be protected cover thickness. Tests on HTC also revealed that a
from corrosion by CP method in CSR (Concession- cover thickness of about 27 mm is sufficient to secure
aire Supplementary Requirements). The application a lifetime of 200 years against the penetration of chlo-
of this method guarantees a 60-year design life for rides (Figure 5). Such properties are achieved owing
30
to the low W/C ratio of the material as well as the
homogeneity of the cement matrix.
This project having been successfully completed in
2006, Super Bridge 200 has been launched in 2007 as
a succeeding project involved in the top-brand projects
of the Ministry of Sciences and Technology. Super
Bridge 200 targets the development of bridge sus-
tainable for 200 years by exploiting the so-developed
ultra-high strength concrete to cable-stayed bridge
structures. The expected outcomes are reduction of
the maintenance costs by 20% and extension of the
lifespan of the bridges by 200%.
Figure 6. Key technologies of HIPER-CONMAT.
3.1.2 Corrosion protections for steel bridges

The previous sections dealt mainly with the corrosion amount of alloying elements such as Cu, Cr, Ni pro-
protection of concrete bridges. However, long-span tect steel from excessive rust which result in the
bridges and super-long-span bridges are demanding increase of durability. It enables the developed steels
lightweight materials, which mean that these bridges to acquire weathering indices over 5.8. Those weath-
will make use of steel rather than concrete. ering steels also feature high weldability, which mean
External corrosion protection is a prerequisite for weld preheat is not required. Hybrid slab systems com-
sea side located steel bridges. They are generally bining fiber-reinforced polymer (FRP) are also have
required to be galvanized or coated with metal spray been developed to improve durability of structure by
and paint. Heavy-duty anti-corrosive painting system lightening the weight of slab and prolongation of main-
is widely applied for steel bridges. Electric reactions tenance period from previous 1015 years to over 20
between steel of bridge and ionized metal such as zinc, years.
aluminum are used to protect corrosion. The Golden
Gate Bridge adopted modern inorganic zinc-rich paint 3.1.3 Cable protection
system for the first time in large bridge structures The cables of long-span bridges as main structural
(Kline 2009). Stonecutters Bridge used a high-build members are required for long term endurance. The
epoxy based paint system on zinc rich primer has main cables of suspension bridges cannot be replaced
been specified (Mahmoud 2006). High-build inorganic usually, which means that their durability must be
zinc-rich and epoxy resin painting system was applied imperatively verified and secured.
to Inoshima Bridge and Akashi Kaikyo Bridge (Chen Stay cables of suspension bridge are protected from
et al. 1999). corrosion conventionally by the following method.
Internal corrosion protections are also important. Each cable consists of high-strength steel wires coated
For Golden Gate Bridge, CP for steel bridge was with zinc. These galvanized wires are covered with
used. For Incheon Bridge, dehumidification system an anti-corrosion paste, wrapped with annealed galva-
was applied. This system aims to control the inter- nized wires and painted. This system was first used on
nal environment to provide a relative humidity of less the Brooklyn Bridge about 100 years ago (Stahl et al.
than 60%. This will effectively prevent corrosion of 1996). However zinc-coating is not sufficient in severe
the internal surface of the steel boxes and eliminate the environmental condition near sea water that additional
need for the sophisticated corrosion protection scheme corrosion protection is essential.
used on the outside. Youngjong Bridge(2000), the first bridge with
Similarly to the reinforcement, researches and three-dimensionally profiled suspension cables and
applications on the increase of steel durability itself self-anchoring, used the advanced dehumidification
have been implemented. HPS50W and HPS70W process for cable protection. Dry air is produced by
developed in USA show developed performance not a dehumidification device containing silica gel. The
only in strength but also in weathering and fatigue air is carried through piping, enters the cable via an
characteristics. These types of steel were verified inlet cable band passes through the cable, and exits via
with experiment and included in the AASHTO LRFD an outlet cable band. Dry air is continuously injected at
Bridge Design Specification. a rate of 0.25 m3 /min with the pressure of 300 mmAq.
Similarly in Korea, HIPER-CONMAT (High Per- The relative humidity, designed to remain below 40%,
formance Construction Material Research Center), a has been continuously monitored on the bridge since
new-born research center established at the Research the system entered service in 2000.
Institute of Industrial Science and Technology (RIST) Among the projects implemented by the Super-
in the form of a research instituteindustry coop- Long Span Bridge R&BD Center that will be intro-
erative program extending for a period of 5 years, duced in Chapter 6, the Core Research Project
has developed HSB500W, HSB600W and HSB 800W 2 intends to increase durability of cable material
with increased weathering and fatigue performances in itself and cable system for super-long-span bridges.
the HSB (High-performance Steel for Bridge) project. First, zinc-aluminum alloy coating method for high-
Stabilized rust layer composed with Fe and a small strength cable has been developed to achieve highly
31
corrosion-resistance coating system. Zinc-aluminum
alloy coated cables are expected to have enhanced
corrosion resistance more than 2 times compared to
former zinc coated cables. Techniques and design stan-
dards for high-durability ceramic/heavy duty painting
method are also developed to ease maintenance even
in harsh conditions like near the sea water area. The
painting systems are purposed to have B10 life longer
than 20 years with 80% reliability. Cable dehumidifi-
cation system expected to reach the global standards
is also going to be developed through this project.
Researches and experiments on durability of cables,
anchorages and cable systems using FRP to improve
corrosion resistance and fatigue performance are also
under progress. These researches may contribute not Figure 7. Design life of concrete structures: a two-phase
only to the steel durability but also to the guarantee of modeling of deterioration (Tuuttis Model 1982).
the 200-year design life.
reach a tremendous level regard to the importance of

3.2 Limitations of the current practice in the bridge as unique access and the time required for
super-long-span bridges its construction. From such a point of view, a service
life exceeding 200 years is obviously necessary.
The afore-mentioned durability design methods may Due to these evident characteristics of super-long-
have some limitations for direct application to super- span bridges, several limitations of the current practice
long-span bridges. One reason is that super-long-span can be highlighted as follows.
bridges require service life significantly longer than
common bridge and another reason is that improved
solutions should be applied for the main elements that 3.2.1 Lack of quantitative indicator and
are not replaceable. assessment methods
Since super-long-span bridges are usually more Common design codes suggest implicitly 50-year
important than common bridges in economic and (Eurocode, ISO) and 75-year (AASHTO) of design
social aspects, higher risk should be guaranteed all life. But these codes fail to specify explicitly a target
along a longer design life. However most of the pre- service life and show no numerical basis (Gehlen et
vious design methods do not explicitly mention the al. 1999; Sarja et al. 1996). The limited consideration
service life. of extended service life may partially be the result of
Since super-long-span bridges are expected to have the lack of clear specifications on this topic. Differ-
higher demands in steel not only for the cables but also ent service life levels might be appropriate in future
for the girder, deeper attention should be dedicated on specifications depending on factors like initial cost,
the lifetime design. However, the 2006 National Bridge importance, and average daily traffic.
Inventory data in USA recorded no less than 73,798 Researches on the quantitative prediction of design
bridges as structurally deficient, that is approximately life for RC bridges have been performed and applied.
12% of the bridge stock. These data revealed that the The ACI Life-365 program (ACI Committee 365
structural deficiency rate is substantially greater for 2000) and DuraCrete Project (DuraCrete Report 2000)
steel bridges than concrete bridges with a value for are representative examples. These approaches use
steel bridges 5.2 times that of prestressed concrete numerical models of the deterioration mechanism of
bridges and 2.7 times that of conventional reinforced RC member to clarify chloride ion properties near rein-
concrete bridges. forcement steel. If the calculated chloride concentra-
On the other hand, bridges are maintained through- tion exceeds the limit concentration for steel corrosion
out a series of repairs and replacements of the members it is assumed that the failure has occurred. And the
and the end of the lifetime occurs when maintenance corresponding time is considered as design life.
cost exceeds newly construction cost. For instance, the The DuraCrete approach has been developed dur-
failure of the I-35W Bridge in Minneapolis on August ing a European research project (19961999) and is
2007 may be cited as an example of the user costs internationally the only available probability-based
associated with the replacement or repair of a major service life approach. This approach computes proper
bridge located in urban area. The Minnesota Depart- design details such as minimum cover depth and max-
ment of Transportation established a user cost related imum chloride diffusion coefficient guaranteeing the
to this failure of $400,000 per day. This cost reached service life of the concrete structure. Chloride ion
$165,200,000 for the 413-day period that the bridge ingress model based on Ficks 2nd law is adopted
was out of service. This cost remained relatively low to determine limit state function. Selected design
because of the bridge was replaced within a very short details confirm target reliability (i.e. target failure
period of 11 months. In the case of super-long-span probability) for expected design life. Specified envi-
bridges, the construction costs to be burdened will ronmental categories are also considered to represent
32
Figure 8. Example of target service life vs. failure
probability.
Figure 9. S-N fatigue curve.
severe environment conditions. Coefficients of con-
crete properties are acquired by experiment and not be established and verified to assess the design life.
by theoretical basis to enhance the accuracy. The long history of concrete allowed the development
This approach is used for Busan-Geoje Fixed Link. of model with a relatively high degree of reliability.
The environmental conditions of the bridges and But steel members need more verifications and exper-
immersed tunnel are classified into 3 zones that are iments. The establishment of a reliable deterioration
splash zone, submerged zone, and atmospheric zone, model for steel would enable to suggest quantitative
and limit states are defined as the time when chloride design life using reliability-based design.
ions reach the reinforcement. Accordingly, the cover Protection methods such as CP or painting sys-
depths and chloride diffusion coefficients are deter- tem need also to be quantified. Those methods are not
mined so that the probability of those limit states to be expected to last for 200 years without additional main-
violated (i.e. failure event) remain less than 10% for tenance. Periodic monitoring system and maintenance
100-year life (Daewoo E&C 2005). are essential. Currently, the expected life of these pro-
However, these methods present some lack for tections is not suggested quantitatively but implicitly.
being applied to super-long-span bridges such as For reasonable maintenance planning, durability and
model imperfections of deterioration model due to reliability of each protection system should be verified.
complexity of deterioration mechanisms. According
to Luping (2007), even DuraCrete method which uses 3.2.2 Fatigue load problem
simplified deterioration model for RC can overesti- The Korea Bridge Design Code (KBDC 2008) sug-
mate the actual life of structure. This imperfection gests 2 million cycles for fatigue life of reinforcement
error can be neglected for bridges with relatively short assuming a 50-year life. If the fatigue life is extended
design life but may become significant for prolonged to 200 years, the number of cycles to be considered will
design life and increased importance of super-long- increase which leads to decrease of allowable fatigue
span bridges. This error can be lessened by using stress as shown in the following S-N fatigue curve.
improved deterioration models which are more sophis- This result can be verified using the following equa-
ticated even though the analyzed lifetimes from those tions. For load-induced fatigue considerations, each
models also show some differences with actual exper- detail shall satisfy:
imental results. There are some researches to provide
a correct model based on the actual physical or elec-
trochemical processes considering distance from the
surface, time, the interactions between chloride and
where = load factor for fatigue limit state; (f ) =
cement hydrate, the influence of moisture transport in
live load stress range due to the passage of fatigue load;
parallel with diffusion within the concrete, etc. (Nils-
(F)n = nominal fatigue resistance (MPa).
son 2009). Although these types of model would be
Nominal fatigue resistance:
better than those based on simple Ficks 2nd Law, these
methods usually require finite lamina time stepped
calculation to solve the sophisticated mathematical
equations that are hard to apply field application
directly. Accordingly, research should be implemented
to provide more reliable and practical models. where N = (365)(t)n(ADTT )SL ; t = design life (year);
On the other hand, the development of assessment A = detail category constant taken from Table 2
methods usually focused on the deterioration of con- (MPa3 ); n = number of stress range cycles per
crete and reinforcement steel and, not on steel element. truck passage; (ADTT )SL = single-lane ADTT (Aver-
Even if RC member is important in pylon or pier of age Daily Truck Traffic); and (F)TH = constant-
super-long-span bridges, the importance of steel mem- amplitude fatigue threshold in Table 2 (MPa).
ber such as the non-replaceable girder is more crucial In Equation 2, the nominal fatigue resistance
due to the increased span length. It means that deterio- decreases as the total number of fatigue load passage
ration model for steel members according to time must N increases. Since N is proportional to design life t,
33
Table 2. Detail category constant, A, and constant-
amplitude fatigue threshold (F)TH .
Detail category A 1011 (MPa) Threshold (MPa)
A 82.0 165.0
B 39.3 110.0
B 20.0 82.7
C 14.4 69.0
C 14.4 82.7
D 7.21 48.3
E 3.61 31.0
E 1.28
the fatigue resistance would decrease if the design life

changes from 100 years to 200 years.
In general case, fatigue load is not the governing
Figure 10. The highway live load model in Canada
factor. But as allowable fatigue stress decreases due to (OHBDC 1983).
increased design life, fatigue may become a governing
factor and should thus be verified.
In addition, the modified design fatigue truck HBS and HBSW that will likely be adopted in super-
according to extended fatigue life affects the reliability long-span bridges should be imperatively verified to
index. The limit state function based on Miners rule achieve a 200-year design life.
for fatigue failure is (Nyman 1985):
4 VEHICULAR LIVE LOAD
Vehicular live load model used in the design of super-

where Df = damage to cause failure; V = ADTT; long-span bridges has to be determined according
t = design life (day); ND = design number of cycles; to the truck weight and traffic characteristics in the
S = stress range ratio; M = moment ratio; G = nation or region where the bridges are located. Because
distribution factor ratio; I = impact factor ratio; this vehicular live load is highly uncertain factor for
H = headway ratio; and L0 = loadmeter value. These the lifetime of the bridge, a probabilistic approach is
random variables (Df , V , S, M , G, I , H , L0 ) are known to be the most reasonable choice to determine
determined as shown below. the live load model (Nowak 1993). In the AASHTO
LRFD design specification (AASHTO 2004), the
HL-93 load model proposed by Nowak (1993) is used.
The HL-93 load model is determined based on the
maximum load effect that bridges endure for the life-
time of 75 years. The maximum load effect is estimated
by truck surveys, weigh-in-motion measurements, etc.
It is assumed that extreme vehicular load event is
related to the distribution of only heavy trucks.Accord-
where D = nominal design value; k = k-th bridge site; ingly, only probability distribution function of upper
Srt = true stress range; i = weight of i-th truck pas- 20% total truck weight are assumed as normal and
sage; D = gross weight of design fatigue truck; and the 75-year maximum truck weight is estimated by
n = total number of trucks. linear regression on the probability paper. In addi-
In Equation 3, the limit state function for the fatigue tion, OHBDC (Ontario Highway Bridge Design Code,
failure depends on the fatigue life t. As the design 1983) specifies the highway live load as the larger of
fatigue life increases, the function value g decreases, either a truck load or a lane load as shown in Fig-
which results in the decrease of the reliability index. ure 10. OHBDC truck model is based on commercial
Moreover, as the design fatigue life increases, the vehicle weight surveys conducted in Ontario from
COVs of the random variables M and L0 may increase 1967 to 1975. OHBDC truck weights 700 KN which
due to further uncertainty and consequently the reli- is significantly more than AASHTO standard truck.
ability index may decrease additionally. These results Meanwhile, DB-24 and DL-24 load models used in
show that the fatigue truck loads may be the governing KBDC (2008) were developed on the basis of HS20-
factor with a 200-year extended fatigue life. 44 load model of AASHTO in the 1970s. However,
Together with theoretical investigation, the fatigue because of the growth of the commercial traffic and the
stress of steel needs also to be experimentally verified. increase of truck weight, new vehicular live load model
Especially, the fatigue life of improved steels such as should be developed. Therefore, KBRC (Korea Bridge
34
Figure 11. The new live load model in Korea (Hwang 2009).
Figure 13. Route of the seaside roads and long-span bridges

to be constructed in Korea.
Figure 12. Sites installed with BWIM in Korea.
Design & Engineering Research Center) proposed a

new vehicular live load model (TL truck load) based
on the traffic data of Korea. The new model is shown Figure 14. Moment and shear bias factor by span length
in Figure 11 (Hwang 2009). (HL-93).
The procedure adopted for the development of the
TL truck load model is similar to that of Nowaks load in 200 years or not. In view of the current trend,
study (Nowak 1993). In order to establish statisti- trucks are becoming heavier but one cannot affirm that
cal database of traffic load, BWIM (Bridge Weigh-In this trend will continue. The necessity for CO2 reduc-
Motion) system is installed at 8 sites in Korea (Fig- tion could lead to the shift from road transportation to
ure 12). From the obtained data, the representative rail transportation. The adoption of innovative mate-
types of vehicles are chosen and the probabilistic rials in the vehicle industry may also lead to lighter
distribution function of total vehicle weight is deter- trucks. Therefore, the future transportation system
mined. The maximum load effect is estimated and should be considered with the estimation of current
TL truck load model is proposed. Differently from vehicle load.
Nowaks study, the Extreme Type I distribution was The second issue lies in the data collection area. The
identified to be more proper than the normal distribu- data collection areas (Figure 12) selected in order to
tion by analyzing the probability distribution of upper make TL load in inland region, but the regions where
data (Hwang 2009). Thereafter, 100-year maximum long-span bridges will be constructed are likely coastal
truck weights are estimated by linear regression on areas (Figure 13).
the probability paper. It is anticipated that there will be substantial dif-
In the current stage, the two vehicular load modes ference between the information collected in the cur-
are compared. Then, we have to consider if the two rent inland region and the vehicle characteristics of
load models can be applied to super-long-span bridges naval region, so numerous studies have been con-
with a 200-year lifetime. There seems to be three major ducted in Korea to establish the statistical data for
issues about this problem. super-long-span bridges.
First, the HL-93 load model is made by assuming The third issue is that the current HL-93 load and the
that lifetime of bridge is 75 years and the TL truck TL truck load model are made only for bridges shorter
load model of Korea is made by assuming that lifetime than 100 m. This implies that verification is required
of bridge is 100 years. If we design the super-long- if constant bias factor (ratio of the mean maximum
span bridge which has lifetime of 200 years, 200-year moment/shear and HL-93 maximum moment/shear)
maximum truck weight should be estimated by linear (Figure 14) is guaranteed for super-long-span bridges.
regression on the probability paper. In the current Korean Design Code for Cable
However, it is still questionable if this simple extra- Supported Steel Bridge, legal enforcement spec-
polation will be able to predict the actual maximum ifies to apply reduced factor with span length
35
Korea, therefore a rational live load model should be
developed based upon the collected data in the future.
5 EXTREME EVENTS
Extreme events are also important features of concern

in the design of bridges. Extreme event could be nat-
ural hazard like earthquake, landslide, typhoon, flood
and fire, or manmade hazards such as bombing, col-
lision and terror attack. The consequences of extreme
events on the integrity of the bridge can be tremendous
as we could see during the Kobe earthquake in 1995.
Predicting the effects, loads and eventual dam-
Figure 15. Bird view of Seohae Bridge in Korea. ages accompanying the occurrence of extreme event
is a major task when designing bridges for extreme
events. Therefore, the estimation of the likelihood of
the occurrence of a given duration and intensity of
the event through its return period is crucial. Risk
assessment involves the prediction of the chances of a
specific set of events to occur together with their conse-
quences. The exceedance probability curve constitutes
a helpful tool for decision-making.
However, long-span bridges are by nature and loca-
tion more vulnerable to such extreme events. Ensuring
a service life of 200 years means also that higher risk
should be guaranteed. Accordingly, the current design
method which adopts a probabilistic approach only in
hazard analysis to determine occurrence rates of the
extreme events and uses a deterministic procedure for
Figure 16. Annual variation of vertical deflection at the
center of the main span of Seohae Bridge (Park et al. 2003). the structural analysis should shift onto a fully proba-
bilistic approach, i.e. consequence-based design may
be needed.
(0.57 + 300/(500 + L)) to uniform load (Korean
5.1 Features of extreme events in super-long-span
Design Code for Cable Steel Bridge 2006). ASCE
bridges
Report (ASCE 1981) states that there are plenty of
margins in the live load. Such margins were also been The consequences of accidents in super-long-span
verified in Seohae Bridge (Figure 15) and Youngjong bridges would likely be much larger in terms of socio-
Bridge in Korea (Koh et al. 2008). economic losses than normal bridges. Besides, the
Specifically in the case of Seohae Bridge, the struc- extended lifetime required for such bridges increases
tural behavior of the cable-stayed bridge was observed the probability of the occurrence of accidents. There-
and analyzed during the first 2 years following its com- fore super-long span bridge should be designed to
pletion. Results showed that the annual variation of satisfy for much higher risk level. In general a quan-
the vertical deflection in the stiffening girder satis- titative index of risk is used in extreme event design.
fies the allowable design limit and that the deflection The risk R is defined as follows:
due to live load represents only 25% of the design
limit. The stress range in the stiffening girder due to
live loads showed good correlation with the volume of
traffic monitored during 2 years. Stress margin appears
to remain considerable since measured stresses repre- where Pi = probability of i-th accident within a given
sent only 5 to 12% of the design stress. Accordingly, time (i = 1, 2, . . . , n); Ci = consequence of i-th acci-
the actual highway bridge design specifications in dent in the given time; and n = number of possible
Korea are producing excessively conservative struc- accidents. Super-long-span bridges have higher risk
tures (Park et al. 2003). than that of common bridges because of their extended
For these purposes, the Honshu-Shikoku Bridge lifespan, the higher occurrence of possible accidents
Authority suggests cutting down the designed live load (n) to happen and the larger consequences such as cost
according to span length for designing super-long- of repair or failure (Ci ).
span bridges, and the Korean Design Code for Cable In consideration of the social and economical
Steel Bridge also applies the method reducing the uni- importance of super-long-span bridges, the societal
form lane load when span length exceeds 200 m. But acceptable risk should be regulated strictly. The soci-
this cut down does not rely on measured truck data in etal acceptable risk is based on FN (Fatal Number)
36
Table 3. Return period and risk rate of earthquake.
10% of exceedance within a given lifetime
Design life (years) Return period (years) Risk rate
5 48.0 0.40
10 95.4 0.57
20 190.3 0.73
50 475.1 1.00
100 949.6 1.40
200 1898.7 1.89
Figure 17. Example of super-long span bridges FN-curve.
curves. FN-curves are a graphical presentation of

information about the frequency of fatal accidents in
a system and the distribution of the numbers of fatal-
ities in such accidents. They plot the frequency F(N)
of accidents with N or more fatalities, where N ranges
upward from 1 to the maximum possible number of
fatalities in the system. Because the values of both F Figure 18. Seismically over-designed piers and correspond-
and N sometimes range across several orders of mag- ing arrangement of reinforcing bars.
nitude, FN-graphs are usually drawn with logarithmic
scales. FN-curves themselves are simply a means of Table 4. Basic design wind speed on design life increase.
presenting descriptive information about fatal accident
frequencies and fatality distributions. In this respect 0.6 of Non-exceedance probability
they are similar to histograms, and indeed present the
same information as histograms in a different way. Design Return Basic design
These curves show three regions bordered by two life (years) period (years) wind speed (m/s)
curves in logarithmic scale. Values of F for high values
of N are often of particular political interest, because 5 10.3 25.9
these are the frequencies of high-fatality accidents. 10 20.1 28.6
25 49.4 31.9
The curves are acceptable risk higher where risk could 50 98.4 34.5
be accepted, lower where risk could not be accepted 100 196.3 37.0
and, model where ALARP (As Low As Reasonably 200 392.0 39.5
Practicable) where all possible measure should be
undertaken to reduce the risk. These tolerable crite-
ria vary case by case. Figure 17 illustrates an example where identical risk rate is accepted, the earthquake
of FN-curve where the criteria of risk level are adjusted load is determined by the product of the seismic site
for super-long-span bridges which are assumed to have coefficient and risk rate.
10 times more risk consequence than common bridges. Table 3 presents the return period of the earthquake
with respect to the design life. It can be seen that earth-
quake load for a 200-year service life is 1.35 times
5.2 Lifetime design against extreme event, current
than that of 100 years. Simple application of the load
practice
would imply that stiffer foundations and piers should
Extreme loads are different from ordinary loads be designed to resist increased earthquake loads, which
and bear uncertainties, which require a probabilistic consequently will lead to increase of the construc-
approach. Extreme loads are predicted probabilisti- tion costs and loss of constructability due to the larger
cally based on observed data. Especially, occurrence amount of reinforcing bars to be arranged (Figure 18).
probability of natural disasters such as earthquake and
typhoon is expressed as return periods, and some spe- 5.2.2 Wind loads for lifetime design
cifications (e.g. AASHTO LRFD) state return periods In the case of wind loads, the aerodynamic stability
as much as 50 to 75 years. The question is how to man- should be secured through the stability of the vibra-
age these return periods for super-long-span bridges tions induced by the dynamic loads. Of equivalent
designed for a service life of 200 years. importance, need is also to obtain the material prop-
erty which resists to static wind loads. At this time,
5.2.1 Earthquake loads for lifetime design static wind loads increase with the rate of lifetime as
For earthquake loads, the load is determined by means shown in Table 4.
of the average return period exceeding a probabilistic Because FD (Static wind loads per unit length) is
threshold value during the specified period. In the case linear to the square of the design wind speed, design
37
Figure 19. Rendering of Gwangyang Bridge with main span
of 1545 m in Korea.
wind loads of 100 and 200 years are determined as

follows.
Accordingly, a 200-year lifetime design implies that

the design wind load should be increased by about Figure 20. Gradually optimized cross sectional shapes of
14% compared to that of 100 years. Since design wind Gwangyang bridge in Korea.
load generally governs the design of the superstruc-
ture of long span bridges, higher design wind load
requires higher resistance strength of the girder. More-
over, higher design wind speed needs also larger flutter
onset velocity, which is related to the vibrational char-
acteristics and shape of the cross-section. Therefore,
increased wind speed implies strengthening of the
girder and improvement of the cross-section.
Gwangyang Bridge provides a good example of the
optimization of the cross-section of the girder for the
satisfaction of the flutter onset velocity. Gwangyang
Bridge will cross Gwangyang Bay and will connect
the cities of Gwangyang and Yeosu passing through
Myodo Island in the southern coast of Korea. The Figure 21. Extrapolation of an extra-large vessel growth.
completion of the bridge is scheduled to happen in
2012 just before Yeosu Expo 2012. The Gwangyang-
Yeosu area being located in the southern seashore is
regularly swept by typhoons in summer. During the
Typhoon Maemi in 2003, a maximum instantaneous
wind velocity of 49.2 m/s was recorded at the Yeosu
Weather Station far away by about 18 km from the
bridge site. Considering that the bridge superstruc-
ture will rise at 81 m above the sea level, high wind is
likely to occur at the deck. The critical wind velocity of
Gwangyang Bridge is determined 81.6 m/s. Therefore,
the aerodynamic stability constitutes a main concern
in the design of the bridge. The flutter onset velocity Figure 22. Harbor facility development plan in Korea.
has improved from 67 m/s of initial cross section to
120 m/s after optimization of cross-sectional shapes of supersized cargo ship (Figure 21) and the harbor
(Kwon 2008). facility development plan (Figure 22) constitute very
important data which can influence the design values.
5.2.3 Ship collision prevention for lifetime design Various risk assessment models have been devel-
Risk of vessel-bridge collision is decided according to oped about collision between vessels and bridges.
the magnitude and quantity of the vessel passage on Most of these models rely on formula (7), which calcu-
the bridge site. The magnitude of design vessel and the lates the annual frequency of bridge elementscollapse
passing quantity can be forecasted using linear extrap- (AF). The inverse of AF (1/AF) is equal to the return
olation of statistical data. For example, the emergence period (in years). For a specific member, the sum of
38
all the AFs computed for each class of vessels cor- limited historical ship collision data of US waterways.
responds to the annual frequency of collapse of this Geometric condition, effects of current and traffic vol-
specific structural element (PIANIC 2001). ume are considered in the calculation of PA. However,
the human error factor, which represents the most
important factor in ship aberrancy, is not reflected
sufficiently in the current assessment methodology
where AF = annual frequency of bridge element col- of ship aberrancy probability. Geometric probability
lapse due to vessel collision; N = annual number of (PG) which is concerned with the model of the loca-
vessels classified by type, size, and loading conditions tion of an aberrant vessel and the collision velocity also
which can strike the bridge element: PA = probability need to be improved. Based on historical bridge col-
of vessel aberrancy; PG = geometric probability of a lision data of USA, a normal distribution is assumed
collision between an aberrant vessel and a bridge pier to model the location of an aberrant vessel which can-
or span; PC = probability of bridge collapse due to a not represent actual aberrant ships path with accuracy.
collision with an aberrant vessel. The same observation can be done for the ship collision
According to this formula, the annual frequency of velocity. Calculated ship collision risk of super-long-
bridge collapse depends on the predictions of the future span bridges may differ from actual value because
annual number of vessels (N ) and vessel size. More- of the lack of available statistics and may result in
over, if design is performed assuming a higher level erroneous designs of the bridges.
of risk for extended lifespan of the bridge, a smaller Another limitation of the current practice concern-
target AF should be adopted in the design. Since the ing ship collision is the formula adopted to estimate
values of PA and PG cannot be controlled or adjusted the ship collision force. The impact force formula pro-
by the designer, the value of PC must be reduced posed inAASHTO LRFD is based on studies dedicated
to strengthen the resistance capacity of the bridge to the collisions between ships (AASHTO 1991). In
elements to satisfy the current design specifications. case of ship-ship collision, the absorption of kinetic
Supplying additional protection facilities like dol- energy due to large deformations of each ship is the
phins constitutes another way for reducing the collapse main mechanism. However ship-bridge collision cases
probability. The type and size of the ship collision pro- are ship-to-rigid-body collision, which results in the
tection facilities depend on the impact loads of design underestimation of the impact force by the formula of
vessel. AASHTO (Petersen et al. 1993; Svensson 2006).
Both options have naturally direct influence on the The prediction or estimation of extreme events that
construction costs. Therefore, for more economical may occur in the future is very difficult and involves
design against vessel collisions the pre-estimation of a wide range of variables. Even though a probabilis-
future vessel type, size and quantity should be carefully tic approach is adopted, the corresponding prediction
considered. is likely to lose its accuracy as much as the consid-
ered design lifetime lengthens. In current practice,
simple extrapolation will increase extreme load and
5.3 Limitations of the current practice in
make it a governing load in the structural design.
super-long-span bridges
Therefore, change of the design paradigm should
As mentioned earlier, if the design loads are extrapo- give priority to the economy and sustainable design
lated for 200-year lifetime using the current practice, against extreme events. Under extreme loads the con-
the resulting structures are likely to be excessively sequence of damages are more important than other
overdesigned. Especially, the three types of extreme criteria related to the deflection or serviceability. As a
events considered above have been seen to influence result, when extreme events are considered in design,
different parts of bridge. That is, earthquake loads the design paradigm should shift toward risk-based
govern the design of the substructure, wind loads design or consequence-based design to minimize the
determine the design of the superstructure and vessel consequences and damages.
collision influence the design of the piers or require Innovations are needed to transcend the current
additional protections. Besides, if more stringent cri- practice and to devise reasonable approaches in cop-
teria are applied in the design of the structural parts, ing with extreme events. In case of natural disasters
the feasibility of the construction project can be ruined such as earthquakes and typhoons, the probabilities of
due to the loss of economical efficiency brought by extreme events are site-dependent. Careful site investi-
larger sections and stronger structural elements. gation should thus be conducted to assess realistically
Since ship collision assessment requires proba- and reasonably the probability of the uncertainties
bilistic approach because of its rareness, previous related to these events. In the case of Honshu-Shikoku
design specifications such as AASHTO LRFD sug- Bridge Authority, the special windproof design stan-
gest methodologies based on a probabilistic approach. dard forAkashi Kaikyo bridge was established in 1990.
However, the lack of statistical data impedes direct Design specifications should prescribe the application
application of these methodologies to the design of of improved structures such as airfoil shaped sec-
super-long-span bridges for the following reasons. tions and multi-box girders used in Messina Bridge
The AASHTO LRFD suggests base rate of proba- or Gwangyang Bridge so as to cope efficiently with
bility of ship aberrancy (PA) as simple values based on extreme events. Solutions like seismic control devices
39
against strong earthquakes and gales should also be Cho (2009) suggested a probabilistic analysis method-
developed to be applicable for super-long span bridges. ology using 3D ship maneuvering simulation data to
In addition, in case of manmade disasters like improve the accuracy of the results. The simulation
ship collisions, cooperation with other industrial fields data for the given ship and bridge location reflect
should also be accounted. As an example, recent navi- more accurately the effects of various weather/current
gation technology of ship has experienced tremendous conditions and even human factors on ship collision
evolution. GPS and laser range finder can reduce accidents. However, the time and cost required to per-
drastically the occurrence of ship-bridge collision. In form such simulations constitute also an obstacle in
Korea, the volume of vessel has recognized an increase gathering a sufficient number of simulations to estab-
of 800% within two decades. But the annual ship col- lish reliable probabilistic model of aberrant path and
lision frequency has just increased less than two times collision velocity. For that reason, random vector gen-
(KMI 2003). This proves that manmade disasters can eration method from multivariate normal distribution
be reduced relatively easily by technical tools. is applied to the limited simulation data in order to
generate and acquire sufficient and meaningful data.
This methodology makes it possible to calculate ship
6 SUPER LONG SPAN BRIDGE R&BD collision risk more accurately over the long service
PROJECT life. Besides more convenient to use and more reli-
able methodology have been developed using fast-time
The Ministry of Land, Transportation and Maritime simulation with advanced auto-pilot function. The col-
Affairs (MLTM) of Korea has set up the VC-10 lapse mechanism of ship-to-pier according to various
(Value Creator-10) Program for Construction Tech- types of bridges and ships will also be investigated.
nology Innovation. Technology Road Map for each In addition, a ship collision risk assessment program
program was established by experts from academia, including dolphin protections will be developed. An
research institutes and various fields of construction expected outcome of the program will be the optimal
industry. One of the programs is Super Long Span design and placement of dolphin protections.
bridge R&BD project, launched in 2009 under the These results from the researches about extreme
supervision of Korea Expressway Corporation, which events will also be applied to prepare for the design
defines four major programs: (1) development of key specifications for super long span bridges together
technologies for planning and design of long-span with the development of aerodynamic cross-section of
bridges, (2) development of high-performance cable cable-supported bridges. Disaster management system
systems and high-performance materials for long-span will be developed for a series of disaster scenar-
bridges; (3) development of highly efficient and inno- ios. Based on those scenarios, disaster factors will
vative cable erection equipments and methods, and be monitored in real time, and the damages will be
cost effective construction methods for offshore mega analyzed accurately by probabilistic risk assessment
foundations; and (4) development of IT-based disaster technique. The so-gathered disaster information will
mitigation and maintenance technologies. be exploited for bridge users safety as well as for the
Lifetime design is one of the most important themes implementation of immediate actions.
of the project. To that goal, research is actively con- Other important topic of the R&BD project is the
ducted in diversified fields involving materials, live durability improvement of materials and bridge sys-
loads and extreme events as discussed in the previ- tems. HSB800W steel featured not only high strength
ous chapters. The detailed researches of each program and improved fatigue performance but also improved
related to the features discussed above are as follows. weathering performance is under development. Devel-
Live load models will be developed to be suitable for opments of zinc-aluminum alloy coating method,
super-long-span bridges. For that purpose, measured dehumidification system and high-durability paint-
truck data of domestic bridges have been collected and ing method have also been performed to enhance
analyzed. System reliability assessment methodology corrosion resistance of high strength steel cable for
and reliability based load combination method using super-long-span bridges. Increase of cable material
the developed model will also be developed through durability itself is also supposed to be attained by
this project. The final outcome of this topic is the devel- using FRP cable. Developed FRP cable and anchorage
opment of reliability-based design specifications for systems will show improved fatigue performance and
super long span bridges. durability. These researches will contribute to increase
Related to the extreme events, the extreme loads material durability and eventually to secure 200-
caused by earthquakes, typhoons or ship collisions will year design life of super-long-span bridges. Details
be defined more specifically and reasonable design of the assignments can be found in the website
solutions will be suggested. As mentioned in sec- (www.longspanbridge.org).
tion 5.3, the current probabilistic approach adopted
for the assessment of ship collision presents several
limitations due to the lack of statistical data. Accord- 7 CONCLUSIONS
ingly, improved methodology for the assessment of
ship aberrancy probability and geometric probability This paper reviewed current practices for service life
has been studied using ship maneuvering simulation. design and addressed issues that will be faced in the
40
lifetime design for cable-supported super-long-span gratitude for Dr. Y.J. Kim at the Korea Institute of
bridges. Research directions that will help designers Construction Technology and Prof. D.H. Ha at Konkuk
in achieving extended lifetime of 200 years or more University for their contribution.
for the super-long-span bridges have been derived.
The main issue when deciding upon a specific ser-
vice life is to clarify the event and feature, which will REFERENCES
identify the end of the service life. In view of the cur-
AASHTO 1991. Guide Specification and Commentary for
rent practice, service life of 100 to 150 years seems Vessel Collision Design of Highway Bridges. Washington,
to be achievable and has already been implemented in D.C.
recent major bridges owing to special specifications AASHTO 2004.AASHTO LRFD bridge specifications. Wash-
by the owners of the bridge projects. ington, D.C.
The lack of specification provisions prescribing ACI Committee 365. 2000. Service-Life-Prediction State-of-
explicitly the length of the service life is an obvious the-Art Report, ACI 365.1R-00. Farmington Hills.
obstacle to the implementation of extended service life Aitcin, P.C. 2003. The durability characteristics of high
for major bridges. Even though each long-span bridge performance concrete: a review. Cement and Concrete
is unique, the development of relevant specifications Composites 25(45): 409420.
Bartholomew, M. 2007. How long will your bridge last?
providing sufficient margins should be implemented The need to predict service life of bridge components.
for the designers to guarantee the service life of their Western Bridge Engineers Seminar, Boise, Idaho, 2326
structures. September 2007.
This paper addressed three major topics for which Bartholomew, M. 2009. Design for service life, bridge birth
limitations of the current practice are likely to hap- certificate and concrete structures management concepts.
pen for the design of super-long-span bridges in the AASHTO Bridge Sub-Committee Meeting, T-9 Technical
future. These topics are material durability, live load Committee for Bridge Preservation, New Orleans, LA, 6
model and extreme events. Currently, design prac- July 2009.
tice assumes that securing material durability for Chen, W.F. & Duan, L. (ed.) 1999. Bridge engineering
handbook. Florida: CRC press LLC.
200 years will guarantee equivalent or longer life- Cho, H. H. 2009. Probability Analysis of Ship-Bridge Colli-
time of the bridge. However, one cannot assume the sion Using Ship Maneuvering Simulation (Ph. D. Thesis).
reliability of such approach considering the harsher Seoul National University. Korea.
conditions in which super-long span bridges will be Daewoo E&C 2005. Busan-Geoje Fixed Link-Concrete Dura-
exposed. Accordingly, the application of high perfor- bility. Internal Report.
mance materials together with appropriate protection DuraCrete Report 2000. DuraCrete Final Technical Report,
alternatives should be seriously conceived and dedi- Document No. BE95-1347/R17. Gouda: CUR.
cated research should promoted. In the case of load Frangopol, D.M. & Furuta, H. (ed.) 2000. Life-cycle cost
models including live loads and extreme events, the analysis and design of civil infrastructure systems. Proc.
US-Japan Workshop on Life-Cycle Cost Analysis and
need for probabilistic approach has been highlighted. Design of Civil Infrastructure Systems, Honolulu, Hawaii.
Should the designer assume higher risks or maximize Gehlen, Ch. & Schiessl, P. 1999. Probability-based dura-
safety with more conservative design? Although fur- bility design for the Western Scheldt Tunnel. Structural
ther study should be conducted, the authors estimate Concrete 2: 17.
that the cost-effectiveness of the structure is a design Gluver, H. & Olsen, D. 1998. Ship Collision Analysis.
priority. This means that, from a lifetime perspective, a Roterdam: Balkema.
certain level of risks should be accepted within a prob- Gucma, L. 2009. Method of Ship-Bridge Collision Safety
abilistic framework so as to ensure both reliability and Evaluation. Reliability & Risk Analysis: Theory & Appli-
cost-effectiveness. cations 2(2): 5063.
Hwang, E.S. 2009. Development of Vehicular Load Model
Consequently, extending the service life for bridge using Heavy Truck Weight Distribution (I) Data Col-
structures and especially super-long-span bridges lection and Estimation of Single Truck Weight. KSCE
requires a paradigm shift in the current design practice. Journal 29(3A): 189197.
Rather than focusing unilaterally on material dura- Hwang, E.S. 2009. Development of Vehicular Load Model
bility, the extension of service life should rely on using Heavy Truck Weight Distribution (II) Multiple
more systematic lifetime design procedures involving Truck Effects and Model Development. KSCE Journal
probabilistic approach, consequence-based design and 29(3A): 199207.
some marginal increase of costs. Imai, K. & Frangopol, D.M. 2001. Reliability-based assess-
ment of suspension bridges: application to the Innoshima
bridge. ASCE Journal of Bridge Engineering 6(6):
ACKNOWLEDGEMENTS 398411.
Incheon Bridge Corporation 2004. Concessionaire Supple-
This work is a part of a research project, performed at mentary Requirements Internal report.
Korea Bridge Design & Engineering Research Center International Navigation Association(PIANIC) 2001. Ship
Collisions due to the Presence of Bridges. Brussels,
of Seoul National University and supported by Korea Report of WG19
Ministry of Land, Transportation Maritime Affairs KICT 2009. DurabilityAssessment of ultra-high performance
(MLTM) as a Core Research Project 1 of Super Long concretes Internal report.
Span Bridge R&BD Center. The authors acknowledge Kline, E. 2009. Durable Bridge Coatings. Modern Steel
the financial support. The authors also express their Construction 11: 4650.
41
Koh, H.M., Lee, H.S., Kim, S. & Choo, J.F. 2008. Monitoring OHBDC 1983. Ontario Highway Bridge Design Code.
of Bridges in Korea. In Encyclopedia of Structural health Ontario Ministry of Transportation Communication.
Monitoring. John Wiley & Sons, Ltd. Park, C.M. & Park, J.C. 2003. Evaluation of structural
Korea Road and Transportation Association 2008. Korea behavior using full scale measurements on the Seohae
Bridge Design Code. KRTA. (in Korean) cable-stayed bridge. Proceedings of annual conference of
Korea Maritime Institute 2003. Korean Shipping statistical the Korean Society of Civil Engineers (KSCE), Korea,
handbook. KMI. (in Korean) 2003: 571576.
Kwon, S.-D. 2008. Aerodynamic Design of Long-Span Pedeferri, P. 1996. Cathodic protection and cathodic pre-
Bridges in Korea. Proceeding of the 4th International convention. Dipartimento di Chimica Fisica Applicata
ference on Advances in Wind and Structures(AWAS08), Politecnico di Milano, 6 February 1995.
Jeju, 2008. Petersen, P.T., Valsgaard, S., Olsen, D. & Spangenberg, S.
Kwon, S-.D. 2008. Wind resistance design of Kwangyang 1993. Ship impacts: Bow Collisions. International Jour-
Bridge. Proceeding of the 4th International Conference nal of Impact Engineering 13(2): 163187.
on Bridge Maintenance, Safety and Management (IAB- Raupach, M. & Dauberschmidt, C. 2003. Durability of
MAS08), Seoul, 1317 July 2008. marine structures made of high-performance concrete.
Lafarge 2007. Concretes for the 21st century. The Press In Ying-shu Yuan (ed.), Advances in Concrete and
Theme Book of Lafarge. Structures; Proc. International Conference ICACS 2003,
Luping, T. & Gulikers, J. 2007. On the mathematics of Xuzhou, 1719 September 2003. Bagneux: RILEM.
time-dependent apparent chloride diffusion coefficient in Rostam, S. 2005. Service life design of concrete structures
concrete. Cement and Concrete Research 37(4): 589595. A challenge to designers as well as to owners. Asian Jour-
Mahmoud, K.M. (ed.) 2006. Advances in Cable-Supported nal of Civil Engineering (Building and Housing) 6(5):
Bridges: Selected Papers. 5th International Cable- 423445.
Supported Bridge Operators Conference, New York City, Sarja, A. & Vesikari, E. (eds.) 1996. RILEM Report 14
2829 August 2006. London: Taylor & Francis. Durability design of concrete structures. London: E&FN
Nagai, M. 2004. Feasibility of a 1,400 m spa steel cable- SPON.
stayed bridge. Journal of bridge engineering: 444452. Sarja, A. 2010. Reliability principles, methodology and
Nilsson, L.O. 2009. Models for chloride ingress into con- methods for lifetime design. Materials and Structures 43
crete from Collepardi to today. International Journal of (DOI 10.1617/s11527-009-9486-y): 261271.
Modeling, Identification and Control 7(2): 129134. Stahl, F.L. & Gagnon, C.P. 1996. Cable Corrosion in Bridges
Nowak, A.S. 1993. Live load model for highway bridges. and Other Structures: Causes and Solutions. New York:
Structural safety 13: 5366. ASCE Press.
Nyman, W.E. & Moses, F. 1985. Calibration of bridge fatigue Svensson, H. 2006. Protection of Bridge Piers against Ship
design model. Journal of Structural Engineering 111(6): Collision. Keynote Lecture, Proceeding of International
12511266. Conference on Bridge Engineering, Hong Kong, 2006.
42
Managing old bridges
P.G. Malerba
Department of Structural Engineering, Politecnico di Milano, Milan, Italy
1 INTRODUCTION Conservation or damaging states of the structural

elements and of the joints, of the bearing supports
In the design practice, we are used to refer to recur- and of the auxiliary devices.
rence intervals of centuries. For instance, in the Euro-
All these issues concern environmental influences
pean countries, 200 years is the return period usually
on the bridge service life. The loads consist of self
assumed for floods. But, if we consider the real life
weight, of the stream flow and of the wind. The distor-
of a structure, how long do two hundred years last?
tions may be caused by settlements of the foundations
In fact, two hundred years ago, the material charac-
of the piers and of the abutments.
teristics, the building technologies and the theoretical
But a bridge is primarily a transportation link.
know-how were very different from the present ones.
Hence, many bridges modifications are caused by new
As regards to the materials, for instance, the indus-
traffic needs. Again, we remain struck by the strong
trial process to produce cast iron was invented at the
differences among the traffic loads given by present
end of eighteenth century, while a steel with modern
codes and those assumed at the time of construction:
characteristics was patented at the middle of 1800.
for instance, one hundred years ago, the maximum load
The first experiments on reinforced concrete were car-
for road bridges were 13.515 kN/m and the maximum
ried out in the second half of the nineteenth century.
rail bridge loads were 3060 kN/m. In those times, the
Mrsch published his famous book Der Eisenbeton-
speed was also lower and usually two lanes only were
bau in 1902, (Mrsch 1902). The first applications
sufficient.
of prestressed concrete structures were proposed by
This paper gives an account of studies and rehabili-
Freyssinet in 1928 (Freyssinet 1950).
tation works carried out on a group of bridges located
At the same time, the nineteenth century also gave
in the North of Italy and belonging to the main typolo-
us the basis of the strength of materials and of the
gies used in the years between 1850 and 1970. Part
theory of structures (Timoshenko 1953). In particu-
of these bridges lie on the reach of the Po river which
lar, with the growing use of steel in structures, more
delimits the southern border of Lombardia.
complete investigations of truss structures became
The others bridges cross minor rivers.
necessary and the first methods to deal with statically
Recalls will be made on the surveys and monitoring
indeterminate structures were proposed.
activities and on the problems addressed during the
The great arch and truss steel bridges date back
rehabilitation works.
to those times, while the first long span reinforced
Synthetic descriptions of the main interventions
concrete bridges appeared in the first decades of the
on the foundations, on the main structure and on the
twentieth century (Trojanovic 1960; Leonhardt 1982).
special devices will be given.
An epic history of the greatest cable suspended
bridges and of their designers was written by Henry
Petrowsky (Petrowsky 1996).
Dealing with old structures, we cannot avoid fram- 2 BRIDGES AND ENVIRONMENT. SURVEY
ing them into the context of their time, before assuming OF A GROUP OF BRIDGES ALONG THE
any sort of decision and/or choosing any sort of PO RIVER
intervention.
The major aspects that must be taken into consider- 2.1 The Po basin
ation usually regard:
The Po is the main Italian river. It rises in the western
Possible changes in the geo-morphology of the Alps and flows to the Adriatic sea, 652 km away. Its
territory where the bridge is located, due to settle- hydrographic basin is 74,970 km2 wide and receives 43
ments and to variations in riverbed geometry and tributary rivers. The total length of the embankments
in bathymetry conformation). is 3564 km.
Possible changes in the attitude of the bridge, made With the exception of the upper reach, until the end
evident by vertical irregularities of the road level, of nineteenth century the crossing of the Po was carried
by rotations of the piers or by excessive vertical out through river ferries and floating bridges. Due to
displacements of the deck. the wide span of its main branch, traditional masonry
43
Table 1. Characteristics of the main bridges examined (see Figure 1.a).
Total Length
Year of length over the Piers in Piers in
River Place construction [m] river [m] flood-plains riverbed Pier and foundation types
1 Po Casalmaggiore 1958 1,205.38 580.00 3 + 25 6 Pile/pier type

2 Po Viadana 1967 1,670.00 734.00 36 + 6 5 Hexagonal columns on piles
3 Po Borgoforte 1961 1,137.21 471.83 9 + 3 + 0 4 Pile/pier type
4 Po San Benedetto Po 196466 613.00 613.00 5 4 Double blade piers
5 Po Ostiglia 1929/1947 511.00 511.00 1 5 Masonry piers
6 Po Piacenza 1908/1947 1,096.00 607.00 12 + 5 + 3 2 Masonry piers
7 Serio Montodine 1970 64.00 64.00 0 1 Masonry piers
8 Oglio Pontevico 1970 90.00 90.00 0 2 Pile/pier type
9 Oglio Sarnico 1970 87.00 87.00 0 5 Pile/pier type
10 Oglio Montecchio 1 1970 270.00 270.00 9 2 Circular piers on piles
11 Oglio Montecchio 2 1970 90.00 90.00 1 2 Circular piers on piles
12 Oglio Breno 1970 403.00 403.00 15 2 Circular piers on piles
13 Po Pieve Porto Morone 1961 1,250.00 1,250.00 10 5 Rectangular r.c. columns on piles
14 Po Becca 1912 1,040.00 1,040.00 3 9 Masonry piers
and stone arch bridges involved limited spans resting General visual inspection.
on a high number of piers, having basements in an Geometrical survey, with dimensional cross-checks
insidious and wandering riverbed. with the original drawings. Detailed description of
The first long span bridges appeared at the begin- local damage states.
ning of the twentieth century with the diffusion of the Survey of the overall attitude of the bridge, with
steel-truss girders. Many of these road and rail bridges, respect to horizontal and vertical references. Verti-
although after some reconstruction work, are still oper- cality checks of the piers. Levelling of the roadway.
ating. After WWII, many new bridges were built, with Bathymetric survey, in order to draw the geometry
a wide use of prestressed reinforced concrete. The of the riverbed, to detect possible scouring signals
main challenge during the design and erection phases near the piers.
was not represented by the spans (usually 5070 m Underwater surveys, aimed to detect cracks or
on average), but by the interaction with severe fluvial clefts in the submerged parts, as well as traces of
hydraulics, characterized by cyclic floods, which often erosions or scour holes at the piers basis.
overtopped the embankments, invaded the floodplains Vertical boreholes in the piers aimed to measure
and sometimes upset countries and villages at the two their buried depth. Echo-soundings, used to check
sides of the river. the actual depth of the piers above the piles.
We have a fairly complete knowledge of the his- Geognostic boreholes, with the performance of
torical floods of the Po river since a remote age. The SPT and CPT tests and with collection of disturbed
main floods of the last hundred years occurred in 1926, and undisturbed soil samples.
1951, 1994 and 2000.
The relative cost of each survey activity is shown
in Figure 1c. The mean cost of the surveys versus the
2.2 A wide surveying campaign overall length of the bridges is shown in Figure 1d.
After the 2000 flood, the Compartment for the Lom-
bardia region of the Italian Agency for Roads (ANAS) 2.4 Laboratory and office activities
promoted a campaign aimed to survey the state of
the piers, of the basements and of the foundations of The results of the laboratory tests carried out on the soil
the main bridges crossing the Po. Such a campaign samples were used to define the load carrying capacity
was meant to provide a first evaluation of the bridges of the original foundation systems, as well as to design
state and to detect possible critical conditions of the the strengthening works on the insufficient ones.
piers and foundations, paying particular attention to The office activities concerned in hydraulic engi-
hydraulics causes of instability, like erosion and scour. neering and structural assessments.
The bridge main characteristics are listed in Table 1. For each bridge a comprehensive report was com-
Their locations are shown in Fig. 1a. piled. These reports gave a final assessment of the
actual state of the bridge and highlighted its possi-
ble critical faults. When the safety of the structure
2.3 Surveying and monitoring was at stake, suitable suggestions for urgent inter-
ventions were provided. Each bridge was classified
Surveying and monitoring consisted in the following
and given a priority level which implied recommenda-
activities:
tion for ordinary or extraordinary maintenance activ-
Preliminary exam of each bridge on the basis of the ities, or, in the worst cases, for radical strengthening
documents available in the archives. works.
44
Figure 1. (a) Position of the main bridges listed in Table 1; (b) maximum hydrometric at the Piacenza gauging station in the
last two centuries; (c) cost distribution for different types of test; (d) mean cost of the surveys per meter vs overall length of
the bridges.
2.5 Main results drawn from the bridge inspection As regards to the body of the piers, it was not
Floods are a common experience for people who live at found in a bad situation. Some masonry piers pre-
the Po riversides. Now, comparing the recent surveys sented losses of mortar among the masonry courses.
with the historical data, it is possible to observe a rising Both masonry and reinforced piles presented traces of
trend of the maximum flood levels and, in recent years, collision with the small boats and ships which sail the
a higher frequency of the flood events. middle and final reaches of the river.
According to the experts of environmental For the main bridges, the position and the ori-
hydraulics, such a phenomenon is mainly due to entation of the piers with respect to the main flow
anthropic factors, like a progressive waterproofing of was judged correct. In some minor bridge, spanning
the basins, due to urban and infrastructural grow- over tributary rivers, some cases of wrong foundation
ing, the removal of expansion zones and the increase basement were found (Figs 2c, d).
of river reaches confined by embankments. Climate Another general consideration regards the sound-
changes may also have contributed to these effects. ness of our probabilistic design procedures, tuned to
Figure 1b shows the maximum hydrometric levels values of return periods (100, 200 years). The surveys
recorded by the Piacenza measuring station during last of the most recent bridges, for which it was possible
two centuries and confirm these remarks. to compare the present riverbed profile with that of
Another element which was confirmed is the depth thirty/forty years ago, showed cases of strong riverbed
of the scour in the rapid transient phase as determined changes, with the movement of the main current from
according to the recent Po Basin Authority specifica- one alignment to another and also with the growth of
tions, which agrees with the most widely recognized temporary islands downstream the old main current.
formulations (Hamill 1999). These values of scour The underwater surveys reported that, even after
depth appeared quite higher than those assumed in the a long time from the end of the flood event, a sys-
past (Figs 2a, b) and strongly condition the load car- tematic encumbrance of debris remained at the basis
rying capacity computation of the foundations and of and along the body of the piers. This is a problem of
the piles. ordinary maintenance. But, who is in charge of the
45
Figure 2. Scour effects and debris action on piers in the riverbed: (a),(b) collapses caused by the scouring action in bridges
crossing river Pos tributaries; (c),(d) effects of the position, of the orientation with respect to the water flow, and of the depth
of the foundations in a small river; (e),(f) Multi-columns piers acting as grids in the formation of debris rafts.
debris removal? Is it the river Authority or the bridge the foundation systems;
Authority? More simply: who has to pay? the main structures of the bridges;
their complementary or special devices.
3 THE NEEDED REPAIR INTERVENTION

4 INTERVENTIONS ON FOUNDATIONS
Once the data acquired during the inspections and
surveys had been analyzed, different repair interven- Nowadays, the need for an intervention on the foun-
tion were set up. These interventions were carried dations arises mainly after their stability has been
out on: checked according to new codes and regulations,
46
which lead to higher depths of the estimated scour step by step, caring not to cut the reinforcing bars.
holes and higher values of the acting forces. The demolition was accompanied by the contempo-
The strengthening interventions are usually based rary reconstruction of reaches of the new, wider crown.
on two contributions. The first one consists in protect- Twenty horizontal threaded bars having a diameter of
ing the area surrounding the piers by means of big bags 32 mm, passing across the pile body and placed at two
containing massive stones. Such a work stabilizes the different levels (Figs 3b, d, e), strongly connect the
riverbed and leads to less severe expected scour depths. two opposite sides of the crown. Figure 3f shows an
The second intervention consists in strengthening the intermediate phase of the works. Before and during the
pier basement. works, the attitude of the bridge was topographically
Usually the strengthening is carried out in one of surveyed.
these two ways:
when the body of the pier is sufficiently compact
and massive, like in the case of masonry piers, new 5 MAIN STRUCTURES OF THE BRIDGES
piles are driven across the body itself;
when this is not possible or when previous repairing The interventions needed on the main structures of a
interventions occurred in the volume of the pier, bridge may be summarized as follows:
new piles are driven around the perimeter. Strengthening of the carrying structure due to some
A common problem of both these type of inter- lacks in their load bearing capacity, caused, for
vention is the accomplishment of a robust connection instance, by settlement effects, or by corrosion of
between new and old structures. the reinforcing bars or of the prestressing steel in
An example of strengthening is presented in the critical sections;
following. Refurbishment and adjustment of the structure to
new codes prescriptions, involving, for instance,
heavier load conditions;
4.1 An example of strengthening of the foundations Changes in geometry, due to the need of widening
of a bridge of the nineteenth century the road platform.
The Lodi masonry bridge cross theAdda River and was In the following, the case of refurbishment regard-
built in the years 1863-64, by the Milanese Architect ing an old arch bridge is examined.
Gualini. It is made of nine shallow arches spanning
18.90 m each, for a total length of 175 m (Fig. 3a).
In 197071, the deck of the Lodi Bridge was refur- 5.1 The railway bridge across the Gaggione River
bished in order to comply with the new traffic loads 5.1.1 The structure
(Figure 3b), and a weir was built 200 m downstream in The railway bridge across the Gaggione River was built
order to protect the basement of the bridge. In the fol- in 1885 and connects the city of Milan to Varese, 60 km
lowing years, the weir caused some floods involving away.
the town riverside. This brought, after new hydraulic The bridge, shown in Figure 4a, has a total length
assessments carried out in 2006, to the decision of low- of 130 m and is made of a sequence of seven stone
ering the weir. This lowering, which is scheduled for barrel arches, which rest on six stone piers. The barrel
the summer of 2010, will increase the flow velocity arches have internal radius of 5.82 m and their centers
and therefore expose the foundations to more severe are 14.00 m apart. The piers are slightly tapered. The
service conditions. After soil samplings and labora- longest one is 32.60 m high. The free height of the piers
tory tests, a subsequent geotechnical and structural is interrupted by an intermediate service deck made of
assessment showed that the foundation system was not five shallow arches.
safe even before a worsening of the conditions caused In 198586, the bridge was strengthened in order to
by the lowering of the weir would occur, and needed carry new, heavier train loads. The main interventions
strengthening works. regarded the basements of the central piers, the node
This choice was confirmed also by the presence of at the intersection between the piers and the interme-
some foundation settlements already observed in the diate deck, and the strengthening of the intrados and
past. In fact, some works had already been performed spandrels of the arches.
in 1947, when each foundation was strengthened and The intrados and spandrels were enveloped in a layer
its basement was protected against erosion and scour. of shot concrete, 180 mm thick, and reinforced with
The new intervention substantially recalls the old 14 mm bars arranged so as to form a mesh with a
one. Twenty new micropiles, having a diameter of pitch of 200 mm both ways. After a strong hydrosand
0.20 m and 24.70 m long, were drilled alternatively at a blasting in order to improve the chemicalmechani-
5 angle along the sides of the basement (Figs 3b, c). cal adherence, the added layers were linked to the
The micropiles were reinforced by valved steel tubes stone surfaces through a uniform curtain of pins and
having a diameter of 127 and a thickness of 10 mm confined against the original walls through 22 mm
(Fig. 3c). threaded bars, piercing the body of the arches. Other
Their position was defined so as not to interfere with works regarded the river bed stabilization, obtained
the old piles. The old concrete crown was demolished with the introduction of embankments and weirs.
47
48
Figure 3. Lodi Bridge (186364). (a) front view of the bridge; (b) transversal section showing the new deck, made of precast prestressed beams, the position of the new piles and that of the
threaded connecting bars; (c) arrangement of the micropiles at the two sides of a pier; (d), (e) details of the R.C. hooping crown; (g) photograph of an intermediate phase of the works.
Figure 4. Gaggione Bridge (1885): (a) front view of the bridge; (b) transversal section after strengthening, layout of the
relative displacement transducers and distribution of the strains in the depth of section during the loading tests; (c) comparison
between theoretical and experimental vertical displacements.
5.1.2 The problem actual static scheme and how old and new structures
It is quite obvious that, considering this type of mas- contribute to the overall bridge robustness.
sive bridges, any intervention which confines tightly Usually the arch is considered the main bearing
the masonry, and, at the same time, does not add exces- structure, while the superstructures (the walls and the
sive loads to the original structure, leads to an increase filling up, made of light non cohesive material), are
in the safety level. It is less immediate to define the considered as dead loads. Then, why it is so common
49
to see clefts around the arches keystones and even, in The bearing supports can also be considered as crit-
some cases, the loss of central bricks? ical parts in a bridge: their substitution in old bridges
is frequently not easy.
5.1.3 The bridge behavior According to the actual concepts of maintenance,
In December 1986, at the end of the strengthening the stays are considered as special devices that can be
works, severe loads test were carried out. Several substituted when necessary. But when the number of
load positions were examined and the vertical dis- stays is very limited, it is difficult not to involve the
placements were measured. At the intrados and at the entire structure.
lateral walls of the central section, eight electrome- In the following, the refurbishment of the ties of
chanical gauges (relative displacement transducers) one of the three A shaped frames which characterize
were placed as shown in Figure 4b. Figure 4c shows the Polcevera Bridge is presented.
the displacements obtained with the train in the cen-
tral position. The maximum vertical displacement was
v = 0.71 mm. Figure 4b shows the vertical distribution 6.1 The strengthening of the ties of one of the first
of the horizontal strains compared to the stratigraphy tied bridges. The case of the Polcevera Bridge
of the section: the added reinforcement layer and the 6.1.1 The structure
arch rib are in tension; the spandrels above the arch top The Polcevera Bridge was designed by Riccardo
and the infill material are compressed. A fairly good Morandi, built in the years 1960-1964 and put in ser-
matching between numerical and experimental data vice in September 1967. It flies over a large railway
was reached. parking lot and connects the A7 Genoa-Serravalle
Hence, during the service life and for the applied highway to the A10 Genoa-Ventimiglia highway. A
loads, the crown of the bearing arch works in tens- general layout of the bridge is shown in Figure 5. The
ion and the superstructure in compression. Analo- bridge is 1121.4 m long and 18.00 m wide. Its main
gous behavior was found through a FEA for the part is composed by three A shaped frames 90.2 m
effects of self weight. The theoretical final results high, supporting decks 171.9, 171.9 and 145.7 long,
of the analysis gave, for the most compressed fiber, connected by 36.0 m long suspended girders. At a dis-
= 0.28 N/mm2 , a relatively small value which can tance of ten meters from their ends, the long decks
be sustained also by a moderately compacted soil. are suspended to a couple of ties, made of prestressed
These assessments, which trust to the cooperation concrete.
with materials that cannot be defined as structural in This scheme repeats concepts already adopted by
the strict sense of the word (i.e. the filling of the span- Morandi for the Maracaibo Bridge, the Rio Magdalena
drels), cannot be used for safety evaluations. Safety Bridge and the Wadi Kuff Bridge (Morandi 1969).
derives from the certainty that the evolution from the
service to the ultimate state would involve the crushing 6.1.2 The concept of stay according to Morandi
of the filling material, while the line of the thrust lowers This type of ties is probably the most characteristic
until it reaches the extrados of the bearing arch, which elements of the Morandi system. The tie sections are
finally works according to the usual interpretation of shown in Figure 5b. Their construction sequence was
its behavior. no doubt complicated, but had a clear aim: to create ties
From this experience, many suggestions were that behave as an homogenous system made of tendons
drawn for similar interventions carried out in the working in tension and of a prestressed concrete case,
following years. working in decompression, but not in tension, under
the added and traffic loads. In this way, the fatigue
effects in the strands were limited thanks to the reduc-
6 AUXILIARY AND SPECIAL DEVICES tion of stress variations due to variable loads and, at
the same, the strands were protected against corrosion.
The service life of a bridge is strongly influenced by
the regular functioning of all its different components. 6.1.3 Needed repair interventions
A special mention must be made to the water After about 25 years of service, many parts of the
drainage system. It has a relatively modest cost if bridge presented severe damage states. On the ties
compared to other parts of the bridge, but it may of frame No. 11, at the Genoa side, clear corrosion
cause severe damages. Leakages through any points of traces in the strands of the tendons appeared. Minor
weakness in the waterproofing system or in the expan- damages were detected on the tendons at the top of
sion joints may lead to significant reductions of the the antenna of the nearby frame (No. 10) and in other
structural safety. parts of the bridge. In 199294, a recovery program
Other vulnerable systems are the expansion joints was carried out under the guide of Francesco Pisani,
which may exhibit worn rubber covering, torn up who was one of Prof. Morandis aides at the time of the
anchorage bolts and permanent deformations. A cause bridge design and planned the repairing intervention
of these drawbacks can be found in the dynamic effects phases (Martinez y Cabrera 1994). The main interven-
due to road platform discontinuities in correspondence tion concerned the four ties of frame No. 11. The basic
of the joints and to poor attention paid to the details concept of the intervention was to flank each original
during the construction works. tie with a set of 12 additional modern cables, in order
50
Figure 5. Polcevera Bridge (196064). (a) anchorage plates at the top of the antenna; (b) collar fastening the original stays;
(c) lower end ribbed plate; (d) anchorage of the tie to the abutment; (e) anchorage of the stay to deck transverse; (f), g), (h)
phases of demolition and of reconstruction of the end zones of the ties.
to transfer the suspension action from the ties to the is important in order to maintain the original design
stays. behavior of the bridge and to avoid any change in
Through specially designed devices (collars and the deflection and flexural behavior of the deck, with
new anchorage systems) and following the recovery consequences also on the elements of the main frame.
sequence (progressive tensioning phases) shown in Another aim of the progressive tension transfer from
Figure 5c, the new composed stay system, result- the old ties to the new composed stays, was to reduce
ing from the coupling of the old ties to the new cables, the structural risk of excessive compression stresses
maintains its original shape, while the stiffness char- in the concrete ties, avoiding potential bursting effects
acteristics remain very close to the original ones. This when the old cables were being cut.
51
At the end of this process, the compressive stress in of the type of concrete, a correct curing process,
the concrete ties was about 10 N/mm2 , as the original adequate cover and detailing of the reinforcing bars
design assumed. may considerably lengthen the service life and reduce
The repair works concerned also the other two maintenance costs.
frames. On frame No. 10 a local repair, aimed to Particular attention must be paid to the drainage
strengthen the upper end of the ties was carried out. On system. The lack of efficiency of the drainage system
frame No. 9, whose cables appeared less damaged with is one the main causes of damages and corrosion both
respect to the previous ones, no particular interven- in steel and concrete structures.
tions were adopted. Surface protection interventions Any effort to eliminate joints or, at least, to reduce
were carried out on all three frames. their dynamic effects, must be done.
It must be pointed out that the interventions were
carried out without traffic interruptions. Only some
traffic limitations were needed during the demolition ACKNOWLEDGEMENTS
and tensioning phases.
My first acknowledgments are for the members of my
6.1.4 Some final considerations on the Polcevera design team: Emanuele Barbera, Paolo Galli, Chiara
Bridge Malerba, Marco di Domizio, Giacomo Comaita and
The Polcevera Bridge and other Morandi tied bridges Giovanni Mineo.
represent an exceptional reference from the concep- Many of the works presented were carried out by the
tual, aesthetic and technical point of view, which is Author in cooperation with the late Prof. Francesco
even more relevant if related to the times in which these Martinez y Cabrera, to whom my memory and my
structures were built. Nowadays, however, similar gratitude go.
static schemes, though brilliant, cannot be proposed. The hydraulic studies were carried by the ETATEC
According to the modern criteria of durability, Company. Supervisor: Prof. Alessandro Paoletti
the prestressed concrete tie does not appear as a (Politecnico di Milano).
safe solution for elements in tension. Moreover, the All Italian Authorities who promoted special inves-
suspension action entrusted to a limited number of ele- tigations and demanding repairing works are grate-
ments, makes the whole structure little robust and the fully acknowledged. Among these: the ANAS (Italian
maintenance actions quite difficult. Agency for Roads) Compartment for the Lombardia
Modern bridge configurations, characterized by a region, Milan; the Lodi Municipality, Lodi; the Tech-
relative great number of stays (a curtain of stays), are nical Office of the Lecco Province, Lecco; FNM SpA
designed so that, should the failure of one of more stays (North Milan Railways), Milan; SSIF Spa (Swiss
occur, the subsequent loss of suspension action would Italian Railway Company), Domodossola.
be made up for by other suspension elements, making
the cables maintenance and/or substitution easier.
REFERENCES
7 FINAL CONCLUSIONS Basin Po Authority 1999. Criteri per la valutazione della

compatibilit idraulica delle infrastrutture pubbliche e di
interesse pubblico allinterno delle fasce A e B. Delib.
At the end of this presentation, the following conclu- N. 2/99.
sions can be drawn. Freyssinet, E. 1950. Souvenirs, Confrence prononce par
As regards to the relationship between the bridges M. E. Freyssinet a Paris. Beton und Stahlbetonbau
and the environment, on the basis of what it was February.
observed, the actual trend to avoid or to limit the Hamill, L. 1999. Bridge Hydraulics. London: E & FN Spon.
number of piers in riverbed, and to prefer an increase Leonhardt, F. 1982. Brcken Bridges. Struttgart: Deutsche
in the span of the deck, is confirmed and recom- Verlags Anstalt GmbH.
mended. Possible piers in the riverbed must have strong Martinez y Cabrera, F. & Pisani, F. 1994. Rehabilitation of
stays of the Polcevera Viaduct. Proceedings of the Int.
foundations, surrounded by suitable riverbed stabiliza-
Symp. on Cable Stayed Bridges, Shanghai: 640665.
tion devices. The piers must be correctly placed with Morandi R. 1969. Some Type of Tied Bridges in Prestressed
respect to the flow. Concrete. Concrete Bridge Design ACI SP-23: 447465.
As far as existing bridges are concerned, it seems Mrsch, E. 1902 Der Eisenbetonbau. Stuttgart: Wittwer.
that the old massive and well shaped piers behave bet- Petroski, H. 1996 Engineers of Dreams. Great Bridge
ter than some types of piers built in 1960s and 1970s, Builders and the Spanning of America. NY: Vintage
which tend to rake solid debris and are more vulnera- Books, A Division of Random House, Inc. N.Y.
ble. Moreover, a massive pier makes it easier to carry Timoshenko, S.P. 1953 History of Strength of Materials. NY:
out integrative works aimed to helping the original McGraw-Hill Book Company, Inc.
Trojanovic, M.S. 1960 Mostovi od armiranog I prednapreg-
foundations.
nutog betona do 1960. Beograd: Zavod za Izdavanje
Speaking about the main structures, the usual Udbenicka Socijalisticke Republike Srbije.
remarks on the durability of reinforced and prestressed
concrete structures can be recalled. A proper choice
52
The evolving science of terrorist threat mitigation for bridges
James C. Ray
US Army Engineer Research and Development Center
ABSTRACT
Since September 11, 2001 (911), a new threat has

been added to the list of all others that endanger our
nations bridges: Terrorists! Prior to 911, bridge engi-
neers would have considered it ridiculous to lump
terrorism amongst the more common natural threats
such as earthquakes, wind, and flood. Yet, national
and international events since that time have proven
otherwise.
This paper provides a glimpse of the current state-
of-the-art for terrorist threat mitigation for bridges and
how it has evolved from almost nothing since 911. Figure 1. Bridge destroyed terrorists.
The science is still very much in its infancy and has
far to go before it reaches the level of maturity and
definition desired by engineers endeavoring to address
the threat in their bridge designs and retrofits. While
it is very sad that this science even has to exist, it is
exciting for engineers and scientists looking to make
a difference in the world that there is still so much
room for innovation and development. As shown in
the paper, we are still looking for that silver bullet that
will effectively and economically mitigate all terrorist
threats.
Most importantly, the author hopes to inspire engi-
neers and scientists to find and pursue their niche
within this science so that they can make important
and innovative contributions, and ultimately achieve
what we all desire, to make this world a better- and Figure 2. Bridge with piers destroyed by terrorists.
safer place for all.
So, are bridges really threatened by terrorists? include explosive-packed land-based vehicles that
Immediately following 911, there were credible would be deployed against components reachable by
threats made against several major bridges within land and water-based vehicles that would be deployed
the United States, causing them to close for short against any components reachable by water. (2) Hand-
periods of time and to deploy military personnel for Emplaced Improvised Explosive Devices (HEIEDs):
their protection. Other countries, such as England, These include improvised explosive devices that while
have experienced direct terrorist attacks against their not as large as VBIEDs, can be placed in direct-
bridges over the years (Williamson, in press). And as or near-contact with a structural member and cause
shown in Figures 1 and 2, terrorists fully recognize severe localized cutting- and breaching type damage
bridges as viable targets. Thus, the security of trans- due to its close proximity. (3) Non-Explosive Cut-
portation infrastructure against terrorist attacks is now ting Devices: These include any non-explosive devices
an important issue for engineers. such as saws, grinders, and torches that can be used to
There are unlimited possibilities as to the types of cut/sever structural members. (4) IntentionalVehicular
terrorist threats that could be brought against bridge Impact: Like VBIEDs, these include both land-borne
structures (AASHTO 2003). This paper discusses and water-borne vehicles, depending on the location
threats in the following basic categories: (1) Vehicle- of the component of concern. (5) Fire: Fire of size and
Borne Improvised Explosive Devices (VBIEDs): These duration can cause structural members to lose both
53
their stiffness and strength. Thus, a pool fire from
such as a ruptured tanker truck on the deck of a bridge,
adjacent to key components or in the water adjacent to
piers or towers is of concern.
The science for mitigation of these threats is dis-
cussed in terms of the following sub-areas: mitigation
prioritization; vulnerability assessments; and threat
mitigation. Detailed discussions of these items are pro-
vided in the paper, but a brief overview is provided
below.
Mitigation Prioritization: Since 911, most bridge
owners have completed the prioritization of their
infrastructure between individual nodes (i.e. bridges,
tunnels, etc.) and have begun or are ready to begin mit-
igation efforts on their highest priority nodes. At this
point, the question once again arises: Where do we
start? Based upon the myriad of terrorist threats that Figure 3. Explosively damaged girders.
could be brought to bare as well as the large number of
vulnerable structural components on any given bridge,
there are almost unlimited mitigation measures that The paper provides an overview of the vulnerability
can be deployed on a given bridge.Yet, there are always assessment process for terrorist threats and discusses
limited resources. A rational and consistent means is various analytical tools that are available for this
required to assess and compare individual structural purpose.
component criticality and the effectiveness of varied Threat Mitigation: Physical security of any asset
mitigation measures throughout an individual bridge. must essentially comprise a layered and fully inte-
Thus, once again a prioritization is required, this time grated combination of four basic mitigation measures,
at the individual structure level. referred to herein as the Four Ds: Deter, Detect,
Instead of prioritizing among a group of structures, Delay, and Defend. These measures cannot be applied
the owner must now prioritize among the individual independently and must be employed as part of an
components on a given structure to determine which interdependent systematic approach to a layered secu-
are at highest risk and most in need of mitigation rity perimeter around the protected asset, in this case
efforts. The need in this case is to compare individual the bridge component or a specific critical bridge com-
bridge components based on their specific importance ponent. The paper provides discussion on each of
and vulnerabilities. Since most of the high-priority the Four Ds, but emphasis is placed on the Defense
bridges are major structures with potentially massive aspect (better known as hardening), which addresses
replacement costs and economic effects if lost, impor- the scenario where the attacker overcomes the denial
tance should be primarily based on a components methods and carries out the attack before a capable
contribution to overall structural stability; i.e. if the response occurs. Or, with the case of a vehicular-borne
component is sufficiently damaged, the bridge will device (i.e. vehicle bomb), the attack can be carried
totally collapse. However, other factors such as com- out so quickly and with such force that detection and
ponent replacement or repair costs can also factor in. denial methods are essentially of no use. Hardening is
Component vulnerability will be a function of the spe- the only viable defense for this threat.
cific threat type and size used against the component, The type of structural defense employed will of
the likelihood of such a threat, and the components course be threat dependent. Hand-emplaced explo-
resistance to the threat. There are many risk assessment sive threats, non-explosive cutting threats, fire, etc. all
methodologies that can be used for this purpose. This require radically different defensive measures. Signif-
paper provides an overview of a component level risk icant advances have been made since 911 for protec-
assessment methodology developed by Ray (2007) tion of vulnerable components from hand-emplaced
specifically for this purpose. explosive- and non-explosive cutting threats. How-
Vulnerability Assessment: In order to prioritize miti- ever, most of these technologies are proprietary and/or
gation measures on a given structure, the vulnera- Classified and thus cannot be discussed in the paper.
bilities of important structural components to each And while there is always room for improvement, fire
threat must be understood. Since damage from ter- and vehicular impact mitigation technologies are well-
rorist threats is generally very localized (Fig. 3), evolved and require no additional discussion. Thus, the
vulnerability assessments must be accomplished at the major focus of the paper is on hardening against vehi-
individual component level. cle bomb blast, which is a predominant threat against
Total bridge collapse will only occur if the locally which hardening is required.
affected structural components (i.e. column, truss In sifting through the myriad of potential solutions
member, tower wall, cable, etc.) are sufficiently dam- to structural hardening, the most important thing to
aged and structurally important enough to induce a remember is that no matter how exotic or high-tech
progressive collapse of the entire structural system. a proposed mitigation scheme may appear, it must
54
ultimately affect at least one of the three variables of
Newtons Second Law as given in equation (1).
where F = the summation of applied force and

resisting force; m = mass; and a = acceleration.
The response of any structural component to a blast
loading can depend upon many factors, but as shown in
equation (1) it will depend primarily upon: structural
mass; strain capacity and strength of the component;
proximity to the detonation (i.e. standoff); magnitude
of the detonation (i.e. explosive type and weight);
and support conditions for the responding compo-
nent. For points of discussion herein, equation (1) is
expanded to: Figure 4. First few microseconds of a detonation near the
face of a target.
where Fapplied = the applied forces (such as blast load-

ing) and Fresisting = any resisting forces, such as that
due to bending, shear and support reactions of the
responding component.
This basic law certainly does not require review, but
it is used to make several basic but important points
about the role of mass and resistance in the response
of a bridge component to blast. First: The greater the
mass, the higher its inertial resistance to acceleration;
and likewise the greater its strength, the more it will
be able to resist the applied explosive force. Con-
versely, once the mass is moving, the component must
have sufficient strain capacity (i.e. resistance) to over- Figure 5. Approximately one millisecond into same deto-
come the momentum of the moving mass and arrest nation as in previous figure.
its motion. If not, the component will fail and if suf-
ficient momentum remains, it will ultimately fly away In addition, many other high-strength, ductile,
as a fragment. and lightweight advanced materials have evolved in
In addition to these factors, magnitude and proxi- recent years and gained wide acceptance within the
mity of the explosion affect the applied loading design/construction communities. However, the reader
and impulse imparted to a bridge component. Near- is reminded that while these materials can have appli-
contact detonations, such as in Figures 4 and 5, cation to blast mitigation, the light weight (i.e. low
produce extremely severe pressure and impulse load- mass) can be detrimental in terms of inertial resistance.
ings. Because the detonation is so close-in and the Additionally, many of these high-strength materials
pressure durations are generally so short in relation to have a relatively low strain capacity.
the fundamental response mode of the component, the Decrease Applied Load: If sufficient resistance can-
response is more a function of the total blast impulse not be obtained for the component of concern, is it
and not the peak pressure. possible to affect the loading side of equation (2) and
Blast effects can be mitigated by any combination reduce the blast energy that is applied to the com-
of the following basic categories, all of which affect ponent? Blast loadings decrease very rapidly with
either the load- or the response side of equation (2): distance from the point of detonation (USArmy, 2002).
Increase Bridge Component Resistance to Load: Thus, the most effective means to reduce blast load-
Strengthening of a component against blast is accom- ings is to enforce standoff. This can be accomplished
plished in conventional ways, such as component via conventional means such as impact-resistant traffic
thickening, span shortening, etc. It is almost always barriers, etc. However, as previously discussed, this is
accompanied by an increase in the components mass, rarely an option on bridges as this generally requires
and as seen in Equation (2), this is always a good thing. narrowing- or closure of traffic lanes, and our nations
Mass is an important part on the resistance side of the bridges are so heavily taxed with traffic that this is
equation and it can significantly decrease structural generally not an option.
response and damage. Thus, the only remaining option in this category is
Steel has the significant benefit of high mass and to place a mitigation measure between the bomb and
many newer steels also offer high-strength and ductil- the target that has the effect of reducing the amount of
ity. While not as high in mass, concrete can offer very explosive energy that actually makes it to the protected
high strength and ductility if detailed properly. structural component. Numerous energy absorbing
55
concepts have been proposed and explored for this explosive energy does not go away and must be defused
purpose as it is clear that all materials demonstrate in some manner. Essentially, any concept only serves
various phase changes (an energy absorbing process) to re-distribute the energy; either through inertial
as a function of shock pressure. In addition, all mate- resistance, strain energy, or momentum transfer (frag-
rials absorb energy to varying degrees as they undergo mentation). Mitigation designers are encouraged to
gross irrecoverable volumetric strains (i.e. crushing). consider all of the strengths and weaknesses of each of
Beyond energy absorption, are there ways to the concepts and develop a layered hardening approach
completely- or partially shield the component of con- that capitalizes upon the strengths of each. An exam-
cern from the explosive energy? This can conceivably ple of a layered hardening approach for a cellular steel
be accomplished through blast barriers that can be bro- column is provided.
ken into two basic categories: structural barriers and Multi-Hazard Considerations: Terrorist threat miti-
sacrificial barriers. gation cannot be considered alone. In addition to
A structural barrier is essentially a wall in front of terrorism, there are many other hazards that threaten
the protected component that has sufficient strength to a bridge, including earthquakes, wind, water, fire,
stay in place throughout the blast event, collecting and weathering, etc. and a risk-based approach must be
dissipating all of the explosive energy and completely utilized to determine the relative degree of importance
shielding the protected structure. A sacrificial barrier of each threat to a given bridge. And, as funding is
provides shock wave reflection and inertial resistance always limited, the mitigation efforts must be priori-
just like structural barriers, but has minimal structural tized according to the level of risk. The paper discusses
resistance and breaks apart under the blast loading, mitigation measures that can address multiple hazards,
thus minimizing support reaction forces. The exper- such as wrapping of reinforced concrete columns to
imental testing of various materials for blast energy increase both seismic and blast resistance. In addition
absorption and shielding is discussed in detail in to the beneficial multi-hazard overlaps, discussion is
the paper. Materials considered included: elastomers, also given to detrimental overlaps of mitigation mea-
dilatants, porous aggregates, concretes, and water. sures. One example is: Local hardening of a structural
It is ultimately shown that energy absorption or component to increase its blast resistance may add
shielding is not effective for bridge components where detrimental mass and stiffness within the structural
bomb standoffs and shield thicknesses are minimal. system, affecting its seismic resistance.
Certainly the theory is sound, but unfortunately there
is just not enough space to place a sufficiently thick
shield with enough material to significantly affect the REFERENCES
extremely high blast pressures from a near-contact
detonation. American Association of State Highway and Transportation
Officials (AASHTO). 2003. Recommendations for Bridge
Load Path Redundancy: Complete mitigation of
and Tunnel Security, prepared by the Blue Ribbon Panel
damage from very large explosive threats may not be on Bridge and Tunnel Security, Washington, DC.
economically or logistically possible. In many cases, it Ray, J. 2007. Risk Based Prioritization of Terrorist Threat
makes more sense to just limit the extent of damage to Mitigation Measures on Bridges. ASCE Journal of Bridge
the most exposed components and ensure that there Engineering. March/April 2007: pp. 140146.
is enough redundant/residual capacity in other less US Army, Air Force, Navy, & the Defense Special Weapons
exposed members to insure that the structural system Agency. 2002. DAHS-CWE-UFC 3-340-01, Design and
as a whole can continue to function and not undergo a Analysis of Hardened Structures to Conventional Weapons
progressive collapse. Effects, Washington, DC.
Williamson, E., Williams, D., Holland, C., Bayrak, O.,
Layered Hardening Approach: A wide variety of
Marchand, K. (in press). Blast-Resistant Highway
hardening concepts are discussed and each has at Bridges: Design and Detailing Guidelines, Final Report
least some validity and usefulness for specific sce- to the National Cooperative Highway Research Program,
narios. There are also no concepts that just allevi- Project 1272.
ate the extreme loadings without consequences. The
56
Preserving the asset for its intended use
R.A.P. Sweeney
Modjeski & Masters, USA
1 INRODUCTION Table 1. Cost Effectiveness of various situations, Robertson

(1969).
The problems of maintenance are just as challenging
as those of design and construction. This paper draws Work not
heavily on concepts written by J.A. Robertson, FICE, Work Required Required
Superintending Civil Engineer, Ministry of Public
Fair ware Culpable
Building and Works, UK circa 1969 and was my guid- and tear, etc.
Neglect, etc.
ing philosophy throughout my career at CN. Robertson Work Done Cost Cost effective Simple
(1969). adequately effective Only when waste
I hope what I have to say will cause you to Restitution is
think hard about the difficult choices that have to be obtained
made in funding adequate and appropriate structural Work Done Wasteful Wasteful Simple
maintenance. inadequately Rectification Rectification Waste plus
What you begin with is critical to long term mainte- costs and costs and Wasteful
possibly possibly Rectification
nance requirements. If new structures are adequately
indirect indirect and/or
designed to shed water and deal with the elements and costs costs. indirect
are sufficient for the load to be carried through the involved Restitution Costs may
structural life, then maintenance will be less onerous. of costs may be incurred
A good quality protective coating applied in the be Prejudiced.
shop for structural steel, good waterproofing for con- Work not Indirect Indirect costs.
crete and careful treatment of wood products along done costs Possibly
with careful detailing will pay dividends for many loss of
years. Skimping here seems foolish. Opportunity
to seek
We need:
restitution
Good Design
Good Execution
Good Inspection These need to be minimized as much as possible
Good Maintenance through effective maintenance management.
Table 1 shows the main options.
We have all seen diagrams like Figure 1. Note in
2 MAINTENANCE particular that there is a category of ineffective costs
charged to maintenance. These are in effect a total
Maintenance is: Work undertaken in order to keep waste.
or restore every part of a facility to an acceptable
standard. Robertson (1969). 2.1 Inspection Sweeney (2010)
An acceptable standard is not necessarily perfection
Given the fact that unlimited funds are rarely available,
but what is adequate and appropriate to do the job.
there is a need for high quality inspection and sound
The inability to distinguish between what is adequate
engineering assessment. Inspection is the poor mans
and perfection is often a problem for engineers and a
insurance and the best line of defense to ensure safety
major complaint of professional managers. Remem-
and reliability.
ber as Voltaire said: Perfect is the Enemy of Good.
With regard to inspections, one size does not fit all.
Just make sure the Standard of maintenance is good
Inspection procedures for a bridge plant consisting of
enough.
proven robust designs needs to be different than one
Note that often work charged to maintenance is not
where there are weaker structures. Clearly, one of a
always really maintenance.
kind structures need special inspections designed for
Some is ineffective the individual structure, and usage should play a role.
Some is done inadequately Structures that are not used often might justify less
Some is unnecessary. frequent inspections.
57
Table 2. Compare 1964 to today. Sweeney (2006).
1964 Today
50 kg/m (100 lb/yd) 70 kg/m (142 lb/yd)

Wear and corrosion Fatigue
Design E 60 E 100 (E 80)
Average Steel bridge 58 Average age 95
years old
Car loads up 180%
Tonnage up 600%
100 years of service. Observe the changes during my

career and note that many of the bridges in service
today are over 100 years old and were designed for E
40 load.
2.5 Funding maintenance

Figure 1. Cost effectiveness of Maintenance.
The first place to look for needed funds is to areas that
Roughly 28% of North American rail bridges on are wasteful and neither needed nor effective.
the 7 principle railways are timber. The percentage is Engineers especially those who have to campaign
much higher on the independent short lines. What is for adequate budgets need to understand engineer-
not acceptable on a principle main line may be quite ing economics and how it differs from investment
tolerable on an infrequently used line. economics.
2.5.1 Rate of return calculations

2.2 Capital vs. labor Subjecting maintenance funding to rate of return cal-
This brings up the debate as to whether to repair or culations is not appropriate unless it is to determine
replace. The railway industry is a capital intensive and if a different standard of service or abandonment
labor intensive business. Most businesses are either of service is appropriate. Rate of return calculations
one or the other. The decision to replace may depend on always assume that the do nothing options is accept-
having an asset that reduces the need for full time staff able. The do nothing option is never acceptable if work
to perform routine maintenance or if capital is scarce it is needed to keep or restore every part of a facility to
may be the reverse. Tax and other considerations come an acceptable standard.
into play here. To emphasize again, rate of return calculations are
appropriate in deciding whether an asset should be
abandoned or put to other use.
2.3 Being complacent As an example many years ago the state of repair and
revenue from traffic on the narrow gage railway on the
Watch out for knee jerk reactions. Not all concrete fail- Island of Newfoundland put its continued existence in
ures are due to lousy concrete nor are all steel failures doubt. Maintenance was a continuing draw on funds
due to fatigue. Abraham Maslow, a noted American with no return and little or no long term prospects.
Psychologist quipped: If the only tool you have is a Simply put the cost of minimum annual maintenance
hammer, you tend to see every problem as a nail. exceeded revenues by a large margin.
Also watch out the next major failure type may have Studies to find ways of increasing revenues showed
already happened but as a profession we may not yet no long term prospects. A relatively simple rate of
be aware of it. We need to document and be aware of return calculation that was continuously negative con-
the cause of all failures. vinced the Government owner to abandon the line and
This is the main reason why Engineers in this field place their investment funds into the nearby highway
must not lose their engineering skills and become infrastructure as a more useful investment. Instead of
preoccupied with management and funding although pouring money down the drain this has proved to be a
these are essential concerns. more useful spur to economic activity in that area.
2.5.2 Present value calculations

2.4 Up-Dating budget allocations
Present value calculations are useful in deciding
Reducing maintenance to doing nothing or what we between two viable options, for example do I paint
have always done before is not necessarily adequate or now or do steel repairs later.
responsible. Periodic re-evaluation is necessary. In performing such a calculation the interest rate
Budget allocations when structures were new are chosen has to be consistent with the realistic remaining
clearly not adequate when structures are approaching useful life of the structure, should be in constant dollars
58
bridge plant that is expected to be needed for quite
some time to come was totally irresponsible even if
he intended to retire in 9 years. Fortunately, his staff
politely ignored his order.
Suppose it had been possible to run the structures
into the ground where and how would it be possible
to earn enough money to completely re-build? As all
managers know the first duty is to maintain share-
holder value and preserve the enterprise. Substitute the
word the peoples for shareholder if the enterprise is
Government owned.
Figure 2. AAA Bonds, Jones (1982). 2.6 Cutting corners

During times of economic hardship delaying mainte-
nance may be acceptable provided there is an under-
standing that making up for these delays will be more
costly in the future. Painting of structural steel is a
good example.
On one railway the decision was made in the 1930s
to stop painting steel bridges. The penalty came in
the 1960s and 1970s when major steel repairs and
replacements became necessary as a result of serious
corrosion damage. This increased expenditure became
a permanent fixture of maintenance budgets and will
Figure 3. Real rate of AAA bonds, Jones (1982). remain so for many years to come as corrosion resis-
tant steel was not introduced until the 1970s and a
and must not include inflation Jones (1982). This is no painting policy remained in effect. On the average
not an investment decision where not investing is an we are looking at another 30 to 50 years of steadily
option. Using investment interest rates are just wrong increasing expenditures on corrosion damage because
in spite of what your investment advisor or venture of the decisions made in the past.
capitalist advises. Another railway maintained a painting policy and
Venture Capitalists tend to want their money back their corrosion repair bills remain far lower in
quickly and this is appropriate when deciding whether comparison.
to invest or not. But once an asset is in place and A present value comparison clearly indicates that
needed, if its useful life before major upgrading is say the continuous maintenance painting option is the less
70 years then choosing an interest rate that completely costly option with two exceptions.
discounts all costs after 20 or 30 years is wrong. The first concerns those structures that were aban-
Although inflation has not been a major concern for doned long before their structural life was exhausted.
the past few years, I suspect it may be in the future. In North America many lines were abandoned as
To illustrate the concept of the effect of inflation Companies consolidated and traffic in some areas
consider the annual yield rate (%) of AAA corporate declined to levels where service could no longer be
bonds (Figure 2) and compare it to the real rate without economically provided.
inflation (Figure 3). The second is that period of time when paint-
It is seen that the average yield of these bonds over a ing became more costly than structural steel renewal
25 year period was 2%. The traditional time value of as paint manufacturers struggled to develop environ-
money has been more of an accounting of inflation and mentally effective acceptable systems that could be
risk rather than a reflection of the true earning power applied economically. This struggle is still ongoing
of capital stock in recent years. particularly for field applied re-coating.
2.5.3 Inappropriate policies

2.7 Dealing with upper management
Various Tax regimes also tend to use distorted interest
rates and depreciation rates that are useful as pub- In dealing with upper management or the owners it is
lic policy but are total irrelevant to the structural important to understand that the man of knowledge is
maintenance decision process. expected to take responsibility for being understood
Any policy that does not preserve the asset to an and not the other way around. The maintenance man-
acceptable standard throughout its full useful life is ager must take responsibility for his contribution and
not a good policy. ensure that his product that is, his knowledge is
There was one Railway Chief Engineer who ordered effectively understood and used. Drucker (1985)
that all maintenance funding decisions be made on the Thinking out loud to senior executives not only
basis of 10 years of remaining life. This for a large wastes their time but can be very counter-productive.
59
These people need to be given clear actionable effec- Table 3. Maintenance Levels.
tive choices that will accomplish the desired results.
They need to be respectfully led to do the right thing. Level Functional Safety of Preservation
Performance Persons and of the asset
Quality and Property in and the
reliability all aspects amenities
3 MAINTENANCE MANAGEMENT
Minimum Nil (Not All legal All legal
Applicable) Requirements Requirements
Maintenance management can be divided into 5 main To be met To be met
possible action levels:
Preservation Nil (Not Fully safe Prevention of
1. A minimum level to comply with legal, statuary and Applicable) in all deterioration
contractual requirements. respects for of asset and
2. A slightly better level designed to preserve the asset. reintroduction amenities
to service
3. A normal level of economical good practice.
on 7 days
4. A higher level (for facilities that are operationally notice
vital, structurally hazardous,
Normal Availability Fully safe Meets
5. An even higher level for assets of prestige value,
Factor at in all specified
etc.) full specified respects for condition of
Note that in level 1 it may not be possible to restore performance regular use facilities and
the facility to adequate operational condition with- not less amenities
than 0.85
out considerable expenditure since only the minimum
legal and similar requirements are met. Operationally Availability Fully safe Meets
At level 2 it would seem that reintroduction to Vital Factor at in all specified
full specified respects for condition
service in a week would be a reasonable expectation.
performance regular use of facilities
At level 3, the asset should be available for service not less under specified and amenities
85% of the time at a minimum. than 0.98 overload and
At levels 4 and 5, the asset should be available 98% abnormal
of the time under specified overload and abnormal con- conditions
ditions. The difference with level 5 is that the facility Prestige Availability As above To be in first
should also be in impeccable condition. Factor at class order
In deciding what to do and how much to do the full specified and
answers to the following questions need to be quan- performance impeccable
titively assessed as each comes with a rising level of not less condition
cost: than 0.98
Usable with work

Usable with inspection
Generally reliable 5 SAFETY
No service interruptions
Safety comes in two main aspects, the ability of the
Table 3 summarizes the options. structure to fulfill its function safely and safety of
personnel using, inspecting and maintaining it.
4 BRIDGE MANAGEMENT SYSTEMS
5.1 Structural safety
These are useful and essential tools to determine where
resources need to be focused but unless one has an Structural safety means being strong enough and with
army of persons to input data their usefulness on a par- geometric properties that allow its use in a safe manner.
ticular structure is often overrated. It is very important Make the best of existing conditions. Do not con-
to understand the difference between decisions that demn a structure because it is not what it ought to be but
affect a population of bridges and those that affect a ask what it is good for under probable and not under
particular unique structure. Management Systems at practically impossible conditions of traffic, Waddell
the individual project level are and need to be far more (1916).
comprehensive in detail. The expense of some man- A Phoenix truss that at first glance looked to be seri-
agement systems would be better allocated to fund ously overloaded upon close examination proved to be
maintenance. adequate given the weekly inspection that it received
If you think you can effectively manage a bridge and the very low volume of traffic. The end circular end
plant from an air conditioned office then think again. posts had been filled with concrete greatly increasing
We all should know the science of large numbers. In their capacity and the bottom chord was pre-stressed.
my country the chances of winning the Lottery are More interesting was the fact that the wrought iron
about one in 14 million yet someone wins almost every web members of this through truss were strength-
week. The chances of having a rogue bridge are about ened with welded plates. The welds broke regularly
the same. Think of the I-35 disaster. but a welder examined all of these welds before each
60
6 REDUCING THE STRESS STATE
OF THE RAILROAD
This has been a slogan for the Research Group of

the Association of American Railroads for quite some
time. The objective is to reduce the stress state so as
to prolong the useful life of all assets including bridge
assets.
The movement of a steel wheel on steel rails is ide-
ally supposed to be a very low impact smooth operation
so efforts need to be made to make it so or as nearly as
possible so. Correcting potential problems early can
save bigger problems later
Removing joints with continuous welded rail
reduces impulsive loads.
Figure 4. Stress range tests results for 100 steel bridges. Keeping approaches smooth and well tamped and
supported reduces the bump and bounce at the end
train passage and any failed welds were repaired. The of the bridge.
Good Car behaviour is important.
structure proved adequate for its use.
Wheel and rail condition. The smoother these are
5.1.1 Real stresses the less the impulsive loads.
Car loading/overloading.
Actual stresses or stress ranges often differ from what
Watch for the effects of corrosion and fatigue on
was assumed in design. Some times for the better and
sometimes not. steel spans.
Although many North American steel bridges were
Load testing and strain gage work often indicates
that live load stresses are less than expected but as the sufficiently over-built that poor bearings can be
figure 4 shows this is not always the case. effectively tolerated this is not always the case.
Apply Ultrasonic Impact Treatment to details that
are sure to cause problems.
On timber bridges keep the geometry tight and well
5.2 Worker safety
aligned. If you dont they shake apart. This can also
There needs to be provision for inspectors to safely do apply to steel structures.
their task and for workers to accomplish maintenance. On timber bridges mechanical wear and environ-
One thing that is very important to remember is mental decay should be dealt with sooner rather than
that ladders, walkways, hooks and railings need to be later.
inspected. Often these are made of much thinner mate- My personal view is that a bit of extra ballast on a
rials than the structures to which they are attached concrete deck as long as the structure can handle the
and thus will deteriorate much sooner than the main extra weight is a good idea. It certainly attenuates
members unless special provisions are made to ensure impulsive loads.
adequate maintenance. Watch out for hidden problems in concrete and
Simply putting hand rails for inspectors to tie off to pre-stressed concrete spans. Reinforcement deteri-
on small simple structures can be adequate. oration can be a critical issue.
Major structures need more elaborate devices for
inspection.
5.2.1 Safe work methods 7 RESEARCH

One of the most frequent causes of serious failures is
in the area of instability. Instability comes in various The role of test facilities in giving guidance to what and
forms of buckling in compression and brittle fracture when things need to be done is very important. Most
in tensile situations. During repair and maintenance of us are aware of the long term impact of the ASHTO
work it is essential to ensure against these two. Road tests in the late 1950s on pavement design and
During the retrofit of a large timber trestle, two maintenance but it influenced bridge work as well.
6 6 inch by 12 foot struts (152 152 mm by 3.66 m) The future US Federal Highway 20 year Long-Term
that had been removed were inadvertently not replaced Bridge Performance Program (LTBPP) will develop a
before a work train attempted to cross the bridge. The better understanding of how current highway bridges
bridge collapsed like a deck of cards killing several perform under a variety of service conditions. This
people. program will monitor bridges in a number of States
Laying out the sequence of a job so workers are and will be of long term benefit to bridge maintenance
working on a safe surface seems so obvious but practice.
sadly I have seen a few tragedies that resulted from Recently in the railway area, research test tracks
thoughtless job sequencing. in Pueblo, Colorado, and the Czech Republic are
61
providing guidance on what is critical for long term
maintenance.
At the US Department of Transportations FAST
facility in Colorado that is managed by the Transporta-
tion Technology Center Inc. (TTCI) of the Association
ofAmerican Railroads there are three pre-stressed con-
crete bridge spans and a steel bridge with two spans.
A train of 315 000 lb. (1.4 MN) cars circles a track
loop at 40 mph (64 kph), reversing directions daily, and
can accumulate the same traffic in a month that a very
heavily travelled mainline railroad would see in a year,
Sweeney (2007).
7.1 Open deck fastenings

For many years open timber decks have been fastened
with a number of devices. All were somewhat of a Figure 5. Crack into bottom flange.
maintenance headache. The FAST facility was used to
test a number of devices and within a few years of test- (sleepers) as well as approach conditions have been
ing a simple innovative solution was found that greatly monitored.
increased the time between needed maintenance.
7.6 Hybrid-composite bridge Test at FAST
7.2 Movable rail joints
A hybrid-composite beam span was installed and has
These joints are a constant maintenance headache, accumulated over 120 MGT with no adverse effects
add impulsive loads into a bridge and are the bane noted yet.
of every track maintainer. Testing at the FAST facility
has developed a rail joint for movable bridges that lasts
more than twice as long as previous versions. 7.7 AAR timber testing
Research work was performed at an AAR Affiliated
7.3 Continuous welded rail on bridges lab (Texas A & M University) to evaluate the effect of
interior timber rot in circular piles with varying Length
Recent work by the TTCI Group of the AAR has to least Dimension (L/D) ratios, Sculley (2004).
clarified the behavior of CWR on bridges for mild Additional work was carried on to determine the
winter to hot summer conditions. This has gone a shear behavior of timber stringers under large cyclic
long way to clarify the seemingly contradictory prac- loading. The result was to increase allowable stresses
tices of various railroads. This testing better defined from 75 to 150 psi, Fry (2004).
those instances where it is possible to eliminate rail
expansion joints. 7.8 Preparing for increased loading
Further work is required at very cold temperatures.
These and other tests are on-going and were initiated
to ensure that there would be no serious economic sur-
7.4 Steel bridge tests prises as North American Railroads increased loads
The Steel Bridge at the Fast facility has a serious flange by 20%.
crack that has been in service carrying 315,000 lb. The result has been good information on what to
(1.4 MN) cars for over 1400 MGT (Million Gross Tons retrofit or strengthen and also some good indications
(US)) or over 8 million cycles of loading. Careful mon- of deterioration rates on many components.
itoring of this crack with and without rail joints on
the span has given information on the threshold stress
7.9 CN timber tie (sleeeper) tests
range for which there is no crack growth and the stress
range for which growth occurs. In evaluating timber bridge ties for possible movement
Given that there are 47 other cracks in this welded to higher axle loads, then current accepted practice
bridge, various repair and retrofit schemes have been indicated that most of that Railways bridge ties used
evaluated as well. on open decks would be seriously overstressed.
A number of non-destructive testing tools are being Given the tremendous cost of replacing these ties, it
evaluated as well. was decided to embark on a testing program to estab-
lish the capacity of a large representative sample of the
bridge ties it was purchasing as well as those that had
7.5 Pre-stressed concrete bridge Tests at FAST
been in service for quite some time.
The pre-stressed concrete bridges have accumulated The result of the study, Madsen (1999), indicated
over 800 MGT so far with few problems. Major the CN Ties to be much stronger than analytically
studies on ballast depth, ballast mats and different ties predicted.
62
8 FUTURE TO DO LIST 9 CONCLUSIONS
Issues that need further research to increase mainte- Preserving the asset for its intended use involves work
nance effectiveness. undertaken in order to keep or restore every part of a
facility to an acceptable standard. This paper expanded
We need a methodology to use anecdotal data. There
on that concept and how to achieve it.
are thousands of structures in service that can tell
us a lot.
We need a methodology to use the research test data REFERENCES
that did not lead to failure. A lot of tests get stopped
before failure occurs, yet we have no agreed way of Drucker, Peter F., The Effective Executive Harper & Row,
making use of this information. NY, 1985.
We need a much better understanding of the real FRA Track Safety Standards, Federal Railway Administra-
load environment. tion, Office of Safety, Washington, DC, 2005.
The gap between design and prudent limits of use Fry, G.T., Borchers, C.R., Otter, D.E.et all, Fatigue Testing
needs further investigation. of Full Scale Solid Sawn and Glued Laminated Railroad
Bridge Stringers, R-969, AAR-TTCI, Pueblo, CO., 2004.
The effect of infrequent loads on fatigue:-does one
Jones, Byron. W., Inflation in Engineering Analysis, John
car in 1000 or 10,000 really effect the life? Wiley, NY., 1982.
What light vehicles can be ignored in fatigue life Madsen, Borg, Sweeney, Robert A.P., Shear Strength of
evaluations? Douglas FirTimber BridgeTies.Transportation Research
How do we avoid legal or political concerns from Board, Washington, DC 1999.
preventing the dissemination of valuable research? Robertson, J.A. The Planned Maintenance of Buildings and
Capacity problems will force higher loads and more Structures Paper 7184S. ICE, London, UK, Oct. 1969.
frequent loads on our bridges. What can we do to Sculley, P.J. Fry, G.T. et all, Static Tests of Timber Bridge
affordably accommodate these? Piles Draft Report R04-02sa, TTCI, Pueblo, CO., 2004
Sweeney Robert A.P., The Art of Railway Bridge Engi-
Deflection criteria. As capacity problems get worse, neering Today Keynote Address, 7th International Con-
the pressure for higher speeds on railways will grow. ference on Short and Medium Span Bridges, CSCE,
Current design deflection limits on steel and timber Montreal, Canada, 2006. Sweeney, Robert A.P., Otter,
in North America are quite liberal and in fact as Duane, A Review of North American Railway Bridge
speeds increase the deflection of the span will absorb Research IABSE Symposium Weimar, IABSE Zurich,
most of the tolerance now allowed for rail., FRA Switzerland, 2007.
(2005) Sweeney, Robert A.P., Unsworth, John F., North American
Educators and engineering supervisors need to Railway Bridge Inspection Practice Paper submitted and
remember to teach is so far as possible that Engi- accepted by J of Bridge Engineering, ASCE, Washington,
DC, 2010.
neering is an Art that uses Science and not just
Waddell, J.A.L., Bridge Engineering Vol. 1 and 2,
applied science. J. Willey & Sons, New York, NY 1916.
Engineers need to develop simple intuitive tools to
ensure their conclusions are in the right ballpark.
Particularly in this computer age.
It is very important that older engineers tell their
stories so that mistakes are not continually repeated.
Continuous relevant learning and improvement are
a must.
63
On longevity and monitoring technologies of bridges a survey study by

Japanese Society of Steel Construction
E. Watanabe
Kyoto University, Chairperson of Board of Directors of Regional Planning Institute
H. Furuta
Professor of Informatics of Kansai University, Takatsuki, Osaka
T. Yamaguchi
Professor of Osaka City University, Osaka
M. Kano
JIP Techno-Science Co., Osaka, Japan
ABSTRACT In the world of construction industry, the era of the

new construction seems to be ending in most of the
JSSC, Japanese Society of Steel Construction, has developed countries and how to maintain and manage
organized a special task committee and conducted the existing facilities are considered to be one of the
a study on the recent development of maintenance, most important paradigms. On the other hand, as the
renewal and retrofitting for buildings and bridges of socio-economic activities continue in a large extent,
Japan. This paper describes the outline of the study the scale of the production, consumption and scrap
by a working group of bridges and consists of two have been significantly increasing and the exhaus-
parts. The first part is concerned with the survey on tion of resources and the destruction of environment
the technologies for the longevity of bridges whereas are being rapidly accelerated throughout the world. It
the second part is specifically concerned with the is needless to say that the construction environment
structural health monitoring technologies reflecting should be drastically changed from the era of scrap to
recent developments. In view of the fact that important the ecological use of existing stock of infrastructure.
terminologies such as lifetime and service life have Therefore, the structural longevity and elongation of
been used differently throughout the world, some fun- the lifetime of infrastructure should be regarded as
damental terminologies are reviewed, discussed and increasingly important subjects at the present time and
redefined. Furthermore, recent developments on the in the future.
structural health monitoring technology are reviewed According to the white paper of the Government
and the results are summarized in a matrix form. Lastly, of Japan, the lives of typical infrastructure such as
voices of some local governments, owners of express- roads and bridges are thought to be approximately
ways and railroads are taken into account in the form of 60 years and that of harbors and coasts is 50 years,
questionnaires and replies to propose future structural respectively (MLIT 2002). Furthermore, according
health monitoring system for traffic infrastructures. to Nikkei Newspaper, the annual cost of mainte-
nance, management, replacement of public infras-
tructure and restoration from natural disasters of
1 INTRODUCTION Japan would be doubled from that of 2004 to 2030,
namely would become from 5 trillion JPY to 10 tril-
In recent years, with some exceptions the number of lion JPY. According to the detail of the white paper
children is significantly decreasing and young people of Japanese Government, the following facts on the
tend to seek easy jobs which do not necessarily require infrastructure are worth mentioning. Assuming the
them working hard with perspirations in the developed total investment remains as it is, the figures from
countries in particular and in some developing coun- those in the year of 2000 will increase or decrease
tries. In this view point, the education of engineers respectively annually to those in the year of 2025 in
and important traditional technologies that have been Japan:
acquired through generations are going to be forlorn (1) The investment for the maintenance is predicted
and these facts are in fact becoming deplorable social to increase, specifically, from 3.8 trillion JPY to
problems. 6.2 trillion JPY.
64
(2) The investment for the renewal is predicted to 2.3 Function and performance
increase, specifically, from 0.3 trillion JPY to 3.7
Performance refers to the structural capacity based
trillion JPY.
on the field inspection data and the structural health
(3) The investment for the newly-built is predicted to
assessment taking into account the deterioration of
decrease, specifically, from 15.9 trillion JPY to 9
infrastructure. On the other hand, the function implies
trillion JPY.
updated capacities meeting the current standard of the
(4) The stock of the infrastructure will be vastly
wheel loadings, natural forces such as wind and earth-
accumulated.
quake excitations, river flow, design traffic flow and so
forth.
2 LIFETIME OF BRIDGES It is a general practice to decide the maintenance
plan based on the field inspection data and the struc-
2.1 Definition of lifetime tural health assessment of bridges taking into account
their deterioration rate. However, the old bridges built
It is quite confusing to know that the definition of the
in the past were based on different old codes from
lifetime varies considerably from places to places. In
now with respect to loads, earthquake-resistant design
this paper, the lifetime may be defined as the period
and river conditions and so forth, thus their func-
of time since structures have started to be in service
tion does not correspond to the present design codes.
until they cease to be used for some reasons or the final
Thus, the judgment on the bridge replacement solely
stage when they are possibly in service any further
on the basis of the bridge health assessment alone
(JSSC 1991).
may lead to wrong decisions. From this standpoint,
the final decision requires the functionality in addi-
2.2 Definition of expected lifetime tion to the health assessment as explained in the
following. In this paper, the countermeasure such as
The expected lifetime may be defined as the period
earthquake-resistance retrofitting is not considered for
of time in which structures are expected to satisfy
mere convenience. However, base-isolation will be
the demand performance, to possess the physical
regarded as a method to improve the performance in
load-carrying capacity and to fulfill the serviceability.
this paper.
2.4 Classification and concepts of lifetime

Lifetime may be classified into physical, functional
and economic lifetimes (Kato 1983) as shown in
Figure 1 and Table 1.
(1) Physical lifetime
It refers to the lifetime of an existing bridge
which may have to be renewed by a new bridge
upon significant reduction of the load-carrying
capacity or by the troubles encountered during
the service but deemed difficult to meet these
demands by just reforming the present bridge. As
shown in Figure 1 (a), it represents the period
from the initial state of a bridge to the time when
its performance decreases and coincides with the
performance corresponding to the serviceability
limit. It frequently happens to restore the per-
formance by intermediate repair, retrofitting or
replacement works.
Table 1. Comparison of Lifetimes.
Figure 1. Chronological change of lifetime.
65
(2) Functional lifetime inspection, system, timing, judgment, measure and
It refers to the lifetime of an existing bridge budget to prolong the lifetime beyond the expected
which may have to be renewed by a new bridge lifetime.
when the widening of roadway width or the up- The keywords for the longevity may be summa-
grading of the traffic loads are legally demanded rized as:
but deemed difficult to meet these demands by
(1) Increase of bridge stocks
only reforming the present bridge. As shown in
(2) In view of the limitation for budgetary appropria-
Figure 1(b), it represents the period from the ini-
tion, the scenario of renewal of every bridge does
tial state of a bridge to the time when its demand
not exist any more.
performance is changed beyond the present per-
(3) The national and local governmental principle for
formance.
the preventive management is supported by the
(3) Economic lifetime
authorized experts.
It refers to the lifetime of an existing bridge
(4) The concept of longevity for those newly-built and
which may have to be renewed by a new bridge
already existing are different.
when the up-grading the existing bridge is judged
(5) The performance-based specification is encour-
much more expensive than building a new bridge.
aged for the longevity by adopting innovative and
In Figure 1(c), it represents the period from the
development technologies.
initial state of a bridge to the time when its cost
for the repair and retrofitting become greater than Figure 3 shows the ratio of the bridges with the
that for the replacement. lifetime over the age of 50 to those under 50 (MLIT
Table 2 shows several examples of physical, func-
tional and economic lifetimes.
2.5 Chronological change of reasons for

replacement of bridges
Figure 2 shows the chronological change of reasons
for bridge replacement in Japan (PWRI 1997, BMSG
2004). Figure 2(a) describes the reasons for bridge
replacement during the period of 19771986. These
reasons may be listed in the order of larger percentage:
(1) Improvement of road alignment
(2) Improve the functional obsoleteness
(3) Structural defects such as corrosion
(4) Insufficient load-carrying capacity
(5) Insufficient earthquake-resistance
After 10 years, some of the reasons for replace-
ment of bridges have changed as shown in Figure 2.
For example, the replacement due to damage corre-
sponds to only 1/5 of the total replacements. Most
of the replacements are due to the improvement of
road alignment, refit of river flow and widening of
road width. The percentage of the improvement of the
functionality changed from 57% to 76%.
2.6 Definition of technology toward longevity

The longevity technology refers to the technology
and all methods of management including those for
Table 2. Examples of physical, functional and economic

lifetimes.
Figure 2. Chronological change of lifetime of Japanese

roadway bridges.
66
Table 3. Chronological characteristics of replacement of
steel bridges (NILIM 2004).
Figure 3. Chronological change of ratio of the bridges with

the age over 50 to those under 50.
Figure 5. Target lifetime for bridges.
2.8 Management for longevity

The lifetime can be made longer or shorter than the
initially expected one depending on the management
method.
Figure 4. Performance and lifetime of bridges (Abe 2008).

2.9 Setting of basic parameters for determination
of lifetime for steel bridges
2008). It may be apparent that after the year of 2030, The National Institute for Land and Infrastructure
the ratio of bridges over the age of 50 becomes more Management, NILIM, showed the statistics of the
significant. number of replaced bridge plotted against the year of
replacement.
The statistical distribution is assumed to follow the
Normal distribution and the parameters are so deter-
2.7 Target lifetime mined as to minimize the errors between the achieved
record and the prediction. Table 3 shows the average
The target lifetime is not defined by the structures but
lifetime and standard deviation of bridges built from
by the manager.
1920 to 2000 (NILIM 2004).
(1) It is the matter of consciousness of the managers For those bridges built during the period of so-called
and the public opinion. rapid economic progress from the year 1961 to 1980
(2) The limit performance for management varies the statistics show that the average lifetime is 70 years
depending on the demand performance of the with the standard deviation of 20 years. Thus, if the
time. normal distribution is assumed and 95% fractile of
(3) Theoretically, steel structures are considered to be non-exceedance is assumed, this fractile corresponds
ever-lasting structures. to 1.65, the longevity of prolonged lifetime of 35
(4) The target lifetime may be considered to be 1.5 more years can be expected (20 1.65 = 35) as shown
time of the lifetime of existing structures as shown in Figure 5. While those bridges built in recent years
in Figure 4 (Abe 2008). from 1981 to 2000, the average lifetime is 100 years
(5) Even at the end of the target lifetime, the structures with the standard deviation of 30 years. By the similar
may not have to be renewed if they posses the token, the longevity of 50 more years may be expected
sufficient structural health. (30 1.65 = 50).
67
Table 4. Expected average lifetime of bridges for different
probabilities of non-exceedance.
Figure 7. Causes of damage of bridge parts and elements.
locations of damaged steel parts or members, causes

of damage and the type of damages.
It will be seen that the fatigue and corrosion are the
most popular damages for steel bridges.
Figure 6. Vision of longevity technology and expected
developments of technologies for each step of construction
projects. 3.5 Anti-fatigue technologies
For road bridges, the fatigue design has been intro-
If different fractile other than 95% is assumed, the duced in Japan only quite recently. The counter-
expected average lifetime for steel bridges may be as measure consist of reduction of stress concentration,
listed in Table 4. Thus, for recent bridges built after reduction of residual stresses. Not only in the main-
1980, the expected lifetime may be regarded to be tenance but also they should be taken into account at
150 years. the planning and design steps. Recently, the following
two countermeasures are to be noted.
3 PRESENT STATUS OF TECHNOLOGIES FOR (1) anti-fatigue steel The steel itself has a desir-
LONGEVITY OF BRIDGES able characteristics of fatigue-resistance through
the adjustment of chemical ingredients and metal
3.1 Change of social demand texture.
The philosophy of essential maintenance, namely, (2) UTI, Ultrasonic Impact Treatment Impact of
rebuilding bridges when they become older than the ultrasonic wave is applied to the weld in order to
lifetime in the past is nowadays shifting to the philos- have the following effects:
ophy of preventive maintenance and building durable (a) Tensile residual stresses can be converted to
bridges to prolong their service life. compressive.
(b) Reduce the stress concentration by smoothing
3.2 Type of damages for steel bridges out the surface of weld.
(c) Increase the surface toughness.
Most damages are caused by the defective structural
characteristics and environmental effects. The most
popular damages of steel bridges are considered to be 3.6 Anti-corrosion technologies
fatigue and corrosion.
As has been described above, the corrosion is regarded
as one of the most serious problems for steel struc-
3.3 System for longevity technology tures. Not only to those in the area near the sea
zone susceptible to salt water but to those cold area
Figure 6 shows the vision of longevity technology and where anti-freeze is used inevitably, bridges tend to
expected developments of technologies for each step be corroded much faster than in the other locations.
of construction projects. Furthermore, attention must be paid to some specific
locations of bridge parts of elements where water tends
to be stagnant and pooled or the humidity is easily
3.4 Cause of damages for bridge parts and
kept high. Sometimes dewdrops become also harm-
structural members
ful unless they dry out. Figure 8 shows an example of
Figure 7 shows the causes of damage of bridge parts dehumidifier used to keep inside a girder dry (Kaneko
and elements of steel bridges. Described are the 1999). Figure 9 also shows an example of dehumidifier
68
Figure 8. Dry air dehumidifier inside a bridge deck.
Figure 10. Cathodic protection (without external source

power).
protection without external power source (Ishida

2006).
3.7 Other technologies

(1) Technologies for planning and design stages
The assessment is required taking into account
LCC considering the construction site and selec-
Figure 9. Dry air infusion system for cables. tion of structural type.
(2) Inspection technology
(a) Anti-fatigue inspection: In the past, the crack

used for main cables of a suspension bridge (Kitagawa inspection was first done by flesh eyes. After
2001). confirmation of problems such as peeling-off
of painting some detection tests were carried
(1) Improvement of corrosive environment
out. However, this method has many short-
(a) This is to dry up inside girder by removing comings such as cost, labor, and impossible
humidity to prevent dewdrops using either when eye observation is impossible. A wire-
dehumidifiers or ventilators. less detection technology may be applied by
(b) Clean-up technologies: This is to remove the placing IC tags at the place where cracks are
dirt and salt from the surface expected to initiate.
(b) Anti-corrosion inspection: It may be very use-
(2) Cathodic protection methods
ful to find out corrosion at the place where eye
Because of the severe corrosive environment,
observation is impossible such as reinforce-
corrosion protection should be made, especially
ments in concrete.
for the parts just below the M.L.W.L. when severe
local corrosion occurs. For such parts, cathodic
protection is generally applied.
4 EXPECTED TECHNOLOGIES FOR
(a) Cathodic protection without external electric
LONGEVITY
power
(b) Cathodic protection with external electric
4.1 Current attitude of road owners
power
(c) Cathodic protection in the air Depending on the demand performance, road owners
are interested in the following structural maintenance
(3) Other anti-corrosion methods
matters:
The other coating methods include (a) painting,
(b) organic lining, (c) petrolatum lining and (d) inor- 1) Essential maintenance to repair a great damage.
ganic lining. The inorganic linings include metal lin- 2) Preventive maintenance to repair a light damage
ings such as titanium-clad lining (JTS 2000, Nippon 3) Up-grading of wheel loads and earthquake
Steel 2000), stainless steel lining, thermal spraying retrofitting
using zinc, aluminum and aluminum alloys (JAPH 4) Daily maintenance such as cleaning up road sur-
1999). Figure 10 shows an example of cathodic face, drainage pipes.
69
4.2 Several problems to be solved The methodology of investigation is as follows:
There are many problems remain yet to be solved. (1) to collect the information of related books, reports
These may be summarized below: including committee reports, and seminars that are
published in Japan
(1) Difference of management level (demand perfor- (2) to collect references on typical international con-
mance) ferences on the structural health monitoring
Not only the natural environment but also socio- (3) to understand the present trend of technologies
economic environment are to be considered for for structural health monitoring and evaluation of
the determination of lifetime and maintenance performance
management. (4) to classify the existing technologies by summariz-
(2) Limitation of budget (evaluation of LCC) ing the target of application and methodology in
Balancing of cost for maintenance management the form of matrix
by essential management and that by preventive (5) to summarize the prospect of application of struc-
maintenance must be established. tural health monitoring technologies after review-
(3) Effective use of data (data sharing) ing references
This is effective when combined with manage-
ment system and good for accountability.
(4) Succession of technology 6.1.2 Target references of investigation
The education of young engineers and out The references reviewed are shown in Table 5.
sourcing are important.
(5) Effective inspection method (frequency and
methodology) 6.2 Reference survey
Priority in inspection, remote inspection and 6.2.1 Summary of references
structural health monitoring technology are Totally 136 references were reviewed and summarized
expected to be established. by Tables 10 and 11. Although a great number of fig-
ures and tables were contained. They are not shown in
this paper because of the restriction of page numbers.
5 NEEDS FOR STRUCTURAL HEALTH
MONITORING AND REQUIRED
PERFORMANCE Table 5. List of references surveyed.
5.1 Background for survey by sending out

questionnaires to owners of structures
Attempts were made to communicate with several
owners of bridges to find the needs of structural
health monitoring by sending questionnaires to them.
H However, these are omitted again.
6 SURVEY OF STRUCTURAL HEALTH

MONITORING TECHNOLOGIES AT THE
PRESENT TIME
6.1 Summary of investigation

An extensive literature survey were conducted to know
the technologies available at the present time and
future applications.
6.1.1 Purpose and method of investigation

The purpose of investigation is as follows:
(1) to understand the present state of development of

structural health monitoring technology
(2) to construct DB for references on structural health
monitoring
(3) to extract possibilities and problems of application
of structural health monitoring toward longevity of
steel structures
70
REFERENCES
Abe, M., 2008. Activity of bridge maintenance toward the
age of longevity of bridges. Report of Seminar. EXTEC,
84, 6162 (in Japanese).
BMSG., 2004. Guide to management for road bridges. Bridge
Management Study Group, Japan Bridge Engineering
Center, JBEC (in Japanese).
Habel, W.R. et al., 2007. Fiber Optic Sensors For Long-Term
SHM in Civil Engineering Applications, Proc Interna-
tional Conference on Structural Health Monitoring of
Intelligent Infrastructure.
Ishida, Sato, Fujikawa, Sekiguchi and Hashimoto., 2006.
Concept of corrosion protection of steel for D-runway
For reduction of LCC-. 2nd Seminar on report of construc-
tion of D-runway ofTokyo InternationalAirport (Haneda).
8.18.8 (in Japanese).
Figure 11. Number of references with respect to measured JAPH, 1999. Technological Standard and Commentary of
physical quantities and output. Port and Harbor Facilities, Japanese Association of Port
and Harbor, 1 & 2, (in Japanese).
JSSC, 1991. Study on lifetime of steel structures. Report of
Task Committee on Lifetime of Steel Structures, 2930 (in
Japanese).
JTS. Titanium World In pursuit of its unlimited potentiali-
ties and realization of dreams-, 2000. The Japan Titanium
Society, http://www.titan-japan.com, 7, (in Japanese).
Kaneko, Mizoue and Naito,1999. Corrosion protection inside
a box girder by dry air intake New Onomichi Bridge-,
Bridge and Foundation, 33(5), 3134 (in Japanese).
Kato, M., 1983. Recent topics in maintenance. Journal of
JSCE. 68(11): 25 (in Japanese).
Kitagawa, Furuya, Nakamura and Suzumura., 2001. Study on
corrosion protection performance by dry air intake system
for main cables for suspension bridge cables. Proc. JSCE
672(VI50), 145154 (in Japanese).
MLIT, 2002. White paper of the Government of Japan.
MLIT, Ministry of Land, Infrastructure and Transport
Figure 12. Number of references in the matrix form with
and Tourism (http://www.mlit.go.jp/hakusyo/mlit/h14/
respect to measured physical quantities and types of damage.
H14/index.html) (in Japanese)
MLIT, 2008. Suggestions for preventive maintenance of
6.2.2 Representation of matrix roadway bridges. Report of Meeting by experts toward
The result of the survey on sensing technologies and Preventive Maintenance of Roadway Bridges, MLIT, Min-
istry of Land, Infrastructure and Transport and Tourism (in
assessment methods is conveniently represented by a Japanese).
matrix method as shown in Figures 11 and 12. NILIM, 2004. Study on estimation of the lifetime of road
The following observations may be made: bridges, National Institute for Land and Infrastructure
(1) Status of developments of sensing technologies Management, NILIM Report, 223 (in Japanese).
Nippon Steel, 2000. Titanium Clad, Cat. No. TC026, (in
a. Majority of measured physical quantities are Japanese).
strains and accelerations. However, the number PWRI Bridge Laboratory., 1997. Study on replacement of
of instances of stresses, reaction forces, ph and bridges (III), PWRI Report 3512 (in Japanese).
temperature is quite few. Sumitro, S. et.al., 2007. Removable Actual-Stress Sensor,
Proc International Conference on Structural Health Mon-
b. Local deformations and vibrations constitute the itoring of Intelligent Infrastructure.
majority of the diagnosed damages. The number
of instances of direct measurements is quite few
for material deterioration and decrement of rigidity
target damage.
71
Wireless sensing technologies for bridge monitoring and assessment
C.-B. Yun, H. Sohn & H.J. Jung

Korea Advanced Institute of Science and Technology, Korea
B.F. Spencer
University of Illinois at Urbana-Champaign, USA
T. Nagayama
University of Tokyo, Japan, USA
ABSTRACT loggers, computers, and connecting cables, is a big

obstruction. SHM systems generally require coaxial
Recently, there has been increasing need and interest in wires for communication between sensors and the data
adopting wireless sensors to structural health monitor- repository but the installation of coaxial wires in real
ing (SHM) applications. Smart wireless sensors have structures can be very expensive and labor-intensive.
been considered as an alternative tool for convenient For example, it was reported that a SHM system
and economical realization of SHM system. The wire- installed in a tall buildings generally cost in excess
less sensors can provide rich information which SHM of US$5000 per sensing channel (Celebi 2002).
algorithms can utilize to detect, locate, and assess To overcome these problems, there have been ongo-
structural damage caused by progressive deterioration ing efforts to adopt smart wireless sensors to SHM
under operational loadings and severe loading events. applications (Straser et al. 1998, Lynch et al. 2006a).
Many applications of wireless technologies to real Smart wireless sensor is an emerging sensor with the
structures have been made for on-line wireless SHM following essential features: sensing, onboard pro-
of in-service civil infrastructures. This paper presents cessing, wireless data transmission, and power supply
several hardware/software issues and example appli- (Nagayama 2007). When many sensors are imple-
cations of smart wireless sensing technologies. The mented on a SHM system for a sizable civil struc-
first example is smart wireless sensor network, which ture, wireless communication between sensors and
is applied to a cable stayed bridge. The second one data repository seems to be attractive in the aspects
is a multifunctional wireless impedance sensor node, of the cost effectiveness. Dense arrays of low-cost
which is fabricated and applied to a real steel bridge. smart wireless sensors have the potential to improve
The last one is an innovative wireless system for the quality of the SHM dramatically using their
guided wave excitation and sensing based on laser and on-board computational and wireless communication
optoelectronic technologies, which has been experi- capabilities.
mentally demonstrated. This paper presents three wireless sensor and sens-
ing technologies with their hardware/software issues
and applications to monitoring of bridge structures
1 INTRODUCTION as (1) wireless vibration sensor network and applica-
tion to a large scale SHM system for a cable stayed
The modern civil engineering structures are becoming bridge which consists of 70 wireless sensor nodes with
more complex and are expected to be fully functional 420 channels, (2) multifunctional wireless impedance
under severer environmental conditions. Hence, there sensor node and application to local monitoring and
has been a demand to monitor structures in order to damage detection of a steel bridge, and (3) wireless
ensure safety and serviceability to the community. actuation and sensing system using laser and optoelec-
Structural health monitoring (SHM) is a methodology tronic technologies for guided waves based structural
to monitor the performance of a structure and identify health monitoring.
incipient failure conditions, which helps in improv-
ing the safety and life span of the structure. The SHM
system often offers an opportunity to reduce the cost 2 WIRELESS VIBRATION SENSOR NETWORK
for the maintenance, repair, and retrofit throughout FOR BRIDGE MONITORING
the life-cycle of the structure.
However, in the conventional SHM system, the To address critical issues on wireless smart vibration
expensive cost for purchase and installation of the sensors and to promote related R&D activities, such
SHM system components, such as sensors, data as (1) data synchronization and recovery of missing
72
speed in horizontal and vertical wind directions are
measured through analog input interface connectors
on the SHM-W board. The wind data can be acquired
precisely synchronized with the bridge acceleration
data from the SHM-A board. The IBB2400CA battery
board powers the Imote2 using three 1.5-volt batter-
ies. To investigate the feasibility of sustainable energy
harvesting, solar panels and rechargeable batteries
have been adapted on several sensor nodes installed
at the locations hardly reachable.
2.2 Software of wireless sensor network

Figure 1. The 1st (right) and 2nd (left) Jindo Bridges.
An open-source middleware services toolsuite, which
interacts between the target SHM application and oper-
ating system of hardware, were developed to help the
civil engineers easily code into the wireless smart sen-
sor nodes. The toolsuite contains basic middleware to
provide high-quality sensor data and to transfer the
data reliably to the base station via wireless commu-
nication as well as a library of numerical algorithms.
The toolsuite components are categorized into founda-
tion services, tools and utilities, application services,
and continuous and autonomous monitoring services.
Figure 2. SHM-A sensor board (Rice et al. 2010). (Mechitov et al. 2004, Nagayama & Spencer 2007,
Rice et al. 2008).
To operate large-scale wireless smart sensor net-
work in this study, 3 problems are seriously considered:
data, (2) on-board and decentralized processing, (3)
1) energy saving software architecture, 2) data inunda-
autonomous operation, (4) power management and
tion, and 3) continuous and autonomous operation. To
energy harvesting, and (5) environmental hardening,
save batteries of large number of nodes, the network
an international test bed was developed on a cable-
is allowed to sleep default and wakes up periodically
stayed bridge (the 2nd Jindo Bridge shown in Figure 1)
for a short time to listen to broadcasted commands.
in Korea through a trilateral research collaboration
To prevent data inundation from large number of
among Korea (KAIST), the US (University of Illi-
sensor nodes, preset sentry nodes wake up at a pre-
nois at Urbana-Champaign), and Japan (University
set interval and measure a short period of data to
of Tokyo).
make a decision on the measurement based on preset
threshold. Autonomous SHM network management
software is utilized for the base station to enable the
2.1 Hardware of wireless smart automatic, continuous monitoring with reduced power
sensor network consumption.
The key hardware components of a wireless smart
sensor node are an Imote2 (Crossbow 2007), a sen- 2.3 Deployment of SHM system on
sor board, and a battery board. An Imote2 can be Jindo Bridge
interfaced with a SHM-A multi-scale sensor board to
measure 3-axis acceleration as well as temperature, The developed hardware and software framework has
humidity, and illuminance, or a SHM-W sensor board been deployed on the 2nd Jindo Bridge to realize SHM
to measure wind speed and direction interfaced with system using wireless sensor network. The network
3-axis anemometer.The SHM-A sensor board contains was divided into two sub-networks with different radio
a 3-axis accelerometer, thermometer, hygrometer, and channels: one on the Jindo side and the other on the
luxmeter, whose signal is digitized by the embedded Haenam side as shown in Figure 3. A total of 70 nodes
signal conditioning chip with user-selectable sampling are deployed on the bridge. They contain SHM-A
rates and programmable digital filters as shown in sensor boards to measure 3-axis acceleration mainly,
Figure 2. This board allows to measure data from except one node which contains SHM-W board inter-
other types of sensors such as anemometers and strain faced with an anemometer to measure wind speed and
gages. Four sampling frequencies (10, 25, 50, 100 Hz) direction. The Jindo subnetwork consists of 33 nodes
have been pre-programmed on the SHM-A board for with 22 nodes on the deck, 3 nodes on the pylon, and 8
bridge monitoring applications. The SHM-W board nodes on the cables. The Haenam sub-network consists
was developed by modifying the SHM-A board to of 37 nodes with 26 nodes on the deck, 3 nodes on the
have three external 05 V input channels to be inter- pylon, and 7 nodes on the cables. The sensor nodes
faced with a 3-D ultrasonic anemometer. The wind were enclosed in water-tight plastic enclosures for
73
Figure 3. Locations of sensor nodes deployed on the 2nd Jindo Bridge.
Figure 4. Various types of wireless sensor nodes installed on the 2nd Jindo Bridge.
protection from harsh surroundings. The deck/pylon

nodes are mounted using one-directional magnets
attached to the bottom of enclosure, and cable nodes
are mounted on an aluminum plate with round inter-
face to fit the cables. An anemometer is installed on a
5 m tall at the center of deck to prevent any interrup-
tion of the bridge. Most of the nodes are powered by
3 D-cell batteries, while 8 nodes (2 nodes on the tops
of both pylons, and 5 cable nodes and 1 deck node
of Haenam sub-network) are powered by solar panels
with rechargeable batteries.
Each sub-network is controlled by a correspond-
ing base station located at the tops of the concrete
piers supporting the steel pylons of the 1st Jindo
Bridge. The locations of base stations are selected
to secure consistent line-of-sight communication with
the sensor nodes. The base station is composed of an
industrial personal computer (PC), an uninterrupted
power supply (UPS) backup, a gateway node, and
ADSL internet modem, and it is covered by an envi-
ronmentally hardened enclosure. The parameters of
the wireless sensor network are set after a series of
laboratory and preliminary in-field tests.
The performance of the system has been evalu-
ated in terms of hardware durability, software stability,
power consumption, and energy harvesting options, as
well as the accuracy of the identified modal properties
by an output-only modal identification method shown
in Figure 5. Figure 5. Identified mode shapes of the 2nd Jindo Bridge.
74
3 WIRELESS VIBRATION SENSOR NETWORK
FOR BRIDGE MONITORING
3.1 Electromechanical impedance-based SHM

In this paper, multifunctional wireless impedance
sensor nodes are presented for low-cost and low-
power excitation/sensing, structural damage detection
using embedded algorithms, and energy scavenging.
The electromechanical impedance-based SHM tech-
niques are embedded in the nodes. Recent advances Figure 6. Proposed wireless impedance sensor node.
in online SHM, including low-cost impedance mea-
suring chip, on-board computing, and radio-frequency Table 1. Features of the proposed wireless impedance
(RF) telemetry, have improved the accessibility of the sensor node.
impedance method for in-field applications (Grisso
and Inman, 2005, Mascarenas et al., 2007 & 2009, Frequency Range 1 100 kHz
Overly et al. 2007). Based on those prior researches, Frequency Resolution >1 Hz
the present impedance-based wireless SHM node is Impedance Range 1 k 1M
Temperature Range 20 120 C
developed by focusing on the following three objec- Temperature Resolution >0.03 C
tives: (1) to develop in-field adjustable impedance- Operating Frequency 2.4 GHz IEEE 802.15.4
based wireless sensor nodes with on-board algorithms Outdoor Transmission 150 m
for structural diagnosis, (2) to incorporate the energy Range (2dBi Dipole Antenna)
scavenging system for maintenance-free wireless 5 V AC-DC Adapter;
sensor node, and (3) to investigate the feasibility Commercial batteries
of the impedance-based SHM system to real bridge Power Supply Options (3.67.2 V);
structures. Ni-MH rechargeable batteries
with
Solar Panels
3.2 Wireless impedance sensor nodes 150 100 70 (mm);
Feature 310 (g)
The proposed wireless sensor node is composed of
four functional subsystems: sensing interface, compu-
tational core, wireless transceiver, and power supply.
(1) The sensing interface includes an interface to sensor node developed in this study and the features
which a piezoelectric sensor and a temperature sensor are described in Table 1.
can be connected, and an impedance chip (AD5933) For autonomous SHM using wireless sensor nodes,
for exciting a piezoelectric sensor and measuring it is strongly required to construct the embedded data
the impedance signals. (2) The computational core analysis system. In the proposed sensor node, mul-
consists of a microcontroller and a serial flash mem- tifunctional algorithms are implemented for temper-
ory for computational tasks and system operations ature/power measurement, impedance measurement,
with various embedded algorithms. (3) The wireless and analysis engine for both structural damage detec-
transceiver is an integral component of the wire- tion and sensor self-diagnosis. Here, two algorithms
less system, which is composed of a RF transceiver are embedded on the microcontroller for the struc-
(CC2420), a balun transformer, and an antenna to com- tural status monitoring: the RMSD metric and the
municate with a base station and/or other wireless temperature compensated CC metric calculated by
sensor nodes and to broadcast the structural condi- EFS method. Sensor self-diagnosis is simply carried
tion. (4) The sensor node can be operated by one of out calculating the slope of the imaginary part of
three type power supply systems: 5 V AC-plug DC admittance.
adapter, 3.67.2 V battery, or 5 V solar power system.
The power can be monitored on the microcontroller
3.3 Applications
using a general ADC, which transforms the analog
signals acquired from batteries to the digital signals. A field test has been performed to detect damages
For stable power supply to the sensor node during including loosened bolts and notches on the Ramp-G
operations, LDO (Low-dropout regulator) is mounted bridge in Incheon, Korea. The bridge is a traditional
for providing a fixed 3.3 V reference output to the steel box girder bridge with a reinforced concrete deck
sensor node. Solar power system for energy harvest- as in Figure 7. One sensor node was installed near
ing consists of single crystalline silicon solar cells the bolted joint of the outer girder and connected to a
(120 60 mm2 ) to generate the maximum power for PZT patch of 50 50 0.5 mm3 and a NTC thermis-
its size, two AA Ni-MH re-chargeable batteries to tor surface-mounted at a distance of 200 mm from the
stand high temperature and overcharging under sun- bolted joint using epoxy glue. The maximum temper-
light, and a step-up DC/DC solar controller to protect ature variation obtained from a NTC thermistor was
the appliances and the batteries with over discharge 15 C during the test of 2 days, and the voltage drop
prevention circuit. Figure 6 shows the impedance of battery was 0.004 from 3.476 to 3.472. The E/M
75
Figure 9. Temperature-compensated CC indices for multi-
Figure 7. Experimental setup for impedance-based SHM ple damage detection on a bridge.
on Ramp-G bridge.
noted that the condition of hand-tightened Bolt #1 is
Table 2. Damage scenarios for detecting multiple damages not same with original condition (baseline). Then, the
on a bridge girder. CC value decreases as the damage severity increases
due to loosened bolts and artificial notches. Here, the
Damage Damage CC metric for Cases 5 and 6 have similar values,
Case Description Case Description which might be because Bolt #2, which got loosened
in Case 6, is located fairly far from the sensor. Through
1 No damages 5 Additionally loosened
in-field test, it is expected that the actual damages can
Bolt #1 (2 Turns)
2 Loosened Bolt #1 6 Additionally loosened be effectively identified by the developed sensor node
(2 Turns) Bolt #1 (2 Turns) and on-board processed damage indices.
3 Hand-retightened 7 Additionally Notch #2
Bolt #1
4 Notch #1
4 WIRELESS ACTUATION AND SENSING
FOR GUIDED WAVES BASED SHM
4.1 Guided wave-based SHM

Guided wave based SHM techniques have attracted
much attentions, because they are not only sensitive
to small defects but also capable to cover over a long
distance in plate and pipe like structures. A number
of studies have demonstrated the potential of guided
wave based SHM (Raghavan et al. 2007, Giurgiutiu
2008).
Piezoelectric lead zirconate titanate (PZT) elements
could be a good candidate for such online applications
due to its small size, easy installation, low cost, non-
intrusive nature, and wide frequency response range.
Figure 8. Measured impedance signatures at various dam- PZT transducers typically require wires to supply the
age cases. power necessary for generation of guided waves and to
transmit the sensed data, but the installation of wires
in real structures can be expensive and labor-intensive.
impedances were measured at a frequency range of To overcome these problems, there are on-going efforts
4550 kHz for the baseline. Here, it was assumed that to integrate a PZT transducer with a wireless sensor
there are not any damages within the sensing range of unit (Lynch et al. 2006). One of major challenges
the attached sensor. Tests were carried out for seven in such wireless systems is to secure power neces-
damage scenarios described in Table 2. The first dam- sary to operate the wireless sensors. However, because
age was simulated by two-turn loosened Bolt #1, and guided wave based active sensing devices demand rel-
then Bolt #1 was hand-tightened. Cracks and multiple atively high electric power compared to conventional
damages were induced in a sequence. Example cases passive sensors such as accelerometers and strain
of the measured impedance signals at different dam- gauges (Yeatman 2009), existing battery technologies
age conditions were compared with the baseline data in may not be suitable for long-term operation of active
Figure 8, from which significant variations is observed sensing devices.
in the measurements. Figure 9 shows temperature-
compensated CC values, which were computed on
4.2 Wireless actuation for guided wave based SHM
the on-board microcontroller and wirelessly transmit-
ted to a base station. It is found that the CC value The ultimate goal of the research is to develop an opti-
decreases when Bolt #1 gets loosened, and it is almost cal system for guided wave generation and sensing.
recovered as the bolt is hand-tightened. It should be Figure 10 shows an overall schematic of the proposed
76
Figure 12. An overall experimental setup of the proposed
Figure 10. An overall schematic for optics based wireless
wireless guided wave excitation system.
guided wave generation and sensing.
remotely inspect the integrity and performance of large

distributed infrastructure systems such as bridges.
4.3 Experimental characterization

The feasibility of the proposed method for wire-
less guided wave generation has been experimentally
demonstrated. The overall test configuration and the
test specimen are shown in Figure 12. The system
was composed of a laser diode as a power source, an
EOM with an arbitrary waveform generator (AWG)
Figure 11. Integration of optics-based active sensing tech-
for intensity modulation of the laser, an optical ampli-
nology and unmanned autonomous inspection robot. fier, a collimator, a photodiode, a transformer, and PZT
transducers attached on an aluminum specimen.
The output power of the laser diode used in this
wireless guided wave excitation and sensing system. It experiment was 10 mW and controlled by the laser
takes advantage of optical techniques for both guided driver. Using the AWG, a toneburst signal with 2 V
wave generation and sensing. First, a laser wirelessly peak-to-peak voltage was generated at a driving fre-
transmits a generated waveform to a PZT transducer quency of 150 kHz and exerted to the EOM for
node that consists of a photodiode, a transformer intensity modulation of the laser. This modulated laser
and a PZT transducer. Then, the photodiode converts power was amplified up to 80 mW by the optical
the light into an electric signal, and the transformer amplifier and transmitted by optical fiber to the col-
increases the voltage level of the created electric sig- limator. The collimated laser was emitted into air and
nal. The electric input signal excites the PZT attached aimed at the integrated photodiode and transformer
on a structure, and the excited PZT consequently for conversion to an electric signal and amplification
creates guided waves within the structure. Next, the of voltage level as shown in Figure 13(a). Then, PZT A
corresponding reflected waves are measured by the generated guided waves, and the responses were mea-
identical PZT and re-converted into a laser and trans- sured at PZT B. PZT A was also excited by using
mitted back to another photodiode located in the data conventional wire connection to the AWG, and the
acquisition unit for diagnosis. corresponding responses were measured at PZT B for
Because power is remotely transmitted to the PZT the purpose of comparison as shown in Figure 13(b).
transducer, no complex electronic components are The results confirm that the laser-based guided wave
necessary at the PZT node, making it possible to generation technique exhibits the reasonable perfor-
develop a self-contained, rugged, and non-intrusive mance by comparing the conventional wiring guided
PZT node. The transducer itself can be entirely pas- wave generation method.
sive and consequently will have a long lifetime when it
has rugged packaging. In addition, any arbitrary wave-
form can be generated by a laser using an electro-optic 5 CONCLUDING REMARKS
modulator (EOM).
This paper mainly focuses on the excitation aspect This paper discusses hardware and software issues
of the overall system. A further research for wireless on smart wireless sensor and sensing technologies,
guided wave sensing is currently being developed by and presents several recent developments and applica-
the authors using other optoelectronic devices, such tions to bridge monitoring. Firstly, the deployment and
as photovoltaic panels and laser diodes (Park et al. evaluation of a state-of-the-art wireless smart sensor
2010). Finally, as shown in Figure 11, it is envi- network on a cable-stayed bridge are reported. Then
sioned that the develop technology can be integrated multifunctional wireless impedance sensor nodes are
with autonomous moving agents such as robots to presented for low-cost and low-power excitation/
77
REFERENCES
Celebi, M. (2002) Seismic instrumentation of buildings
(with emphasis of federal buildings), Technical report
No. 0-7460-68170.
Crossbow, Inc. (2007). Imote2 Hardware Reference Manual.
Revision A, PN: 7430-0409-01, http://www.xbow.com.
Giurgiutiu, V. (2008) Structural Health Monitoring with
Piezoelectric WaferActive Sensors, Elsevier Inc., London.
Grisso B.L. & Inman D.J. (2005) Developing an autonomous
on-orbit impedance-based SHM system for thermal pro-
tection systems, Proceedings of 5th Intl Workshop on
Structural Health Monitoring, Stanford, CA, September.
Lee, H.S., Park, H.J., Sohn, H. & Kwon, I.B. (2009) A
Hybrid PZT/FBG guided wave generation and sensing
system using a single laser source, International con-
ference on computational design in engineering, Seoul,
Korea, November.
Lynch, J.P. & Loh, K.J. (2006a) A summary review of wireless
sensors and sensor networks for structural health moni-
toring, The Shock and Vibration Digest, Vol. 38(2), pp.
91128.
Mascarenas D.L., Todd M.D., Park G. & Farrar C.R. (2007)
Development of an impedance-based wireless sensor node
for structural health monitoring, Smart Materials and
Structures, Vol. 16, pp. 21372145.
Figure 13. Experimental results. Mascarenas, D.L., Park, G., Farinholt, K., Todd, M.D. &
Farrar, C.R. (2009)A Low-Power Wireless Sensing Device
for Remote Inspection of Bolted Joints, Journal of
sensing, temperature/power monitoring, structural Aerospace Engineering, Vol. 223, pp. 565575.
damage detection using embedded algorithms, and Mechitov, K. Kim, W., Agha, G., & Nagayama, T. (2004)
energy scavenging. Lately, an innovative wireless sys- High-frequency distributed sensing for structure moni-
tem is proposed for guided wave excitation and sensing toring, Proceedings of 1st Int. Workshop on Networked
based on laser and optoelectronic technologies. The Sensing Systems.
Nagayama, T. & Spencer Jr., B.F. (2007) Structural Health
feasibility has been experimentally demonstrated.
Monitoring using Smart Sensors, NSEL Report Series 1,
The wireless sensor and sensing technology is the UIUC. USA. (http://hdl.handle.net/2142/3521).
enabler of realization of cost-effective monitoring Overly, T.G., Park, G., Farrar, C.R. & Allemang, R.J. (2007)
system. The system-on-chip data acquisition system Compact hardware development for structural health
without tethering work lowers the cost of sensors monitoring and sensor diagnostics using admittance mea-
themselves as well as of installation, maintenance, and surements, Proceedings of IMAC-XXV, USA.
replacement. The wireless data transmission removes Park, H.J., Sohn, H., Yun, C.B. & Chung, J. (2010) Develop-
the cost related to the cables, and therefore, pro- ment of a wireless power transmission system for guided
motes the free clustering of sensor networks. The wave generation and sensing via a laser, SPIE vol. 7647
Sensors and Smart Structures Technologies for Civil,
on-board computation enables the quick and easy
Mechanical, and Aerospace Systems, USA, submitted.
on-line monitoring without concerns on data inun- Raghavan, A. & Cesnik, E.S. (2007) Review of Guided-wave
dation. However, there still remain some limitations Structural Health Monitoring, The Shock and Vibration
in wireless communication and energy supply for Digest, Vol. 39(2), pp. 91114.
wireless sensor networks. Wireless communication Rice, J.A., Mechitov, K.A., Spencer Jr., B.F., & Agha, G.
inevitably requires quick and reliable data communi- (2008) A Service-Oriented Architecture for Structural
cation protocol, recovery of missing data, and time Health Monitoring using Smart Sensors. Proc. 14th World
synchronization among sensors. To overcome current Conference on Earthquake Engineering, Beijing, China.
capacity of battery technology, which may not be Straser, E.G. & Kiremidjian, A.S. (1998) A modular, wireless
damage monitoring system for structures Report No. 128,
enough to operate wireless sensor networks in mid-
John A, Blume Earthquake Engineering Center, Stanford
or long-term, power management as well as power Univ., CA.
harvesting from ambient vibration, wind, and temper- Yeatman, E.M. (2009) Energy harvesting-small scale energy
ature, are now actively under development. Multidis- production from ambient source, SPIE vol. 7288 Active
ciplinary researches as well as education can facilitate and Passive Smart Structures and Integrated Systems,
the rapid development as well as application. USA.
78
Mini-Symposia
MS1: Futuristic bridge maintenance technologies
Organizers: C.-B. Yun & B.F. Spencer Jr.
Passive sensors for monitoring corrosion in concrete bridges
A. Abu Yousef & S.L. Wood

Department of Civil, Architectural and Environmental Engineering, University of Texas, Austin, TX, USA
P. Pasupathy & D.P. Neikirk

Department of Electrical Engineering, University of Texas, Austin, TX, USA
ABSTRACT The lack of physical contact between the corroding

element and the resonant circuit in the new arrange-
A class of low-cost wireless sensor has been devel- ment prevents corrosion products from reaching the
oped at the University of Texas at Austin to monitor circuit components. This configuration was selected to
corrosion in concrete bridges. The sensors are pow- address a durability concern in the original threshold
ered and interrogated in a wireless manner using sensor (Puryear 2007).
inductive coupling between an external reader coil Two alternatives for the corroding element were
and the embedded sensor (Figure 1). The sensors are proposed and a number of parametric studies were con-
designed to be placed in concrete during construction ducted in air to understand the behavior of the analog
and interrogated periodically over the service life of sensor. The sensor was also subjected to wet/dry cycles
the bridge. in salt water to assess the change response as corro-
Previous research efforts have focused on detecting sion develops on the sacrificial element. The results of
the onset of corrosion using a threshold sensor. The these preliminary tests are discussed in this paper.
design employed a sacrificial steel wire that is physi-
cally connected to a resonant circuit. The corrosion of
the wire turns the embedded sensor into an open circuit REFERENCES
and alters its phase response. So, when the threshold
sensor is interrogated two limit states can be easily Dickerson, N.P.,Andringa, M.M., Puryear, J.M., Wood, S.L., &
Neikirk, D.P. (2006). Wireless threshold sensors for detect-
recognized (Corrosion / No Corrosion).
ing corrosion in reinforced concrete structures. Smart
In this paper, a new generation of sensors is pre- structures and materials 2006: sensors and smart struc-
sented that can be used to provide analog output tures technologies for civil, mechanical, and aerospace
describing the level of corrosion within the concrete. systems, 6174. San Diego, CA: International Society for
The analog sensor incorporates a sacrificial corrod- Optical Engineering.
ing element that is placed entirely outside the resonant Pasupathy, P., Munukutla, S., Neikirk, D.P., & Wood, S.L.
circuit and interacts with it only by inductive coupling. (2009). Verstile wireless sacrificial transducers for elec-
tronic structural surveillance sensors. IEEE Sensors 2009.
Christchurch, New Zealand: Institute of Electrical and
Electronics Engineers.
Puryear, J.M.H., Dickerson, N.P., Wood, S.L., & Neikirk, D.P.
(2008). Low-cost wireless sensors for concrete bridges.
NDE/NDT for highways and bridges: structural mate-
rials technology. Oakland, CA: American Society for
Nondestructive Testing.
Wood, S.L. & Neikirk, D.P. (2009). Passive sensors for
infrastructure monitoring. Sensors and smart structures
technologies for civil, mechanical, and aerospace systems,
7292. San Diego, CA: International Society for Optical
Figure 1. Wireless interrogation of embedded passive sen- Engineering.
sor using external reader coil.
81
Coupled resonant coil sensors with increased interrogation distance
S. Bhadra, E. Thompson, B. Kordi, G.E. Bridges & D.J. Thomson

Department of Electrical and Computer Engineering, University of Manitoba, Winnipeg, Manitoba, Canada
ABSTRACT the sensor to the interrogator. A delay is introduced

when the system switches from the transmit mode to
Wireless sensors are being explored for many struc- receive mode to eliminate the transient response of the
tural health monitoring applications. Passive wireless interrogator coil.
sensors have the advantage of not requiring a power Using this method a test sensor coils resonant peak
source on the sensor, as the energy is provided by position was determined within 30 parts per million
the interrogation unit. Inductively coupled coil sen- at a greater interrogation distance (20 cm) than that
sor is a class of passive wireless sensor. They operate of the impedance measurement method (15 cm). This
by using a resonant coil coupled to a measurand that accuracy was much higher than some passive wire-
shifts the sensor coils resonant frequency. Coupled less sensors employing an impedance measurement
coil sensors are found in many applications e.g., cor- technique (e.g., 6 cm interrogation distance with less
rosion detection (Andringa et al. 2005), monitoring than 1% measurement error (Ong et al. 2008), 0.6 cm
water content in civil engineering materials (Ong et al. interrogation distance with the peak position within
2008), strain measurement (Butler et al. 2002), and 7740 parts per million (Butler et al. 2002), 8 cm inter-
moisture measurement (Hapster et al. 2002). Most rogation distance with peak position within 1714 parts
commonly measurement of the sensor coils resonance per million (Ong et al. 2002)). For a relative humidity
is done through perturbations to the interrogator coils sensor coil, this method was able to detect changes in
impedance versus frequency.This technique is affected relative humidity with less than 2%.
by the response of the interrogator coil. As the inter-
rogation distance increases, the response of the sensor
coil decreased rapidly and only the interrogator coil REFERENCES
self impedance dominates, which is sensitive to the
background environment. Thus this interrogation tech- Andringa, M.M., Neikrik, D.P., Dickerson, N.P. & Wood, S.L.
2005. Unpowered wireless corrosion sensor for steel
nique requires a small interrogation distance between
reinforced concrete. IEEE Sensors 2005: 155158.
the sensor and the interrogator. Butler, J.C., Vogliotti, A.J., Verdi, F.W. & Walsh, S.M. 2002.
In this paper, we present a new interrogation method Wireless, passive, resonant-circuit, inductively coupled,
which makes use of time-domain gating to elimi- inductive strain sensor. Sensors and Actuators A, 102(1):
nate the response from the interrogator coil and the 6166.
exciting signal. This results in a measurement dom- Hapster, T.J., Stark, B. & Najafi, K. 2002. A passive wire-
inated by the response of the sensor coil, enabling less integrated humidity sensor. Sensors and Actuators A,
detection of the sensors resonant frequency far more 95(23): 100107.
precisely at an extended separation distance. The inter- Ong, J.B., You, Z., Mills-Beale, J., Tan, E.L., Pereles, B.D. &
Ong, K.G. 2008. A wireless, passive embedded sensor for
rogator first energizes the sensor through inductive
real time monitoring of water content in civil engineering
coupling between the interrogator coil and the sen- materials. IEEE Sensors Journal, 8(12): 20532058.
sor coil. This is the transmit mode. After enough time Ong, K.G., Puckett, L.G., Sharma, B.V., Loiselle, M.,
has elapsed to allow the induced energy in the sensor Grimes, C.A. & Bachas, L.G. 2002. Wireless, passive
coil to reach equilibrium, the system switches back to resonant-circuit for monitoring food quality. Proceedings
the receive mode where energy is coupled back from of SPIE, 4575: 150159.
82
Concentration-dependent piezoelectricity and strain sensitivity of ZnO

nanoparticle-polymeric thin films
D. Chang
Department of Chemical Engineering & Materials Science, University of California, Davis, Davis, CA, USA
K.J. Loh
Department of Civil & Environmental Engineering, University of California, Davis, Davis, CA, USA
ABSTRACT
Recently, zinc oxide (ZnO) nanomaterials (such as

ZnO nanoparticles and nanowires) have been highly
regarded as potential candidates for the design of
next-generation piezoelectric nanocomposites. Previ-
ous studies have demonstrated that ZnO nanoparticle
(NP)-polyelectrolyte (PE) thin films exhibit piezoelec-
tricity and generate electric potential under applied
strain. Thus, unlike traditional strain sensors (e.g.
strain gages, MEMS strain sensors, piezoresistive
materials, among others), ZnO-NP-PE thin films are
self-sensing materials that do not require a constant
power source for operation. As such, this novel sen-
sor is low cost, of small form factor, and can be
easily embedded within various types of engineering
structural components. Figure 1. Piezoelectric response of ZnO-PSS-PVA films.
In this study, a versatile thermal evaporation and
annealing procedure is proposed for fabricating piezo- generated voltages have also been validated for films
electric zinc oxide-based thin film strain sensors. fabricated with a weight fraction of less than 33%
Here, ZnO NPs are dispersed in poly(sodium 4- to 37%. For films evaporated with lower ZnO weight
styrene sulfonate) (PSS) solutions and mixed with fractions, piezoelectricity is still observed, but the per-
poly(vinyl alcohol) (PVA) for film fabrication. The formance is dramatically lower than those fabricated
nanoparticle and PSS concentrations are varied for with 35% weight fraction zinc oxide NPs. On the
optimizing thin film piezoelectric/self-sensing perfor- other hand, greater ZnO weight fractions resulted in
mance. Upon film fabrication, the piezoelectricity of films exhibiting high noise floors and no piezoelec-
12 unique as-fabricated films (characterized by dif- tric response to applied strains. It is hypothesized that
ferent weight fractions) is experimentally investigated high concentrations of zinc oxide induce clumping
by epoxy-mounting them onto thin poly(vinyl chlo- and agglomeration of nanoparticles that diminish their
ride) cantilevered beams; a traditional metal-foil strain piezoelectric performance.
gage is also mounted adjacent to the films for measur- Second, the piezoelectric thin films normalized
ing the induced strains during testing. Then, an initial generated voltages are directly compared with mea-
displacement is applied to the free-end of the can- sured strains. Specifically, since piezoelectric materi-
tilevered beam and released to facilitate free vibration als respond to dynamic strain, the voltage time history
response of the beam. The potential generated by each plots are overlaid with the rate of change of strain.
film is measured and compared against one another From these results, it can be confirmed that there is
to identify the optimal ZnO concentration that yields good correlation between the normalized generated
the highest bulk film piezoelectricity and self-sensing voltages and the rate of change of applied strain. Pre-
performance. liminary results indicate that the strain rate sensitivity
Upon conducting the experimental study, first, of the proposed nanocomposites varies with the abso-
piezoelectricity is validated for several of the zinc lute amount of nanoparticles embedded within the
oxide-based films (Fig. 1). The maximum normalized thin film.
83
Prediction of displacement response of a suspension bridge using

FBG strain sensors
S.J. Chang & N.S. Kim

Pusan National University, Busan, South Korea
H.K. Kim
Seoul National University, Seoul, South Korea
ABSTRACT 80
Displacement (mm)
Fromdisplace mentsensor
In thisstudy
40 From GPS
Predictions of displacement as an important factor in
evaluating the robustness of large structures become 0
the basis to determine the decrease of structural per-
40
formance and the degree of aging in general. It is
however well known that it is not easy to measure the 80
0 20 40 60 80
displace response of large structure like suspension Time(sec)
bridges due to the absence of measurement tech-
niques despite the importance of measurements in Figure 1. Comparison of estimated displacement response
the displacement response. In this study, a method and measured displacement response.
estimating displacement responses is proposed using
a mode decomposition technique. This is a method
Table 1. Comparison of estimated maximum displacement
estimating the total displacement response combined and measured maximum displacement
with the each displacement response about the major
mode of the structure and the quasi-static displacement From displacement
responses. In this study From GPS sensor
Figure 1 shows that the estimated displacement
response and the displacement response measured are LC1 35.6 mm 37 mm 31.98 mm
identical. The outcome of comparison of the biggest LC2 32.5 mm 50 mm 34.03 mm
amount of deflection of the inferred displacement LC3 60.6 mm 63 mm 60.68 mm
response and the displacement response measured is
seen in Table 1. As shown in Table 1, it was identified
100
that the inferred displacement response by means of
50
mode decomposition technique and the displacement
Displacement (mm)
0
response measured is similar. As shown in the Fig- 50
100
ure 2 was estimated that there is a similarity between 150
Using 16 FBG sensors
Using 8 FBG sensors
measured displacement response and that with the use 200 Using 6 FBG sensors
Using 4 FBG sensors
of 6 sensors. But when 4 sensors was used, it was 250
From displacement sensor
300
observed that there is a big error. Thus when estimat- 0 20 40 60 80
ing the displacement response of simple span girder, it Time(sec)
is expected that the reliable inference of displacement
response can be made with the use of 6 censors. Figure 2. Comparison of estimated displacement responses.
When using mode decomposition technique, it is
possible to infer the displacement response while using REFERENCE
strain signals at a smaller number of points and to infer
the displacement response at all of points. In addition Chang, S.J. & Kim, N.S. 2008. Estimation of Displacement
to this, reliable measurement of strain signal in the Response from the Measured Dynamic Strain Signals
large structures such as suspension bridge can be made Using Mode Decomposition Technique, Journal of the
with the use of FBG sensor with which multiple mea- Korean Society of Civil Engineers. 28(4-A):507-515.
surement is possible, generating little electric noise.
84
Structural health monitoring of a cable-stayed bridge using

acceleration data via wireless smart sensor network
S. Cho, J. Park, H.-J. Jung & C.-B. Yun

KAIST, Daejeon, Korea
S. Jang, H. Jo & B.F. Spencer Jr.

University of Illinois at Urbana-Champaign, Urbana, IL, USA
T. Nagayama
University of Tokyo, Tokyo, Japan
J.W. Seo
Hyundai Institute of Construction Technology, Yongin, Korea
ABSTRACT
There have been increasing interests in the wireless

smart sensors in the field of structural health monitor-
ing (SHM) of civil infrastructures due to its wireless
communication as well as embedded computation
capability. The SHM system based on the wireless
smart sensor network (WSSN) may alter the conven-
tional tethered SHM system which is suffered from the
costly cabling work between sensor and data repos- Figure 1. The 1st (right) and 2nd (left) Jindo Bridges.
itory and inundation of raw data which may not be
informative. Based on the advantages of wireless sen- Output-only modal identification (ID) is carried out
sors, several researches to monitor long-span bridge to extract the modal properties of the bridge from the
structures using wireless sensors have been reported ambient acceleration data of the deck and the pylons.
(Weng et al. 2008, Pakzad 2009). The extracted modal properties from the output-only
The test-bed of the SHM system using state-of-the- modal ID method are validated by comparing with
art WSSN technology has been constructed on a cable- those from the wired monitoring system and the FE
stayed bridge (The 2nd Jindo Bridge shown in figure analysis. Tension forces are also estimated on 10 cables
1) in Korea (Cho et al. 2009). A total of 70 wireless with large tension forces using 3-axis ambient accel-
smart sensors are installed with high spatial density eration data measured on the cables by a vibration
on the bridge, facilitating measurements of 3-axes of method. The estimated tension forces are compared
acceleration underneath of the deck, on two pylons, with the tension forces obtained from the previous reg-
and on the cables. The measurement has been carried ular inspection. Based on the results, discussions are
out in two sub-networks during the past 4 months using made on the monitoring performance using the WSSN.
an autonomous monitoring system. The test-bed is to
provide important insight into the opportunities and
REFERENCES
challenges for WSSN technology on long-term moni-
toring of large civil infrastructures. This effort is a part Weng, J-H., Loh, C-H., Lynch, J-P., Lu, K-C., Lin, P-Y., &
of trilateral collaboration between Korea (KAIST), the Wang, Y. 2008. Output-only modal identification of a
US (University of Illinois at Urbana-Champaign), and cable-stayed bridge using wireless monitoring systems.
Japan (University of Tokyo). The details of constructed Engineering Structures, 30: 18201830.
test-bed can be referred to Cho et al. (2009). Pakzed, S.N., Fenves, G.L., Kim. S., & Culler, D.E. 2008.
This study focuses on the SHM technique using the Design and implementation of scalable wireless sensor
collected acceleration data at the test-bed. As prior network for structural monitoring, Journal of Infrastruc-
ture Systems, 14(1): 89101.
works, a finite element (FE) model is constructed
Cho, S., Jang, S.A., Jo, H., Park, J., Jung, H-J., Yun, C-B.,
based on an in-depth study of the detailed drawings, Spencer Jr., B.F., Nagamaya, T., & Seo, J. 2009. Cable-
and the acceleration data from the existing wired mon- stayed bridge test-bed for long-term structural health
itoring system is analyzed to be used as a reference monitoring using smart wireless sensor network. The
data of WSSN performance. The acceleration data col- Proceedings of the 1st International Conference on Com-
lected from the current WSSN are analyzed offline. putational Design in Engineering. Seoul: COSEIK.
85
Smart wireless tension force monitoring system for stay cables
S. Cho & C.-B. Yun

KAIST, Daejeon, Korea
J.P. Lynch
University of Michigan, Ann Arbor, MI, USA
ABSTRACT The system is validated from the laboratory test

using scale-down cable model with a variety of cable
Tension force of a bridge cable is a structural para- tension and sag conditions. 18 test cases were car-
meter of interest for cable-stayed bridges in which the ried out on the cable with various tension forces with
cable is a primary load carrying member. It should be corresponding sag conditions. The results of the exper-
monitored regularly during its service life to ensure imental study are summarized as follows: (1) Consis-
the bridge is in safe operating condition without over- tent results can be obtained for the first three natural
loading. Though tension measurement devices, such as frequencies regardless of the sensor location along
load cells, hydraulic jacks, and strain gauges, are avail- the cable using the proposed peak-picking algorithm.
able to measure tension forces directly on cables, they (2) The cable tension forces are estimated with good
are rarely used due to their high-cost and tough main- agreements with those measured using strain gauges
tenance. Vibration methods which estimate the cable for all the cases regardless of tension forces and sag-
tension force using modal properties and geometries of ging conditions, which indicates the applicability of
the cable are now widely used because they require less the present system to actual cable-supported bridges.
invasive instrumentation and are easier to implement (3) Embedding Welchs method is a good solution for
cost-effectively. more accurate estimation of the natural frequencies by
Meanwhile, there has been an increasing interest removing the nonstationarity of short-duration signals,
in the wireless sensor technology in the structural which is a limit of the memory-constrained wireless
health monitoring field. Low-cost wireless sensors sensors.
without any cabling work are the enabler of realizing
the structural health monitoring system economically
and effectively on a large structure. The wireless sen- REFERENCES
sors also have a big benefit of embedded computation,
which can infer some important information from the Lynch, J. P. & Loh, K. 2006. A Summary review of wire-
less sensors and sensor networks for structural health
measured data by itself (Lynch & Loh 2006, Lynch
monitoring. Shock and Vibration Digest, 38(2): 91128.
2007). Lynch, J. P., Wang, Y., Loh, K., Yi, J. H., & Yun, C. B. 2006.
In this study, a smart wireless tension force esti- Performance monitoring of the Geumdang Bridge using a
mation system is developed by embedding a vibration dense network of high-resolution wireless sensors. Smart
method for cable tension force estimation into a wire- Materials and Structures, 15: 15611575.
less sensor. Developed system is composed of two Lynch, J. P. 2007. An overview of wireless structural health
parts, low-cost hardware and automated software. The monitoring for civil structures. Philosophical Transac-
low-cost hardware is composed of a wireless sensing tions of the Royal Society of London. Series A, Mathe-
unit (Wang et al. 2005), a signal conditioning board matical and Physical Sciences, 365(1851): 345372.
Zui, H., Shinke, T., & Namita, Y. 1996. Practical formulas for
(Lynch et al. 2006), and a MEMS accelerometer. An
estimation of cable tension by vibration method, Journal
automated peak-picking algorithm to extract natural of Structural Engineering, 122(6): 651656.
frequencies of a cable from response spectrum without Wang, Y., Lynch, J. P., & Law, K. H. 2005. Validation
human intervention is developed and embedded into of an integrated network system for real-time wireless
the wireless sensors in conjunction with a vibration monitoring of civil structures. Proceedings of the 5th
method. To extract the accurate natural frequencies International Workshop on Structural Health Monitoring,
of the cable with the limited data storage of the Stanford, CA, September 1214.
developed system, Welchs method is additionally
coded to be embedded into the developed system.
86
An active sensor placement optimization strategy using data-driven

Bayesian experimental design
E.B. Flynn & M.D. Todd

University of California, San Diego, California, United States
ABSTRACT Past studies have determined optimal arrangements

through development of numerical or analytical mod-
Active sensing for structural health monitoring is the els and simulation of sensor network performance
process of imparting energy into a structure, detect- (Flynn & Todd 2009). This approach is necessary in
ing the resulting structural response, and extracting design situations where there are too many damage
features from the response that correlate with the struc- modes to observe experimentally and/or in situations
tures state of health (Farrar and Worden 2007). A where the structure is too large or complex to instru-
commonly-exploited implementation of this approach ment densely. In the application at hand however, it was
is the use of piezoelectric transducers to launch and possible to take advantage of the relatively low num-
detect ultrasonic elastic waves to probe for defects ber of possible damage states as well as the repeated
(Giurgiutiu et al. 2002). patterns of joints in the full structure. This enabled the
Given typically limited actuator/sensor budgets, it is implementation of data-driven models derived from
very desirable to place these transducers on the struc- experiment with direct observation of all possible
ture in a way that optimizes a certain objective. Using a damage modes.
general formulation of Bayesian experimental design, We considered three optimization scenarios for this
we develop a global optimality criterion within a detec- study. The first two involve variations on the form
tion theory framework based on a cost-weighted sum of the detection loss function with the exclusion of
of the expected occurrence of Type I (false positive) instrumentation cost, while the third incorporates a
and Type II (false negative) errors. per-sensor cost function. Inconsistencies in the opti-
Our approach generalizes to any derived feature mal sets of transducers among the three scenarios
set under some simple assumptions, and it includes demonstrated the significant impact the relative costs
the possibility of non-uniform probabilities of damage have on the optimal transducer arrangement.
occurrence on the structure. The optimization config-
uration space is searched by a greedy algorithm with
a fitness function that evaluates the Bayes risk. REFERENCES
The method is unique in that it makes use of real-
world cost specifications, which allows the optimal Farrar, C. R, and K. Worden. 2007. An introduction to struc-
tural health monitoring. Philosophical Transactions of the
selection of both the transducer locations and the
Royal Society A: Mathematical, Physical and Engineering
number of transducers to implement. Sciences 365, no. 1851: 303315.
The target structure for this study was the Alamosa Flynn, EB, and MD Todd. 2009. A Bayesian approach to
Canyon Bridge in southern New Mexico, USA. This optimal sensor placement for structural health monitoring
is a decommissioned steel beam-reinforced concrete with application to active sensing. Mechanical Systems
bridge which runs along-side the current Interstate-40 and Signal Processing.
bridge. Each of the seven spans of the bridge con- Giurgiutiu, V., A. N Zagrai, and J. J Bao. 2002. Piezoelectric
tains ten steel connections which join the five primary wafer embedded active sensors for aging aircraft structural
longitudinal beams with the eight lateral beams using health monitoring. Structural Health Monitoring 1, no.
1: 41.
14, 44.5 mm bolts spaced 760 mm apart. The objective
was to determine the optimal placement of piezoelec-
tric active transducers at each of these joints in order to
detect the loss of preload on any of the fourteen bolts.
87
Investigation of the control performance of the smart passive system based

on MR damper using hybrid simulation
D.D. Jang, I.H. Kim & H.J. Jung

KAIST, Daejeon, Republic of Korea
J.H. Koo
Miami University, Oxford, OH, USA
ABSTRACT
This paper evaluates the control performance of the

smart passive system employing a magnetorheological
(MR) damper and an electromagnetic induction (EMI)
device for suppressing the vibration of a stay cable. An
MR damper is one of the most promising semi-active
control devices (Dyke et al., 1996), and an EMI device
which consists of permanent magnets and a solenoid
coil can be considered as a power supply for an MR
damper by generating the electricity from the vibration
of structures according to the Faradays law (Cho et al.,
2005). To validate the effectiveness of the proposed
system, a hybrid simulation which is combination of
the experimental test part of the smart passive system
(i.e., an MR damper and EMI device) and the numer-
ical analysis part of a stay cable structure has been
performed with the white noise and generated time
history of wind load (Nakashima et al., 1992). Figure
1 shows the concept and flow of the hybrid simulation.
The displacement at the damper location is calculated
numerically, and it is used as an input command for the
shaking table test with the prototype MR damper and Figure 2. Time histories of the displacement at the
EMI device, and then the damper force which is mea- mid-point of the cable.
sured, is put into the numerical simulation. As shown
in Figure 2, the responses of the controlled case are
much smaller than those of the uncontrolled case. It is
demonstrated from the results that the proposed system
could be the promising control system for suppressing
the excessive vibration of stay cables.
REFERENCES
Cho, S.W., Jung, H.J. & Lee, I.W. 2005. Smart passive system
based on magnetorheological damper. Smart Materials
and Structures. 14: 707714.
Dyke, S.J., Spencer, B.F., Jr., Sain, M.K. & Carlson, J.D. 1996.
Modeling and control of magnetorheological dampers
for seismic response reduction. Smart Materials and
Structures. 5: 565575.
Nakashima, M., Kato, H. and Takaokas, E. 1992. Develop-
ment of real-time pseudo dynamic testing. Earthquake
Engineering and Structural Dynamics. 21: 7992.
Figure 1. Concept diagram of the hybrid simulation.
88
Autonomous structural health monitoring using wireless smart

sensors on a cable-stayed bridge
S. Jang, H. Jo, K. Mechitov, S.-H. Sim, B.F. Spencer, Jr. & G. Agha
University of Illinois at Urbana Champaign, Urbana, USA
S. Cho, H.-J. Jung & C.-B. Yun

Korea Advanced Institute of Science and Technology, Daejeon, Korea
J.A. Rice
Texas Tech University, Lubbock, Texas, USA
ABSTRACT
Structural health monitoring for long-span bridges

using wireless smart sensor network (WSSN) has
drawn significant attention. The benefits of the WSSN
are low-cost, time efficiency, and post-processing-free
operation. However, some drawbacks have remained
with respect to implementation of WSSNs on long- Figure 3. Smart sensor units. Figure 4. Solar panel.
span bridges, including power consumption, man-
ual communication with the sensor network, and limitations on the hardware. In this paper, the solu-
tions for these drawbacks are provided by an energy
efficient autonomous monitoring strategy, combined
with newly developed sensor boards. The elements of
the autonomous monitoring strategy consist of energy
efficient sleeping mode and automatic threshold-based
sensing. A tri-axial acceleration board with tempera-
ture and humidity measurements and a wind speed
sensor board are utilized. To demonstrate the effi-
cacy of this hardware/software solution, 70 Imote2s
loaded with the developed software package have been
deployed on the Jindo Bridge in South Korea. For fur-
ther energy efficiency, several of these nodes are solar
powered.
Figure 1. Jindo Bridges.
Figure 2. Sensor locations.
89
Evaluation of ultimate performance of the reinforced concrete T-girder

bridge using optical fiber sensors
D.-S. Jung
Hybrid Structural Testing Center, Myongji University, Yongin, Republic of Korea
C.-Y. Kim, S.-M. Cho, S.-A. Choi & B.-Y. Seo

Dept. of Civil & Env. Eng., Myongji University, Yongin, Republic of Korea
K.-W. Lee
Maintra Co. Ltd., Seoul, Republic of Korea
ABSTRACT methods used to get load-carrying capacity of real

structure. In this study, both DLT and PLT are per-
Electric resistance type sensor has been widely used formed for short span RC-T type girder bridge which is
for field measurement of civil structure such as bridge. not in use anymore. Fiber Bragg Grating sensor (FBG
This type of sensor shows some disadvantages for sensor) is used together with electric resistance type
long term application, such as decaying, damage by sensor for DLT to investigate the usage of optical fiber
lightning, noise from electromagnetic field, lack of sensor for bridge load-carrying capacity evaluation.
workability etc. It is also known that this type of sensor, FBG sensor can replace traditional electric resis-
even in short term application, gives low credibility tance type sensor for the measurement of civil struc-
measurement due to the effects of electromagnetic tures such as bridges at site due to the strong advantage
wave when it is used for directly exposed structures of being free of environmental conditions, especially
as well as excessive time and expenses are consumed electromagnetic wave. Furthermore, FBG sensor can
for complicated cabling. Studies for utilizing optical reduce measurement error since it could measure much
fiber sensor to compensate shortcomings of existing clearer response. The result of load-carrying capacity
sensors are being carried out actively. test of bridge can be greatly differed due to mea-
Diagnostic Load Test(DLT) that estimates load- surement errors and/or modeling errors of FE model.
carrying capacity by using response and measurement Hence, more reliable load-carrying capacity analysis
from static-dynamic test of vehicle load and Proof can be possible by minimizing measurement errors
Load Test(PLT) that evaluates load-carrying capacity using FBG sensor at site.
directly from applying large load on the structure until
main members show nonlinear behavior are general REFERENCES
AASHTO, 2003. Manual For Condition Evaluation Load and
Resistance Factor Rating (LRFR) of Highway Bridge.
Merzbacher C.I. et al, 1996. Fiber optic sensors in con-
crete structures: A review. Smart Structures and Materials,
196208.
Guerin J.J. et al, 1994. Embedded optical fiber strain gauges
for civil engineering: application to concrete monitoring.
Proc. 2nd Eur. Conf. on Smart Structures and Materials,
224227.
Inaudi D. et al, 1998. Static and dynamic bridge monitoring
Figure 1. Packaged FBG sensor. with fiber optic sensors. Proc. SPIE Vol. 3555: 136146.
Udd E., 1995. Fiber Optic Smart Structures. John Wiely Inc,
New York.
Maher M.H. & Nawy E.G., 1993. Evaluation of fiber optic
bragg grating strain sensor in high strength concrete
beams. Application of Fiber Optic Sensors in Engineering
Mechanics(New York : ASCE), 120133.
Masri S.F. et al, 1994. Experimental study of embedded fiber
optic strain gauges in concrete structures. J. Eng. Mech.
120: 16961717.
Pope C.E. et al, 1992. Integrated fiber optic strain sen-
sor. Optical Design and Processing Technologies and
Figure 2. Response of exciting at 6.0 Hz (FBG sensor). Applications, SPIE, 1779, 113121.
90
Statistical damage assessment based on the extreme value distribution

using vibration responses
J. Kang
University of California Irvine, Irvine, USA
H.W. Park
Dong-A University, Busan, South Korea
ABSTRACT
A baseline statistical distribution is often required

using the dynamic response data of an undamaged
structure so that the existence of damage can be con- Figure 2. The geometry and the measurement point for
firmed in the context of statistical damage assessment. acceleration data of a two-span continuous truss bridge with
Generally, damage-sensitive dynamic response data moving vehicle load.
of a structure manifest themselves near the tail of a
baseline statistical distribution. In this regard, some researchers have paid attention to extreme value distri-
bution for modeling the tail of a baseline distribution.
However, few researches have been conducted to the-
oretically understand the extreme value distribution
from a perspective of statistical damage assessment.
This study investigates the asymptotic convergence of
domain of attraction through parameter estimation of
the generalized extreme value distribution (GEV). The
outlier threshold value error and the outlier occurrence
probability error are proposed to quantify discrepancy
between the theoretical CDF and the estimated GEV
with respect to the sample size [Figure 1]. The effect
of the sample size on the false positive alarms in statis-
tical damage assessment is quantitatively investigated
as well. Finally, the validity of the proposed method is
demonstrated through numerically simulated ambient
acceleration responses of a two span continuous truss
bridge in Figure 2.
REFERENCES
Kang, J. & Park, H.W. 2009. Understanding the asymptotic
convergence of domain of attraction in extreme value dis-
tribution for establishing baseline distribution in statistical
assessment of a structure. Journal of the Korea Institute
for Structural Maintenance Inspection 13(2): 231242
(in Korean).
Park, H.W. & Sohn, H. 2006. Parameter estimation of the gen-
eralized extreme value distribution for structural health
monitoring. Probabilistic Engineering Mechanics 21(4):
366376.
Sohn, H., Allen, D.W., Worden, K., & Farrar, C.R. 2005.
Structural damage classification using extreme value
Figure 1. Quantification of discrepancy between the theo-
statistics. ASME Journal of Dynamic Systems, Measure-
retical CDF of the Pareto distribution and the estimated GEV
ment, and Control 127(1): 125132.
with respect to the sample size.
91
Prestress-force monitoring using impedance-based smart sensor nodes

in PSC girder bridges
J.-T. Kim, D.-S. Hong, S.-Y. Lee & J.-H. Park

Pukyong National University, Busan, Korea
ABSTRACT
In this study, a technique using wireless impedance

sensor nodes is proposed to monitor Prestress-force in
PSC girder bridges. In order to achieve the goal, the
following approaches are implemented.
Firstly, a wireless impedance sensor node is
designed for automated and cost-efficient Prestress-
loss monitoring as shown in Fig. 1. The wireless
impedance sensor node consists of the power supply,
microcontroller, AD5933, and wireless radio. Figure 2. Measured impedance signatures on interface
Secondly, an impedance-based algorithm is embed- plate.
ded in the wireless impedance sensor node for
autonomous structural health monitoring based on the
decentralized wireless sensor network. The overall
procedure of the operation scheme using root-
mean-square-deviation (RMSD), correlation coeffi-
cient (CC), and peak-frequency shift is summarized
as follows:
(1) Broadcast a start command for measuring
impedance signatures by the remote control
server;
(2) Acquire impedance signatures {Zi (), i = 1, . . . , n}
and store in individual wireless impedance sensor
node (WISN);
(3) Broadcast impedance signatures by the captain
sensor node (i.e., WISN i) to all other wireless
impedance sensor nodes;
(4) Compute impedance features {IF i , i = 1, . . . , n}
at individual wireless impedance sensor nodes; Figure 3. Prestress motoring results using RMSD by wire-
(5) Transmit the impedance features extracted at indi- less impedance sensor node.
vidual wireless impedance sensor nodes to the
captain wireless impedance sensor node (WISN i). Thirdly, a Prestress-loss monitoring technique using
an interface plate is proposed to overcome limitations
of the wireless impedance sensor node such as mea-
sureable frequency ranges with narrow band as shown
in Fig. 2.
Finally, the feasibility and applicability of the pro-
posed technique were evaluated in a lab-scaled PSC
girder model for which several prestress-loss sce-
narios are experimentally monitored by the wireless
impedance sensor node. And though embedded SHM
algorithm for prestree-loss of PSC girder, the wire-
less impedance sensor node successfully monitored
prestress-loss in tendon as shown in Fig. 3.
Figure 1. Prototype of wireless impedance sensor node.
92
Tension monitoring of a prestressing strand for concrete bridge

using in-tendon FBG sensors
J.-M. Kim, H.-W. Kim & Y.-S. Kim

Department of Civil & Environmental Engineering, Chonnam National University,
Yeosu, Republic of Korea
Y.-H. Park
Expressway & Transportation Research Institute, Hwaseong, Republic of Korea
ABSTRACT
Since the first post-tensioned concrete bridge was built

in Aue, Saxony, Germany in 1936, the concept has
widely been used all over the world, and is still up
to date. However, it was realized that the post-tension
system is subject to long-term risk such as corrosion
of tendon caused by ingress of water and chloride
irons into partially grouted ducts. The tensile force
of a prestressing strand can be varying due to various
reasons of loss including instantaneous losses such as
elastic shortening, friction, and anchorage set during
transferring the prestressing forces, and time depen-
dent losses due to steel relaxation, concrete creep and
shrinkage that occur after transfer and during the life of
the structural member. Accordingly, measurement of
tensile force of the tendon becomes very important for
long-term maintenance of the bridge as well as design
purpose.
In this study, we propose a novel method for evaluat-
ing tension force of a prestressing 7-wire strand using
in-tendon FBG sensor (Fig. 1). For the sake of demon-
strating effectiveness of the proposed idea, we came
up with 1.0m, 8.0m and 64.0m long tendon prototypes
in which FBG sensors are embedded into the tendon,
and their accuracy and effectiveness are shown with
test results (Fig. 2).
Figure 2. Experimental results of in-tendon FBG sensors

Figure 1. Concept of in-tendon FBG sensor. for 1.0 m, 7.0 m and 64.0 m long tendons.
93
Evaluation of the characteristics of local erosion of fine-grained

soils in the west coast area of Korea
J. Lee, K. Kwak, J. Park & M. Chung

Korea Institute of Construction Technology, Gyeonggi-do, Republic of Korea
ABSTRACT Similarly, the average critical water velocity stood

at 0.57 m/s in Incheon Bridge, 0.95 m/s in Ganghwa
As a recent global trend, the characteristics of soil Choji Bridge, and 2.45 m/s in Hwankyung Bridge.
erosion as the resistance factor against scouring are In addition, the quantitative scour rate-shear stress
quantified and considered in the estimation of scour curve was determined for each target site through
depths in fine-grained soils. As part of such efforts, the tests, and a chart representing the erosion characteris-
characteristics of local erosion of fine-grained soils in tics for each site was presented. These systemized
the West Coast area where several sea-crossing long- data are provided as basic data for calculating more
span bridges have been planned and constructed in accurate scour depths and developing basic designs
recent years are analyzed through scour rate experi- for the construction of long-span bridges in the West
ments. Three sites Incheon Bridge, Choji Bridge, Coast area.
and Hwankyung Bridge were finally selected for
this study; scour rate tests were performed using 29
undisturbed soil samples from the sites. REFERENCES
According to the results of this study on the char-
Briaud, J.L., Ting, F., Chen, H.C., Gudavalli, S.R., Perugu, S.
acteristics of erosion of fine-grained soils in the West and Wei, G. 1999. SRICOS: Prediction of Scour Rate
Coast area, the erosion resistance ability (critical shear in Cohesive Soils at Bridge Piers, Journal of Geotechni-
stress) of most soils except a few samples tended to be cal and Geoenvironmental Engineering, Vol. 125, No. 4,
generally proportional to the undrained shear strength ASCE, Reston, Virginia, U.S.A., pp. 237246.
values of the soils. In addition, fine-grained soils con- Kwak, K. 2002. S-SRICOS method : Prediction of Scour
taining relatively more cohesive clay recorded lower Depth Around Bridge Piers in Cohesive Soils, Journal of
scour rates and far larger critical shear stresses at a Geotechnical Engineering, Korean Geotechnical Society,
steady water velocity than fine-grained soils without vol. 18, No. 2, pp. 1321.
Ministry of Public Administration and Security. 19952003.
cohesive clay; hence the higher resistance to scour.
Annual Disaster Report, National Disaster Prevention and
The average critical shear stress as the erosion resis- Countermeasures Headquarters.
tance ability of soils in the target sites was 1.15 N/m2 Shirole, A.M. and Holt, R.C. 1991. Planning for a com-
in Incheon Bridge, 2.51 N/m2 in Ganghwa Choji prehensive bridge safety assurance program, Washing-
Bridge, and 12.42 N/m2 in Hwankyung Bridge, show- ton, DC., Transportation Research Record No. 1290,
ing differences in keeping with the soil characteristics. pp. 137142.
94
Impedance-based structural health monitoring using neural networks

for autonomous frequency range selection
J. Min & C.-B. Yun

S. Park
Sungkyunkwan University, Suwon, Korea
ABSTRACT
The impedance-based structural health monitoring

(SHM) method comes to the forefront in the SHM
community due to its practical potential for real
applications. Regarding the impedance-based SHM
method, the selection of the optimal frequency ranges
plays an important role in improving the sensitivity of
damage detection since improper selection of the fre-
quency range might lead to erroneous damage detec-
tion results and provide false positive damage alarms.
To tackle this issue, this paper proposes an innovative
technique to autonomously select damage-sensitive Figure 1. Scheme for autonomously selecting the most
frequency ranges using the artificial neural networks sensitive frequency range using artificial neural networks.
(ANN). At first, the impedance signals are obtained in
a wide frequency band, and the signals are split into
multiple sub-ranges of the wide frequency band. Then, frequency ranges can be effectively utilized to eval-
the pre-defined damage ind0 is evaluated for each sub- uate the damage severity for various damage cases in
range by comparing the impedance signals measured real structures.
between the intact and the concurrent cases. Herein,
the cross-correlation coefficients (CC) are used as the
pre-defined damage index. The ANN is constructed REFERENCES
using the CC values for multiple frequency ranges
as multi-inputs and the real damage severity as a Lopes Jr., Park G, Cudney HH, and Inman DJ (2000)
single-output. The ANN is trained using a number Impedance-Based Structural Health Monitoring with
of impedance signals for various damage scenarios Artificial Neural Networks, Journal of Intelligent Mate-
pre-selected, so that sub-sequent damage estimations rial Systems and Structures, Vol. 11, pp. 20614.
may be carried out by selecting the damage-sensitive Koo KY, Park S, Lee J-J, and Yun C-B (2009) Automated
Impedance-based Structural Health Monitoring Incorpo-
frequency ranges autonomously (Figure 1).
rating Effective Frequency Shift for Compensating Tem-
Experimental investigations were performed to val- perature Effects, Journal of Intelligent Material Systems
idate the performance of the proposed technique for and Structures, Vol. 20, pp. 36777.
both automatically determining the frequency range Okafor AC, Chandrashekhara K, and Jiang YP (1996)
and simultaneously detecting loose bolts and artificial Delamination prediction in composite beams with built-
crack damages inflicted on real building and bridge in piezoelectric devices using modal analysis and neu-
structures. It is found that the proposed approach ral network, Smart Materials and Structures, Vol. 5,
using the autonomously selected damage-sensitive pp. 33847.
95
Output only modal identification and damage detection of bridge-structures

using time frequency and wavelet techniques
S. Nagarajaiah
Department of Civil & Environmental Eng. and Dept. of Mechanical Eng. & Material Sc., Rice University,
Houston, Texas, U.S.A.
B. Basu
Department of Civil, Structural, & Environmental Eng., Trinity College, Dublin, Ireland
V.S.M. Vemuru
Department of Civil & Environmental Eng., Rice University, Houston, Texas, U.S.A.
ABSTRACT damping is also presented. In this paper STFT, EMD,

HT and wavelet techniques are developed for decom-
The primary objective of this paper is to develop out- position of free vibration response of MDOF systems
put only modal identification and structural damage into their modal components. Once the modal compo-
detection. Identification of multidegree of freedom nents are obtained, each one is processed using Hilbert
(MDOF) linear time invariant (LTI) and linear time transform to obtain the modal frequency and damping
variant (LTV-due to damage) systems based on Time- ratios. In addition the ratio of modal components at
frequency (TF) techniquessuch as short-time Fourier different degrees of freedom facilitate determination
transform (STFT), empirical mode decomposition of mode shape. In cases with output only modal iden-
(EMD), and waveletsis proposed. STFT, EMD and tification using ambient/ random response the random
wavelet methods are proposed. In addition a Hilbert decrement technique is used to obtain free vibration
transform (HT) approach to determine frequency and response.
96
Safety network system integration for bridge structures in Korea
K.T. Park, Y.J. Yoo, B.C. Joo, J. Lee & Y.K. Hwang
Korea Institute of Construction Technology, Goyang, South Korea
ABSTRACT
In general, structures in service gradually lose their

original performance-level over time due to initial
defects in design and construction, or exposure to unfa-
vorable external conditions such as repeated loading or
deteriorating environment, and in extreme cases, may
collapse. So, in order to maintain the serviceability of
structures at optimal level, advanced structure mea-
suring system which can inform optimal time point
and method of maintenance is required in addition to
accurate prediction of residual life of the structure by
periodic inspection. In Korea, the large-scale project
for the safety network integration for long-term smart
monitoring systems for bridge structures have been Figure 1. Schematic diagram of remote data transmission
performed since 2006, and this is the 4th year of the system.
project.
In this system, various structure types such as
bridge, tunnels and cut slopes were considered with an
emphasis on safety. The purpose of this system is to construction of the 2nd test-bed system. The inte-
integrate the operation centers for these structure types grated safety management network monitoring system
in order to control and evaluate the structures real- will be complemented in 2011 by adding two more
time safety level through measured behavioral data. bridges.
For the bridge structures safety network integration,
four test-bed bridges are integrated by mainly using
the FBG(Fiber Bragg Grating) sensor system. For the
effective monitoring the state of the bridge safety level REFERENCES
in the safety network operation centers, various anal- Korea Infrastructure Safety and Technology Corpora-
ysis techniques for the bridge safety level evaluation tion (2005), Development of Internet-Based Structural
are developed and used. In this paper, the explanations Health Monitoring System, research report.
on various scenes from the evaluation techniques and Wong K.Y. (2004) Instrumentation and health monitoring of
the state of the art of the safety network system for the cable-supported bridges, Structural Control and Health
bridges in Korea are suggested. Monitoring, Vol. 11, pp. 91124.
The test-bed system implemented in this study suc- S. K. Kim, H. M. Koh, J. H. Lee, I. H. Bae (2006). Analysis
cessfully configured link-up items from the existing on the Signal Data of the Yeongjong Bridge Measure-
ment System A collection of learned papers, Earthquake
measurement system and conveyed realtime and his-
Engineering Society of Korea, Vol. 10, Issue 6, pp. 918.
tory data, showing its capability to be the foundation Ministry of Land, Transportation and Maritime Affairs
of further measurement systems integration. (2008), Model Construction of Safety Management Net-
The present research has completed construction works for the Important National Structures and Develop-
and supplementation of the 1st test-bed system and ment of Their Operation Systems (II), research report.
97
Damping properties identified from wind-induced vibration

measurements of a suspension bridge
D.U. Park & N.S. Kim

H.K. Kim
Seoul National University, Seoul, South Korea
ABSTRACT
In a suspension bridge, the damping ratio as an index

suppressing bridge vibration could be considered as
one of the important dynamic characteristics. But, in
general, estimation of damping ratios can be a labori-
ous task when the existing bridges are under ambient
vibration conditions. In this paper, the damping ratios
on Sorok Bridge, a suspension bridge in Korea, were
estimated by using two different damping ratio esti- Figure 1. Damping ratio estimation results of
mation methods that can be properly applied under Hilbert-Huang transform.
ambient vibration conditions.The first is using Hilbert-
Huang Transform and the other is using extended
Kalman filter as a nonlinear system identification tech-
niques. When ambient vibration signals are measured
from a bridge, it is not simple to separate extract
viscous, Coulomb friction and aerodynamic damping
properties from apparent damping ratios directly using
the measured signals. But, if the viscous and Coulomb
friction damping properties are classified as structural
damping, the apparent damping ratios can be separated
into structural and aerodynamic damping properties.
Half-power band width method and logarithmic Figure 2. Results of estimation by extended Kalman filter.
decrement are well known for estimating damping
ratio as an important parameter of bridges. However, And, Distinguishing structural damping and aero-
these methods could be hard to apply for ambient dynamic damping from appearance damping was hard
vibration signals from existing bridges. In this study, only using appearance damping with wind. It could be
Hilbert-Huang transform(HHT) and extended Kalman possible by adding results of exciting test with each
filter(EKF) was used for estimating damping ratio. other RMQ value of acceleration on center of main
In this study, Sorok Bridge was selected as an esti- span.
mation object. Sorok Bridge is a suspension bridge
that has main span of 250 meters, and side span of 110
meters in South Korea. Although Sorok Bridge was REFERENCES
completed in 2008, it had been closed to traffic because
Macdonald, J.H.G. and Daniell, W.E., 2005, Variation of
of incompletion of connecting road. After opening on modal parameters of a cable-stayed bridge identified
March 2009, the bridge was a little traffic because of from ambient vibration measurements and FE modeling,
the same reason until the end of 2009. These condi- Engineering Structures, 27:19161930
tions made a good subject to investigate the relation Yi, J.H., Koo, K.Y., Kim, N.S. and Yun, C.B., Estimation
between wind velocity, acceleration and damping ratio of cable characteristics using extended Kalman filter,
of existing bridge. The Magazine of the Korean Society of Civil Engineers,
At the result as comparing Figure 1 and 2, results 24(6A):12671274
by EKF could be shown more stable than result of Griffin, M. J., 1990, Handbook of Human Vibration, London;
Hilbert-Huang transform. Elsevier Academic Press
98
Debonding condition monitoring of a CFRP laminated concrete beam

using piezoelectric impedance sensor nodes
S. Park & S.-K. Park

Department of u-City Design and Engineering, Sungkyunkwan Univ, Suwon, Gyeonggi, Korea; Department
of Civil and Environmental Engineering, Sungkyunkwan Univ, Suwon, Gyeonggi, Korea
J.-W. Kim & H.-J. Chang

Department of u-City Design and Engineering, Sungkyunkwan Univ, Suwon, Gyeonggi, Korea
ABSTRACT Table 1. Damage index: 1-CC (Cross Correlation

Coefficient).
Recently, carbon fiber reinforced polymer (CFRP)
laminated concrete structures are being very widely 1-CC
used in various engineering fields because of their PZT (Cross Correlation Coefficient)
several advantages. But they always have the risk- Condition PZT#1 PZT#2
iness of structural collapse accidents initiated from
the debonding conditions that might occur between Perfect bonding 0 0
CFRP and concrete surface (Kim et al. 2008). This 2 cm debonding 0.15 0.03
paper employs an electro-mechanical impedance- 4 cm debonding 0.18 0.05
6 cm debonding 0.23 0.07
based wireless structural health monitoring (SHM)
technique applying PZT ceramic patches to identify
the debonding condition of CFRP laminated rein-
forced concrete beam (Park et al. 2003; Park et al. it was verified that the impedance-based wireless
2009; Koo et al. 2008). In the experimental study, SHM technique can be effectively used for debonding
CFRP-reinforced concrete specimens were fabricated monitoring of CFRP laminated concrete structures.
and the impedance signals were measured from the
wireless impedance sensor node according to different
REFERENCES
debonding conditions inflicted between concrete and
CFRP as displayed in Figure 1. To investigate quanti- Kim, S.B., Kim, J.H., Nam, J.W., Kang, S.H. & Byeon, K.J.,
tatively the changes of the impedance measured at the 2008. Bond-slip model of interface between CFRP sheets
PZT patches due to the debonding conditions, cross and concrete beams strengthened with CFRP. Korea
correlation (CC)-based data analysis was conducted Concrete institute, 20(4): 477486.
and its results are showed in Table 1. From this study, Koo K.Y., Park, S., Lee, J.-J. and Yun, C.-B. 2009. Automated
impedance-based structural health monitoring Incorporat-
ing Effective Frequency Shift for Compensating Temper-
ature Effects. Journal of Intelligent Meterial Systems and
Structures. 20: 367377.
Park, G., Sohn, H., Farrar, C.R. and Inman, D.J. 2003.
Overview of piezoelectric impedance-based health moni-
toring and path forward, The Shock and Vibration Digest,
35(6): 451463.
Park, S., Shin, H.-H. and Yun, C.-B. 2009. Wireless
impedance sensor nodes for functions of structural dam-
age identification and sensor self-diagnosis. Smart Mater.
Figure 1. Test specimen with CFRP debonding condition. Struct. 18: 111.
99
Development of a wireless power and data transmission system using laser

and optoelectronic devices for guided wave-based structural health
monitoring
H.J. Park, H. Sohn & C.-B. Yun

J. Chung
CyTroniq Co., Ltd, Asan, Korea
I.B. Kwon
Korea Research Institute of Standards and Science, Deajeon, Korea
ABSTRACT
In recent years, guided wave based structural health

monitoring (SHM) techniques have attracted much
attentions, because they are not only sensitive to small
defects but also capable to propagate over a long dis-
tance in plate and pipe like structures. A number of
studies have demonstrated the potential of guided wave
based SHM.
These guided waves in a structure can be gener-
ated and sensed by a variety of techniques (Su et al. Figure 1. Overall Schematics of the optic-based wireless
2006). The ultimate goal of our research is to develop power and data transmission system.
an optical system for guided wave generation and sens-
ing as in Figure 1. This paper mainly focuses on the 1.5
Wireless
Output Voltage(mV)
1 Wired
excitation aspect of the overall system. The proposed 0.5
wireless technique transmits power necessary for PZT 0

0.5
excitation using laser and optoelectronic devices. First, 1
a desired waveform is generated and the intensity of the 1.5
0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
laser source is modulated accordingly using an electro- Time (msec)
1
optic modulator (EOM). Next, the modulated laser is Wireless
Output Voltage(mV)
Wired
0.5
wirelessly transmitted to a photodiode connected to
a PZT. Then, the photodiode converts the transmitted 0
0.5
light into an electric signal and excites the PZT to gen-
1
erate guided waves on the structure where the PZT is 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
attached to. Finally, the corresponding response from Time (msec)
the sensing PZT is measured. The feasibility of the

proposed method for wireless guided wave generation Figure 2. Comparison of the guided waves generated by
wired and wireless systems.
has been experimentally demonstrated. (Figure 2).
The results confirm that the laser-based guided wave guided waves (Lee et al. 2009). A further research is
generation technique exhibits the reasonable perfor- under way to use other optoelectronic devices, such as
mance by comparing the conventional wiring guided laser diodes and laser vibrometers for wireless guided
wave generation method. Because power is remotely wave sensing and improve the aiming capability of the
transmitted to the PZT transducer, no complex elec- laser source.
tronic components are necessary at the PZT node,
making it possible to develop a self-contained, rugged, REFERENCES
and non-intrusive PZT node. The transducer itself can
be entirely passive and consequently will have a long Kasap, S.O. 2001. Optoelectronics and Photonics: Principles
lifetime when it has rugged packaging. In addition, and Practices. New Jersey: Prentice Hall.
Smith, S.D. 1995. Optoelectronic Devices. New Jersey:
any arbitrary waveform can be generated by a laser Prentice Hall.
using an electro-optic modulator (EOM). This wire- Su, Z. & Ye, L. & Lu, Y. 2006. Guided Lamb waves for iden-
less power transmission scheme also can be expanded tification of damage in composite structures: A review.
to transmit power through optical fibers and generate Journal of Sound and Vibration. 295: 753780.
100
Development of structural health monitoring systems for railroad

bridge testbeds
H.J. Park & C.-B. Yun

M.H. Shin
Korea Railroad Research Institute, Uiwang, Korea
J.J. Lee
Sejong University, Seoul, Korea
S.Y. Park
Korea Maintenance Co., Ltd., Seoul, Korea
ABSTRACT As a part of the project, SHM systems have been

established on railroad bridges. This paper presents
There has been a demand to monitor the health of the current status of railroad bridge health moni-
infra structures, in order to ensure safety and service- toring testbeds. Emerging sensors and monitoring
ability to the community. Structural health monitoring technologies are under investigation. They are local
(SHM) is a methodology to monitor the performance damage detection by monitoring electro-mechanical
of a structure and identify incipient failure conditions, impedances using piezoelectric sensors and structural
which helps in improving the safety and life span of strains using optical fiber sensors; vibration-based
the structure. The SHM system often offers an oppor- global monitoring using accelerometers, and tempera-
tunity to reduce the cost for the maintenance, repair, tures. All of measured data are transmitted to the inte-
and retrofit throughout the life-cycle of a structure. grated control center, where real-time measurements
During the past two decades, the SHM systems are provided under train-transit and environmental
have been widely studied and implemented on many loadings. Overall schematics of national network sys-
infrastructures in the world (Barrish et al. 2000, Sumi- tems are introduced and long-term behaviors of a
tro et al. 2001, Masri et al. 2004). However, in the railroad bridge testbed are investigated using measured
conventional structural monitoring systems, the data data and signal processing results.
acquisition systems were transported frequently to
the monitoring site along with skilled personnel for REFERENCES
acquiring data from the structure (Sridhar et al. 2009).
In addition, each SHM system has been operated and Barrish, Jr. R. A., Grimmelsman, K. A., Aktan, A. E., Instru-
managed independently by its respective management mented monitoring of the Commodore Barry Bridge,
agencies (Kim et al. 2009). Not only those cause much Nondestructive Evaluation of Highways, Utilities, and
time, cost, and efforts for monitoring and assessment Pipelines IV, 2000, Bellingham
a structure, but also make difficult to cooperate with Kim,Y.S.,An, S.S.,Ahn, S.R.,Yang, D.S., The strategic plan-
other related SHM systems. ning of safety management network for infrastructure,
ICOSSAR2009 conference, 2009, Japan
Recently a challenging project has been carried
KISTEC, Key national infrastructure safety management
out for construction of a national network for safety network setup plan project report, 2007
management and monitoring of civil infrastructures Masri, S.F., Sheng, L.H., Caffrey, J.P., Nigbor, R.L.,
in Korea (KISTEC 2007). The main objective of this Wahbeh, M., Abdel-Ghaffar, A.M., Application of a web-
project is to establish real-time SHM systems for dif- enabled real-time structural health monitoring system for
ferent types of infrastructures, such as bridges, cut civil infrastructure systems, Journal of Smart Materials
slops, tunnels, embankments, and playgrounds, and and Structures, 13, 2004, pp. 12691283
integrate each SHM system through on-line networks. Sridhar, S., Ravisankar, K., Sreeshylam, P., Parivallal, S.,
Thus the behavior of each infrastructure could be mon- Kesavan, K., Murthy, S., Development of a Real-time
Remote Structural Monitoring Scheme for Civil Infras-
itored at an integrated monitoring center, which leads
tructural Systems, Structural Health Monitoring, 8,
effective maintenance and management of national 2009, pp. 509521
infrastructures. Additionally damages or abnormali- Sumitro, S., Matsui, Y., Kono, M., Okamoto, T., Fujii, K.,
ties in structural behavior can be observed at an early Long span bridge health monitoring system in Japan,
stage and early alarm system will be operated to Health Monitoring and Management of Civil Infrastruc-
managers and users of the structure. tural Systems, 2001, Bellingham
101
Development of a benchmark laboratory structure for finite-element

model updating
S. Shang, G.J. Yun & S.G. Lee

Department of Civil Engineering, University of Akron, Akron, OH, USA
J. Caicedo
Department of Civil & Environmental Engineering, University of South Carolina, Columbia, SC, USA
S. Narasimhan
Department of Civil & Environmental Engineering, University of Waterloo, Waterloo, Ontario, Canada
ABSTRACT
Finite element models often do not represent the

actual physical characteristics of existing structures.
There are many factors that affect the accuracy of the
numerical model, these include but are not limited
to construction errors such as inappropriate bound-
ary conditions, unplanned loads on the structure, and
material imperfections. The objective of model updat-
ing is to adjust the numerical model of the structure so
that the model predictions are in agreement with the Figure 1. Final 3D Truss Structure and Virtual Reality
test results. Implicit in many model-updating methods Model at the University of Akron.
is that the test results are a true representation of the
actual structure. limited extent. In addition to that, its capabilities to
Recently, a benchmark structure in a two- simulate various physical structural modifications are
dimensional grid form was developed for general deemed the most advantageous in assessing modeling
bridge health monitoring at the University of Cen- techniques and updating algorithms. Finite element
tral Florida (Caicedo et al. 2006; Catbas et al. 2006). models that are close to the physical structure have
However, there has been no benchmark problem that been suggested with introduction to model param-
is devoted to assess various modeling techniques and eters. Experimental impact testing results showed
updating algorithms with realistic three-dimensional that the natural frequencies from FE model are
configurations. fitting well with experimental results. To dissem-
Thus, this paper presents a research initiative toward inate this benchmark structure, sponsorship from
a well-defined benchmark model updating problem the ASCE (American Society of Civil Engineers)
so that participants can implement their updating EMI (Engineering Mechanics Institute) committees
methodologies and compare various methodologies (e.g. dynamics committee and structural health moni-
on a common test-bed. An 8-bay steel truss struc- toring and control committee) will be sought after.
ture (shown in Figure 1) is designed and constructed
to serve as the benchmark test-bed to evaluate FE
model updating strategies. Most of discrepancies from REFERENCES
real-life structures come from neglecting variable joint
Caicedo, J. M., Catbas, F. N., Gul, M., Zaurin, R., and
stiffness at connections, idealizing support boundary
Dyke, S. J. 2006. Benchmark Problem for Bridge Health
conditions, and overlooking any unknown priori struc- Monitoring: Definition Paper. Proceedings of the World
tural flaws or erection errors. Therefore, the structure Conference on Structural Control and Monitoring, San
has been designed with unique features that can simu- Diego, California.
late these often-overlooked structural characteristics Catbas, F. N., Caicedo, J. M., and Dyke, S. J. 2006.
and abnormalities in structural integrity. There are Development of a Benchmark Problem for Bridge Health
provisions in the model to simulate realistic struc- Monitoring. Proceedings of the International Confer-
tural changes due to damage, degradation, erection and ence on Bridge Maintenance, Safety and Management
fabrication errors, etc. (IABMAS), Porto, Portugal.
One of the highlighted design features is its re-
configurability of structural types although it is in
102
Variation of eigen-properties of a PSC bridge due to prestressing force
S. Shin & M.-S. Koo

Inha University, Incheon, Korea
H.-K. Lee
Dong-Bu Eng. Co., Seoul, Korea
S.-J. Kwon
Korail Research Institute, Daejeon, Korea
ABSTRACT 1.30
A finite element formulation for the undamped 1.25

dynamic motion of a beam-tendon system is derived
based on the principle of virtual work. The proposed 1.20
dynamic equation of motion has been verified through

fp/f0
1.15
a comparison of the analytical results with those
obtained from the results of laboratory experiments 1.10
available in the literature, which shows significant
variation in the natural frequencies due to the pre- 1.05 Rect. Cross Section
stressing force. The proposed algorithm has been also T Cross Section
I Cross Section
applied to the analysis of actual PSC girder bridges 1.00
with prestressing. 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55
To compute the first natural frequency in terms of P/Pcr
the prestress directly, a normalized equation of Equa-
tion 1 can be obtained. In the equation, we can observe Figure 1. Variation of fp /f0 with P/Pcr from the proposed
that the variation of the first natural frequency due to model.
the prestress is a function of the relative ratio of the
Table 1. Natural frequencies by Eq. 1 and proposed model.
prestress P and Eulers critical load Pcr of a simply
supported column. Section A-20 A-23 A-25 B-25
P/Pcr 0.0156 0.0283 0.0233 0.0259

fp /f0 (Equation 1) 1.008 1.014 1.012 1.013
fp /f0 (grillage analysis) 1.011 1.018 1.015 1.015
Figure 1 shows the variation of the normalized ratio frequency is attributed to the amount of the prestress-
(fp /f0 ) of the first natural frequency with the increase ing force being negligibly small relative to the critical
of the non-dimensional axial force indicator (P/Pcr ). axial load of a typical PSC girder. Therefore, it can
To investigate the effects of presterssing force on be concluded that it may not be necessary to con-
PSC railway girder bridges, the proposed model has sider the effects of the prestressing force on the natural
been applied. frequencies of actual PSC girder bridges.
Table 1 summarizes the computed critical axial
load for each girder type and the natural frequencies
computed from the proposed finite element model. REFERENCES
The effect of prestressing force on the first natural
DallAsta, A, & Leoni, G. 1999. Vibration of beams pre-
frequency is within 1.5% for all girder types when
stressed by internal frictionless cables. J. of Sound and
Equation 1 is applied. Vibration 222(1): 118.
When the method was applied to actual PSC girder Nabil, F.G. & Ross, B. 1996. Dynamic characteristics of
bridges, the increase of the first natural frequency due post-tensioned girders with web openings. J. of Structural
to the prestressing force was less than 1.5% among Engineering 122: 643650.
the tested cases. This slight increase in the natural
103
Imaging ultrasonic waves in complex structures using a scanning laser

Doppler vibrometer
H. Sohn & J.Y. Yang

Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea
D. Dutta
Carnegie Mellon University, Pittsburgh, USA
M.P. DeSimio & S.E. Olson

University of Dayton Research Institute, Dayton, USA
E.D. Swenson
United States Air force Institute of Technology, Wright-Patterson AFB, USA
ABSTRACT
Ultrasonic wave propagation through metallic and

composite structures is of considerable interest in the
context of non-destructive testing (NDT) and struc-
tural health monitoring (SHM). This paper presents
images of ultrasonic wave fields in specimens varying
in structure and composition. The ultrasonic waves are
generated in the structures using surface bonded piezo-
electric wafer actuators. A scanning laser Doppler
vibrometer is used to image out-of-plane ultrasonic
velocity field across the surface of the structure. The
images thus obtained give valuable insight into the
interaction of ultrasonic waves with various structural
components (like stiffeners, bolts, lap joints and vari- Figure 1. 1.8 mm thick composite plate instrumented with
ation in thickness etc.) as well as with damages (like Kapton coated PZTs.
notches and impact damages). A proper understanding
of such interactions would hopefully lead to improved
damage detection in complex specimens in future.
Elastic waves in solids have been extensively used
for NDT and SHM applications for many years. This
includes the applications of ultrasonic guided waves,
acoustic emissions and vibro-acoustic techniques.
Ultrasonic guided wave based SHM techniques come
with the advantages of moderately large inspection
ranges and high sensitivity to small flaws. It, however,
require the sensors and interconnections to be embed-
ded into the structure, thus adding further complexity
to an already complex structural system (like a bridge
or an aircraft). This limitation of applicability of ultra-
sonic techniques has inspired the development of
non-contact ultrasound measurement systems like the Figure 2. Contour plot of the out-of-plane ultrasonic veloc-
ity field for the specimen shown in Figure 1.
one presented in this paper.
The present study uses a laser Doppler vibrometer laser vibrometry to detect defects in complex struc-
to remotely sense ultrasonic field in a structure. Optics tures. To this effect, the interaction of ultrasonic waves
based generation and measurement of ultrasound has with structural components (like bolts, stiffeners and
been studied in the past (Culshaw et al. 2003). More joints) has been studied in this paper and attempt has
recently, Leong et al. (2005) studied the interaction been made to distinguish such interactions with those
of ultrasonic guided waves with fatigue cracks in a happening at defect locations (notch and impact dam-
plate using a laser Doppler vibrometer. The objective age). An example result for a composite plate with
of the present study is to expand the scope of using impact damage (Figure 1) is shown in Figure 2.
104
Integrated wireless powering and data interrogation for civil

infrastructure monitoring
S.G. Taylor, E.B. Flynn, D.L. Mascarenas, M.D. Todd, D. Dondi, T. Rosing, S. Kpotufe,
S. Dasgupta, K. Lin & R. Gupta
University of California San Diego, La Jolla, CA USA
K.M. Farinholt, G. Park & C.R. Farrar

The Engineering Institute, Los Alamos National Laboratory, USA
ABSTRACT take advantage of this SHM paradigm and do not rely

upon direct human visual inspection have been devel-
Post-event inspection of bridges after unexpected oped. These techniques broadly fall into two general
events (e.g. an earthquake) is an extremely impor- classes: (1) advanced data mining strategies from con-
tant step in the path toward damage assessment and ventional sensor networks (e.g., extracting vibration
functionality restoration. We present how a team of properties from an accelerometer array) and (2) data
structural, mechanical, electrical, and computer engi- mining strategies from remote sensing modes, such as
neers are designing and deploying a variant on sensor satellite imagery or LiDAR, used in conjunction with
networking to address this. The work will present a geographic information systems (GIS). A recent litera-
hybrid approach to sensor array powering and interro- ture review report considering a number of techniques
gation, where both power and interrogation commands for the former class (not necessarily only applied to
are conveyed via autonomous vehicles. A prototype civil infrastructure) may be found in Sohn et al. (2004),
system described here is demonstrated on a full-scale while a more specific report of some of the techniques
bridge. applications to post-earthquake building assessment
A post-event assessment may be particularly impor- may be found in Naiem et al. (2005). A recent arti-
tant if the system is itself a critical link in both the cle discussing the merging of digital photogrammetry
short-term emergency response and in the long-term with a GIS for post-earthquake infrastructural damage
economic viability of the community it serves. An assessment may be found in Altan et al. (2001). In both
example could be a bridge over which rescue vehicles classes of techniques, an appropriate sensor network is
must travel for first response and people must traverse required as a first line of attack in observing the struc-
for safety (short term), and which significant commer- tural system behavior in such a way that suitable signal
cial traffic must utilize (long term). The vast majority processing and damage-sensitive feature extraction on
of these post-event assessments, particularly for civil the measured data may be performed.
infrastructure such as bridges, are currently done by
human visual inspection. In many instances, however,
human visual assessments cannot be made quickly or REFERENCES
efficiently, either a result of inspector unavailability Farrar, C. R., Duffey,T.A., Doebling, S. W., & Nix, D.A. 1999.
or life safety issues posed to the inspector(s). These A Statistical Pattern Recognition Paradigm for Vibration-
aspects are further highlighted when the structure itself Based Structural Health Monitoring. Structural Health
is expansive and/or has a number of areas that are not Monitoring, F. Chang, ed., Lancaster, PA: Technomic,
easily accessible. 764773.
The field of structural health monitoring (SHM) Mascarenas, D.L., Park, G., Farinholt, K.M., Todd, M.D., Far-
is an integrated paradigm (Farrar et al. 1999) of rar, C.R., 2009. A Low-Power Wireless Sensing Device for
networked sensing and actuation, data interrogation Remote Inspection of Bolted Joints. Journal of Aerospace
Engineering, Part G of the Proceedings of the Institution
(signal processing and feature extraction), and statis-
of Mechanical Engineers, 233(5): 565575.
tical assessment (classification of damage existence, Taylor, S.G., Farinholt, K.M., Flynn, E.B., Figueiredo, E.,
location, and/or type) that approaches structural health Mascarenas, D.L., Park, G., Todd, M.D., Farrar, C.R.
assessment in a systematic way. In the last decade, 2009. A Mobile-agent Based Wireless Sensing Network
particularly after the 1994 Northridge, California, for Structural Monitoring Applications. Measurement
earthquake, some automated post-event strategies that Science and Technology, 20(4): 045201.
105
Multi-scale wireless sensor node for impedance-based SHM and long-term

civil infrastructure monitoring
S.G. Taylor, K.M. Farinholt, G. Park & C.R. Farrar

M.D. Todd
University of California San Diego, La Jolla, CA USA
ABSTRACT data from a remote civil infrastructure site when there

are long intervals between on-site maintenance of the
This paper applies a recently developed extremely sensing and data acquisition systems. Such a system
compact, wireless impedance sensor node (WID3, requires a robust sensor network.
Wireless Impedance Device) for use in impedance- While the majority of permanently installed sensor
based structural health monitoring (SHM), sensor networks today employ this wired architecture, devel-
diagnostics and low-frequency vibrational data acqui- opment and deployment of wireless sensor networks
sition. Many SHM projects involving wireless sensor has exploded in recent years. Most wireless sensor
nodes have been conducted, but usually over very network paradigms fall into one of the categories out-
short time frames with most tests lasting only a mat- lined by Farrar et al. (2006). However, while much
ter of days. In this paper, the WID3 is cast as the exploratory work has been done to assess the feasibil-
integral component of a permanent SHM system at a ity of wireless networks for SHM, very little has been
remote civil infrastructure installation, such as a high- done toward a permanently installed wireless sensor
way bridge. The current generation WID3 is equipped network for SHM.
with an Analog Devices AD5933 impedance chip that This paper addresses several of the requirements
can resolve measurements up to 100 kHz, a frequency for a robust and reliable wireless sensor network for
range ideal for many SHM applications. The WID3 a permanently installed SHM system. The key com-
combines on-board processing, data storage, wireless ponents required for such an installation addressed in
communications capabilities, and a series of internal this paper are the measurement devices (sensor nodes),
and external triggering options into a single package to the permanently installed sensors, and the specific net-
realize a truly comprehensive, self-contained wireless working strategies required to collect and analyze the
active-sensor node for SHM applications. The WID3 data. Laboratory proof-of-concept results for various
requires less than 70 mW of power to operate; its aspects of the sensing system are presented, as well as
extremely low duty cycle requirements and its abil- experimental results from field tests at the proposed
ity to operate in various wireless network paradigms permanent wireless SHM system installation site in
make it ideal for permanent monitoring of remote civil southern New Mexico.
infrastructure installations. Experimental results sup-
porting the ability of the WID3 to accurately detect
damage in a permanently installed configuration over REFERENCES
a period of months are presented.
Structural health monitoring (SHM) is the process Farrar, C. R., Duffey,T.A., Doebling, S. W., & Nix, D.A. 1999.
of detecting damage in structures. The goal of SHM A Statistical Pattern Recognition Paradigm for Vibration-
is to improve the safety and reliability of aerospace, Based Structural Health Monitoring. Struct. Health Mon.,
civil, and mechanical infrastructure by detecting dam- F. Chang, ed., Lancaster, PA: Technomic, 764773.
age before it reaches a critical state. A more detailed Park, G., Sohn, H., Farrar, C.R., Inman, D.J., 2003. Overview
general discussion of SHM can be found in Farrar of Piezoelectric Impedance-based Health Monitoring and
Path Forward. The Shock and Vibration Digest. 35:
et al. (1999). In order for any SHM system to be suc-
451463.
cessful, there must be a reliable stream of operational Taylor, S.G., Farinholt, K.M., Flynn, E.B., Figueiredo, E.,
data available for analysis, and little or no interven- Mascarenas, D.L., Park, G., Todd, M.D., Farrar, C.R.
tion should be required on the part of engineers to 2009, A Mobile-agent Based Wireless Sensing Network
obtain such data. This study addresses the reliabil- for Structural Monitoring Applications. Measurement
ity and longevity issues surrounding the collection of Science and Technology. 20 (4): 045201.
106
Energy harvesting and wireless energy transmission for powering SHM

sensor nodes
S.G. Taylor, N.A. Miller, K.M. Farinholt, G. Park & C.R. Farrar
ABSTRACT
The goal of this investigation has been to examine the

feasibility of several different energy harvesting and
wireless transmission approaches to power embedded
sensing hardware. The proper management of energy
resources is essential in the development of a robust
wireless sensing system (Park et al. 2008). For long-
term deployments, innovative power methods must be
developed to supplement, or replace, the finite lifespan
of traditional battery technologies. In addition, sensor
nodes must be designed with energy efficient operation
as a fundamental design criterion. This paper presents
recent investigations into the use of energy harvesting
and wireless energy transmission to power a wireless
SHM sensor node previously developed by the authors
(Taylor et al. 2009).
For this energy harvesting study, field data were first
collected from a local highway bridge and replicated Figure 1. Voltage profile for a 0.1F supercapacitor being
in the laboratory environment through the use of an charged by RF energy at 5.6 GHz.
electromagnetic shaker. Bimorph configurations of the
piezoceramic lead zirconate titanate (PZT) were used energy was being transmitted. The system was capa-
as electromechanical transducers. We observed that it ble of charging the capacitor to a voltage of 3.7V
required 912 seconds to charge a 0.1F super capacitor in an average time of 53 seconds at a distance of
to 3.5V. Once voltage levels reached 3.5V, energy was 1.21.3 meters. In this figure the wireless trans-
released to the sensor node, which powered itself on mission was initiated and approximately 55 seconds
and measured the electrical impedance of three sen- later the power conditioning circuit was triggered,
sors used to monitor bolt preload. The sensor node causing the sensor node to become active and make a
performed a 100 point measurement sweep on each measurement of the piezoelectric sensors, as indicated
sensor, computing the maximum impedance, and stor- by the voltage drop from 56 to 61 seconds.
ing this data to onboard flash memory. This result
indicates that the PZT-based harvester system is capa-
ble of powering a piece of hardware that is capable REFERENCES
of interrogating the structural health of a number of
Park, G., Farrar, C.R., Todd, M.D., Hodgkiss, W., Rosing, T.,
different systems. 2008, Energy Harvesting for Structural Health Monitor-
Another approach to traditional energy harvesting Sensor Networks, ASCE Journal of Infrastructure
ing systems is the use of RF energy transmission to Systems, 14(1): 6479.
remotely power embedded sensor nodes.The operation Taylor, S.G., Farinholt, K.M., Flynn, E.B., Figueiredo, E.,
of the sensor node exclusively from power obtained Mascarenas, D.L., Park, G., Todd, M.D., Farrar, C.R.
through wireless energy transmission was first demon- 2009, A Mobile-agent Based Wireless Sensing Network
strated in the laboratory and then performed at the for Structural Monitoring Applications, Measurement
Alamosa Canyon Bridge. Figure 1 depicts the charg- Science and Technology, 20 (4): 045201.
ing profile within the 0.1F supercapacitor as the RF
107
Feasibility investigation for identifying bridges fundamental frequencies

from vehicle vibrations
T. Toshinami & M. Kawatani

Department of Civil Engineering, Kobe University, Kobe, Japan
C.W. Kim
Department of Civil and Earth Resource Engineering, Kyoto University, Kyoto, Japan
ABSTRACT
Monitoring aged and deteriorated structures includ-

ing bridges have been an important technical issue in
developed countries. A lot of researches on the bridge
condition screening so called bridge health monitor-
ing (BHM) using vibration data of bridges has been
reported (e.g. Doebling et al. 1996). Most of all the Figure 1. Acceleration responses and Fourier spectra of
existing studies relating to the bridge condition screen- vehicle taken from analysis.
ing have focused on the modal properties and quanti-
ties of bridge structures (e.g. Doebling et al. 1996,
Wenzel & Pichler 2006). The fundamental concept
of this technology is that modal parameters are func-
tions of physical properties of the structure. Therefore
changes in physical properties, such as reductions in
stiffness resulting from damages, will cause detectable
changes in those modal properties.
This paper investigates how vehicles vibrations are
correlated with bridge vibrations to clarify feasibility
of extracting frequencies of bridges (e.g. Ling et al.
2005) from vibration data of the vehicle traveling on
bridges as a part of the drive-by-inspection system
which the authors are developing (Kim and Kawatani Figure 2. Responses of experimental vehicleV2 (M=21.6kg
2009). f=3.76Hz) at speed 0.93 m/s.
In this study, a traffic-induced vibration analysis
of a bridge considering roadway roughness is per- frequencies from vehicle vibrations, which is next step
formed as a preliminary feasibility investigation to for this study.
detect bridges frequencies from vehicles vibrations.
A moving vehicle laboratory experiment is conducted
in order to confirm the feasibility of extracting bridges REFERENCES
fundamental frequencies from the vehicles vibration
considering roadway roughness. Doebling, S.W., Farrar,C.R., Prime, M.B. & Shevitz, D.W.
From the analytical study, it is observed that the fre- 1996. Damage identification and health monitoring of
quency (2.34 Hz) near the bridges frequency (2.36 Hz) structural and mechanical systems from changes in their
appears in the vehicles vibration data as shown in vibration characteristics. A literature review, Los Alamos
National Laboratory report LA-13070-MS.
Figure 1. However, it shows difficulties extracting Kim, C. W. & Kawatani, M. 2009. Challenge for a Drive-by
bridges frequency from the vehicles accelerations Bridge Inspection, In: Proc. the 10th Int. Conf. on Struc-
clearly. On the other hand, from the experiment, the fre- tural Safety and Reliability, ICOSSAR 2009. (CD-ROM)
quency (2.44 Hz) near the bridge frequency (2.54 Hz) Ling, C.W., Lin, C.W. & Yang, Y.B. 2005. Use of a pass-
is more easily detectable from the response spectrum ing vehicle to scan the fundamental bridge frequencies.
of the vehicle in comparing with the analysis as shown Engineering Structures, Vol. 27, pp.18651878.
in Figure 2. It needs, however, further investigations on Wenzel, H. & Pichler, D. 2006. Ambient Vibration Monitor-
the condition of high possibility to detect the bridges ing. John Wiley and Sons, p.70.
108
Analysis and prediction for bridge maintenance costs based on life-cycle and
Markov approach
Y.Q. Xiang, C. Zhou & D.M. Guo

Department of Civil Engineering, Zijinggang Campus, Zhejiang University, Hangzhou, P.R. China
ABSTRACT Key words: Bridge life-cycle; Performance degra-

dation; Condition evaluation; Maintenance costs,
The bridge life-cycle is referred to the whole pro- Markov approach; Analysis
cess on the planning, design, construction, operation,
management and maintenance or recycling for bridge.
The traditional maintenance method for bridges pays REFERENCES
main attention to the current performance and manage-
ment, but not thinks about degradation of the structural Frangopol D M, Lin K Y. Life-cycle cost design of deterio-
rating structures. Journal of Structural Engineering,1997,
behavior and the optimal maintenance and manage-
36 (10):13901401
ment during the whole bridge service life-cycle or Frangopol D M. Life-cycle cost analysis for bridge. Journal
predicting life time. These cause the some problems of Structural Engineering,1999, 66 (6):210236
that the just repaired bridge rose to the new defects, R.H. Wei. The technique economic analysis and life pre-
and even the maintenance cost is very high. diction on service bridge [D]. Xian .A thesis submitted
In China there had been many bridges constructed to Changan University for the degree of master of
near 1990s, these bridges have being serviced for engineering science, Feb, 2003
20 years by now. The degradation and damage phe- Q. Zhong. Research on bridge design theoretical based on the
nomenon of bridge became serious. However the whole life cycle cost [D]. Changsha. A thesis submitted to
Hunan University for the degree of master of engineering
available capital for maintenance and strengthening
science, 2005
of bridge is very limited. It is worth to studying and Y.Q. Xiang., S.L. Weng, Y. Song, Y.D. Yao. Health mon-
concerning how to make the available fund play the itoring and evaluation management system of Wenhui
biggest benefit, at the same time, make the bridges get Bridge, First International Conference on Structural
most effective maintenance, ensure the bridges to be Health Monitoring and Intelligent Infrastructure, Nov.
in the accepted service level and operation quality. 1315, 2003, Tokyo
Based on condition monitoring and evaluation of Y.Q. Xiang, J.F. Wang, Advance in Health Monitoring and
the on-line bridges, this paper gives the condition Assessment Theory of Long Span Concrete Bridge and
assessment index system of the typical beam bridge, application, The Proceeding of 4th China-Japan-US Sym-
posium on Structural Control and Monitoring Oct.1617,
analyzes the performance degradation of the type of
2006, Hangzhou
bridges causing by time-varied factors, studies the per- Y.Q. Xiang, Y. Li, C. Zhou etc. Healthy Monitoring and
formance degradation of bridge predicted by using Altering System on-Line of Bridge Structures I: Design
Markov approach and the influence of performance of Monitoring Assessment and Altering System and
degradation on bridge maintenance costs. Finally, Modules. [J]. Journal of Transport Science and Engineer-
based on the analysis thought of the life-cycle, the ing, Vol. 25 (No. 1) March. 2009. pp. 2631
calculation formula of the maintenance costs thinking Y.Q. Xiang, C. Zhou, Y. Li. Healthy Monitoring and Altering
about the influence of interest rate on the maintenance System on-Line of Bridge Structures II Signal Analy-
costs is proposed. The obtained results show that it is sis and Extraction Method for Damage Identification [J].
Journal of Transport Science and Engineering, Vol. 25
not only the least comprehensive costs, but also reli-
(No. 2) June. 2009. pp. 3339
able technique, economic and reasonable maintenance
prediction method when the safety operation can be
ensured during bridge life-cycle.
109
Transmissibility-function-based structural damage detection with tetherless

mobile sensors
X. Yi, D. Zhu & Y. Wang

School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA
J. Guo & K.-M. Lee

School of Mechanical Engineering, Georgia Institute of Technology, Atlanta, GA
ABSTRACT In this research, transmissibility function analysis is

adopted for detecting structural damage using the data
In order to reduce the cost and inaccuracy associated collected by mobile sensing nodes.
with human inspection, prototype mobile sensors are A 2D laboratory steel portal frame is constructed for
developed for the damage detection of steel struc- exploring structural damage detection using mobile
tures. The individual flexonic mobile sensing node sensing data. Hinge connections are adopted at the
consists of three substructures: two 2-wheel cars and a bases of the two columns. Three acceleration mea-
compliant connection beam. Each 2-wheel car con- surement locations are assigned on the left and right
tains a body frame, motors, batteries, a wireless columns, respectively. Five acceleration measurement
sensing unit, as well as infrared (IR) sensors and locations are uniformly assigned on the beam. A steel
Hall-effect sensors with associated hardware circuits. mass block of 0.575 kg is bonded to the left column to
Meanwhile, an accelerometer is fixed at the middle simulate a reversible damage. During the validation
of the compliant connection beam, which serves for experiments, two mobile sensing nodes are assem-
attaching/detaching the accelerometer onto/from the bled for simultaneously measuring the vibration at one
structural surface. When a measurement is to be made, pair of neighboring locations. To reduce the effects of
the two cars are driven towards each other to make experimental uncertainty, measurement is repeatedly
the compliant beam buckle downwards to the struc- taken for 20 times at each pair of locations. Based upon
tural surface. With the help of the small magnets fixed the averaged transmissibility functions obtained for
around the accelerometer, the accelerometer is firmly ten pairs of measurement locations, the damage indica-
attached on the surface. When the accelerometer is tors between undamaged structure and damaged struc-
to be detached, the two cars move in opposite direc- ture are calculated. In addition, repeatability indicators
tions to lift the accelerometer away from the surface are computed for the data collected from the undam-
and straighten the compliant beam. The mobile sens- aged structure, as well as for the data collected from
ing nodes are capable of autonomously maneuvering the damaged structure.The results show that the largest
on ferromagnetic surfaces. damage indicator occurs at the location pair where the
Till now, many vibration-based methods have been damage was introduced, while the repeatability indica-
developed for structural damage detection. Among tors illustrate that the experimental uncertainties have
these methods, transmissibility function analysis has limited effects to the damage localization.
attracted considerable attention because of its effec- In this study, the advantage of mobile sensors is
tiveness in identifying damage using output data only. demonstrated as the high spatial resolution measure-
The transmissibility function Tij between the output ment that requires limited number of sensors and little
degree of freedom (DOF) i and reference-output DOF human effort. Such advantage will allow mobile sensor
j is defined as the ratio between two frequency spec- networks bring transformative changes to future prac-
tra Ai and Aj . Based on the calculated transmissibility tice of structural health monitoring. Future research
function Tiju of the undamaged structure and Tijd of will be conducted to enable the mobile sensing nodes
the damaged structure, an integral damage indica- with the capabilities of autonomously detecting poten-
tor between DOFs (i.e. locations) i and j is defined. tial damages in the structure, as well as with the
The locations with the largest damage indicators are capabilities of maneuvering on more realistic civil
then taken as the most possible damage locations. structures.
110
Long-term structural health monitoring for Tamar Suspension Bridge
K.Y. Koo & J.M.W. Brownjohn

University of Sheffield, Sheffield, UK
P. Carden
Lloyd Register, UK
D.I. List & R. Cole

Tamar Bridge and Torpoint Ferry Joint Committee, Plymouth, UK
T. Wood
Cornwall Country Council, UK
ABSTRACT
For structural health monitoring purposes, it has

become extremely important to understand, model,
and compensate for the environmental variations on
both of static and dynamic characteristics of struc-
tures under ambient operation conditions. This paper
presents a long-term structural health monitoring on
Tamar Suspension Bridge to understand the environ-
mental effects of temperature, wind speed and traffic
volume. The monitoring system consists of two parts:
a static monitoring system for cable tensions, wind,
temperature measurements and a dynamic monitoring
system for deck and cable accelerations. The data-
driven stochastic subspace identification (SSI) method
was implemented for modal parameter identification
of the lower 5 natural frequencies. Environmental Figure 2. Long-term variations of the lower 5 natural
frequencies.
effects were investigated for a record of over 6 month
and it was found that cable tensions appeared to be
dominated by the temperature change while the fre-
quency variations were appeared to be affected by all
of the wind, the temperature and the traffic loading
with different contributions in each mode. Analytical
validation for the observations and identification of a
feasible model are underway.
Figure 3. Wind speed and Temperature effects on the natural

frequencies.
REFERENCES
Fish R., Gill J., Tamar suspension Bridge strengthening and
capacity enhancement. In Bridge Modification 2: Stronger
and Safer Bridges Thomas Telford, London, 1997, ed B
Pritchard.
Brownjohn, J.M.W. and Carden EP, Real-time Operational
Modal analysis of Tamar Bridge. IMACXXVI, Orlando,
USA, 47 Feb 2008.
Brownjohn, J.M.W., Pavic A., Carden P., Middleton, C.,
Modal Testing of Tamar Suspension Bridge, IMAC XXV,
Figure 1. Tamar suspension Bridge, Plymouth, UK. Orlando, USA, 1922 Februray 2007.
111
International collaborative research and education on smart sensors

and monitoring technologies
C.-B. Yun & H. Sohn

Department of Civil & Environmental Engineering, KAIST, Daejeon, South Korea
M.L. Wang
Department of Civil & Material Engineering, Northeastern University, Boston, USA
B.F. Spencer
Department of Civil & Environmental Engineering, University of Illinois at Urbana-Champaign, Urbana, USA
Y. Fujino
Department of Civil Engineering, University of Tokyo, Tokyo, Japan
ABSTRACT monitoring system capable of capturing the global

and local behavior of a structure, thereby leading
Bridges account for a large part of the capital invest- to improvements in structural health assessment. (2)
ment in constructing road networks and represent a US-Korea-Japan joint research on Jindo cable-stayed
key element in terms of safety and functionality. As bridge is presented, which aims to develop an inte-
such, informed bridge management is important to grated middleware services-based system for mon-
provide the public with timely transportation while itoring the state of cable-stayed bridge. It utilizes
maintaining a high level of safety. Similarly, detect- advanced multi-scale sensing, digital signal process-
ing critical damage at an early stage can reduce the ing, wireless communication, and damage diagnostic
costs and down-time associated with its repair. There- methods to take advantage of a smart sensors on-board
fore, smart monitoring systems are being advanced processing capabilities. (3) International education
with a new generation of sensing technologies and activities are finally introduced, which are organized
computational tools emerging from interdisciplinary ANCRiSST to enhance students understanding of the
researches between civil engineering and other engi- cross-disciplinary technological developments on the
neering disciplines. emerging subjects of smart structure technologies and
This study presents recent international collabo- structural health monitoring application. The sum-
rative activities on research and education for smart mer school is a 5 year program among Korea, the
sensors and monitoring technologies in the Asia- US, Japan, and China. The first one was hosted by
Pacific region. (1) A US-Korea joint collaboration KAIST, Korea in 2008. The second one was held at the
is first described exploring various emerging sen- University of Illinois at Urbana-Champaign in 2009.
sors on test-bed bridges in Korea. The heteroge- About 50 graduate students were attended in each
neous set of sensors such as wireless sensors, elasto- year. Students gained valuable experience of global
magnetic (EM) stress sensors, and piezoelectric active engagement in a culturally rich learning environment.
sensors is fused into a single comprehensive structural
Figure 1. International collaborative researches and educations.
112
Piezo paint-based smart tape sensor for bridge diagnosis
Y. Zhang, C. Zhou, Z. Li & C.C. Fu

Department of Civil & Environmental Engineering, University of Maryland, College Park, USA
C. Wang
Key Laboratory for Bridge and Tunnel Engineering of Shaanxi Province, Highway College, Changan University,
Xian, Shaanxi Province, China
ABSTRACT
Fatigue-induced crack monitoring and correspond-

ing retrofit actions will lead to a prolonged life and
enhanced reliability of structural systems. Manual
inspection is plagued with problems such as inten-
sive labor, time consuming process, and subjective
results; therefore it is unsuitable for rapid assess-
ment of structural conditions. In this regard, on-line
structural damage detection methods with automated
procedure are highly desired. Acoustic emission (AE)
is the elastic wave generated by sudden energy release
within a material, it provides real-time information on
damage progression within a structure.
Recently, the feasibility of using piezoelectric paint
to make broadband low-profile surface-mount AE sen-
sor has been investigated by Li and Zhang (2008a,b). Figure 1. Flexible piezoelectric paint sensor mounted on a
circular hollow section (CHS) steel member.
To achieve a low viscosity of fresh paint mixture
(thus suitable for spray painting) and flexibility of
cured paint, the vol-ume fraction of piezoelectric
ceramic powder in the paint is controlled below 50%. based on flexible piezoelectric paint for bridge mon-
The values of piezoelectric charge coefficient d33 for itoring. The use of piezopaint for ultrasonic signal
such piezoelectric paint formulations are comparable measurements is discussed along with a series of ultra-
to PVDF. The flexible piezoelectric paint uniformly sonic tests performed to verify the ultrasonic sensing
cures at ambient temperatures. Figure 1 shows flex- capability of piezopaint. A probabilistic framework for
ible piezoelectric paint tape (with 8 sensor nodes) bridge prognosis is also proposed. Issues associated
mounted on a CHS steel member. Compared with con- with field implementation of piezo paint sen-sors on
ventional acoustic emission sensors, piezopaint has a steel highway bridge in South Korea is discussed in
several advantages: (i) Flexible piezo paint sensor can this paper.
be applied to the near field locations in hot spots, even
directly atop the irregular surface of weldment. (ii) It
has a broadband frequency response in an ultrasonic
frequency range up to 1 MHz, which is due to the high REFERENCES
dielectric loss associated with piezo paint. (iii) The
Li, X. & Zhang,Y. 2008a. Feasibility study of wide-band low-
sensor has a low-profile and can be made at much lower
profile ultrasonic sensor with flexible pie-zoelectric paint,
cost. The piezo paint sensor operates in d31 mode and Smart Structures and Systems, 4(5): 565582.
thus the thickness of piezo paint sensor measures only Li, X. & Zhang, Y. 2008b. Analytical study of piezoelectric
a few millimeters. paint sensor for acoustic emission based frac-ture mon-
This paper describes the recent work on develop- itoring, Fatigue & Fracture of Engineering Materials &
ing a disposable low-profile acoustic emission sensor Structures, 31(8): 684694.
113
MS2: Monitoring & assessment of bridges using novel techniques
Organizers: A. Strauss & D.M. Frangopol
Analysis of the structural response to a moving load using

empirical mode decomposition
M. Bradley, A. Gonzlez & D. Hester

University College Dublin, Dublin, Ireland
ABSTRACT
The ability of empirical mode decomposition (EMD)

to detect structural damage as it happens has been
demonstrated in the literature (Huang & Shen 2005,
Xu & Chen 2004, Yang et al. 2004), but it has not been
tested for a structure with the measurements derived
from the crossing of a moving load. Unlike other avail-
able methods, the EMD method has the capability to
deal with nonstationary and nonlinear data. The EMD
method is based on the assumption that any data con-
sists of different simple intrinsic modes of oscillations.
Each intrinsic mode, linear or nonlinear, represents
a simple oscillation which has the same number of Figure 1. EMD damage detection algorithm applied to the
extrema and zero crossings. The oscillation is also midspan acceleration response: (a) Acceleration response,
symmetric with respect to the local mean. Each of (b) Filtered IMF1 with a cut-off frequency of 100 Hz.
these oscillatory modes is represented by a Intrinsic
Mode Function (IMF).
In this paper, the structure has been modeled as a noise, the spike has been found in the filtered second
one-dimensional simply supported discretized finite IMF more clearly than in the filtered first IMF. The
element beam. Damage has been simulated as a decay result of applying EMD to the simulated accelerations
in stiffness that covers a beam length related to the is shown in the figure below. Figure 1(a) corresponds
severity of the damage. The moving load has been to the midspan acceleration response of a 10 m span
assumed to be constant and driven at uniform speed beam with a 20% damaged section at 6 m from the
over the structure. Then, accelerations due to the pas- bridge support as a load crosses at 5 m/s. Figure 1(b)
sage of the load have been simulated at a number of shows the result of calculating the first IMF of Fig-
sections across the beam. The EMD method breaks the ure 1(a) with an intermittency frequency of 100 Hz.
acceleration signal into the relevant number of IMFs. In the latter, a peak associated to damage is detected
Then, an intermittency or cut-off frequency is applied at 1187 scans when scanning at 1000 Hz (these scans
to the first IMF. Results have shown EMD is able to represent 1187/1000 5 = 5.94 m, very close to the
detect the damage through a spike in the filtered first true damaged location at 6 m).
IMF of the transformed signal. The spike takes place at
the instant the load passes over the damaged location
regardless the measurement point. A sensitivity study REFERENCES
has been carried out varying observation point, velo-
city of the load (5, 10, 15 and 20 m/s), beam length Huang, N. E. & Shen, S.S. 2005. Hilbert Huang transform
and its applications. London, World Scientific.
(10, 15, 20 and 30 m) and severity of damage (5%,
Xu,Y., L. & Chen, J. 2004. Structural damage detection using
10% and 20% ratios of crack height to beam depth). empirical mode decomposition: experimental investiga-
In all cases, EMD has accurately located the damage. tion. Journal of Engineering Mechanics 130:12791288.
Finally, the acceleration has been corrupted with noise Yang, J.N., Lei, Y., Lin, S. & Huang, N. 2004. Hilbert-Huang
to simulate real measurements with signal to noise based approach for structural damage detection. Journal
ratios of 20, 10 and 5. In the presence of high levels of of Engineering Mechanics 130:8595.
117
Measurement tool kit for corrosion and defect monitoring

of bridge tendons
H. Budelmann, A. Holst & H.-J. Wichmann

iBMB, Department of Civil Engineering, University of Braunschweig, Germany
ABSTRACT
One of the most important tasks and challenges of

Life-Cycle-Engineering and bridge management is
to ensure the safety, serviceability and robustness of
the prestressed members during service life. Here,
corrosive and mechanical influences on the steel rein-
forcement may strongly affect the long-term perfor-
mance and durability of internal and external tendons
of concrete structures. On this account, the condi-
tion assessment of prestressed members should be
addressed to the detection of existing damages, such
as grouting faults, corrosion defects and fractures of Figure 1. Schematic diagram of a possible corrosion mon-
the steel elements. itoring strategy for prestressed elements, extended acc.
But unfortunately for grouted interior or for to Holst et al. (2006); NDT/MDT- non/minor destructive
ungrouted exterior tendons no accurate and reliable techniques.
non-destructive methods are available today for the
monitoring of the corrosion state, for the localization A novel type of electrochemical chloride electrode
of fractures or for the measurement of the actual tensile sensor for the measurement of the free chloride
force of tendons. content in the pore solution of concrete and
Several novel techniques for force measurement The electromagnetic resonance measurement
resp. for corrosion and fracture identification at bridge method (frequency domain reflectometry, FDR) in
tendons have been developed by the authors within the combination with electromagnetic field strength
framework of the Collaborative Research Center SFB investigations for the detection and localization of
477 Structural Health Monitoring at the University prestressing steel fractures and of other defects
of Braunschweig in the last decade. based upon microwave technology.
Based upon a the three-level monitoring strategy
(Fig. 1) these redundant NDT/NTE-methods offer an The functionality as well as the application of the
efficient monitoring tool kit for tendons of P/C-bridges distinct methods, showing different stages of develop-
as well as for ground anchors and ropes. ment, were tested and verified under field-conditions,
The main focus of this paper lays on the practical e.g. at a 18 m long prestressed trial concrete bridge and
testing and verification as well as on the specification at first applications at real structures. Based upon this,
of achieved improvements and further developments the pros and cons of the methods are discussed.
of the innovative sensing and monitoring techniques,
listed below:
REFERENCE
Magnetoelastic coiled sensors for the force mea-
surement and material defect detection of pre- Holst, A., Wichmann, H.-J., Hariri, K. and Budelmann,
stressed tensile elements. H. 2006. Monitoring of Tension Members of Civil
Structures New Concepts and Testing. Proc. of the 3rd
Smart calibration-free filament sensors for corro-
Europ. Workshop on SHM, 57 July 2006, Granada, Spain,
sion monitoring consisting of a single or several 117125.
parallel arranged, 0.065 to 0.5 mm thin steel wires,
working as watch-dog-sensors for the identifica-
tion of steel corrosion initiation and progress in the
concrete cover resp. injection grout of ducts.
118
Field application of innovative scour monitoring techniques for bridges
K.C. Chang & C.C. Chen

Department of Civil Engineering, National Taiwan University, Taipei, Taiwan
Y.B. Lin, L.S. Lee & S.C. Wong

National Center for Research on Earthquake Engineering, Taipei, Taiwan
ABSTRACT
Following the global climate change, flood disasters

news has been reported worldwide. Consequently,
Taiwan is well known for its famous topography fea-
tures that the central mountain range bisects the island
from north to south leading to most rivers have short
courses and rapid streams and is subject to several
typhoons and flood events each year during the sum-
mer and fall seasons. Flood water can intensify the
scour menace to bridges since the unsteady nature of
the flow. Furthermore, floating or submerged debris
induced by flood may deepen the scour depth of bridge
foundations. Thus the hydraulic causes of bridge fail-
ure have focused engineering attention on the bridge
scour problem.
However, in-situ bridge scour monitoring is still one
of the rugged works for field application researchers. Figure 1. Layout of the field scour monitoring system.
Challenges include bridges under condition of the
uncertainty, harsh environment of system servicing
events, long-term performance of sensors and pack- ready for presently application research. The purpose
aging modules, ease of installation and cost of deploy- of the field deployment was to evaluate the prelimi-
ment and maintenance. In this paper, we focus on nary design in terms of durability of packaged sensors,
the use of innovative technologies for monitoring reliability of communication under real operational
bridge scour processes. This study applied the inno- environment. With the experience of the success of
vative scour monitoring techniques which have been this system settled in field, further study could be con-
designed and developed in the laboratory condition ducted with acquired field data for scour condition
to field bridge sites. Two developed scour sensing interpretation. More appropriate countermeasures to
techniques were combined into one scour moni- protect the bridge foundation and reduce scour effect
tor tube, which comprises optical fiber sensors and could be drawn up as well.
micro-electro-mechanical systems based sensors and
designed to be installed on adjacent bridge foundations
for scour monitoring. The layout of the field deploy- REFERENCES
ment of bridge scour monitoring system proposed
Lin,Y. B., Chen, C. C., Chang, K. C., Chern, J. C. and Lai, J. S.,
showed in figure 1.
2005, Real-time monitoring of local scour by using fiber
The system can detect the depth of scour near bridge bragg grating sensors, Smart Materials and Structures, 14,
foundation under flood water, and provide reciprocal 664670.
verified result for the monitoring information with two Lin, Y.B., Lai, J. S., Chang, K. C. and Li, L.S., 2006, Flood
sensing techniques included. The system proposed in scour monitoring system using fiber brag optical fiber
this paper has been installed on in-situ bridges and be sensors, Smart Material and Structures, 15, 19501959.
119
Performance evaluation of bridge seismic bearings based

on in-situ quick-release tests
S.S. Chen & I.-S. Ahn

Department of Civil, Structural and Environmental Engineering, University at Buffalo,
Buffalo, New York, USA
ABSTRACT
A nonlinear model-based system identification method

is proposed, formulated, and implemented. The pro-
posed method is applied to quick-release field experi-
ments on a three-span continuous base-isolated bridge
in order to investigate aging and temperature depen-
dent effects of lead-rubber bearings.
A quick-release (or pull-back) test is a free vibra-
tion test method where lateral forces are applied to
the superstructure and released quickly to introduce
a free vibration to the bridge. In the present study,
transverse and rotational rigid-body motions of the
bridge superstructure are formulated into two degrees-
of-freedom dynamic governing equations. To model
hysteretic behavior of lead-rubber bearings (LRBs),
the Menegotto-Pinto model is used. Among several
quick-release experiments the authors investigate: i)
aging effects of LRBs by utilizing two experimental
data sets taken several years apart, and ii) temperature
effects of LRBs by utilizing two experimental data sets
conducted in warm and cold weather.
For given dynamic governing equations and an
input excitation (or free vibration), the resultant
motion is a function of the parameters in the gov-
erning equations. System identification is a process
to determine values of those parameters in a model
that can reproduce the measured responses. Therefore,
system identification frequently becomes an optimiza-
tion problem to find optimal values of the parameters Figure 1. Two-phases System Identification.
to minimize the difference between measured and
reproduced responses. the superstructure can successfully capture free-
Because the existence of a unique solution cannot vibration motion in quick-release tests.
be guaranteed, system identification is referred to as an Regarding aging effects, increases of the pre-
ill-conditioned or an ill-posed problem. Another diffi- yielding stiffness and the post-yielding stiffness are
cult feature of the present problem is the large number observed.
of parameters to be identified. The proposed system Regarding temperature dropping effects, the
identification, therefore, is divided into two phases as decrease of energy dissipation capacity and the
in Figure 1. increase of the pre-yielding stiffness are observed.
Results of applying the proposed method are sum- Current practice does not consider changes in the
marized as follows. pre-yielding stiffness; however, results from sys-
tem identification indicate its effects on response
The two degree-of-freedom governing equations
are significant.
for transverse and rotational rigid-body motion of
120
Experimental assessment of prestressing force in concrete bridges
J. Ciesla
Road and Bridge Research Institute, Warsaw, Poland
ABSTRACT
Assessment of the prestressing force of existing pre-

stressed concrete bridge superstructure is a main object
of the paper. It has been assumed; that proper level of
prestressing force is one of the most essential factors
from the structural safety point of view.
They are some signs that might prove excessive
reduction of prestressing, e.g. residual deflection of
span or visible cracks on the surface of span girders.
But more exact evaluation of prestressing force needs
some authentic and empirically verified procedures.
Figure 1. Flowchart of assessment of safety of prestressed
Suggested experimental procedure of evaluation of concrete structure.
prestressing force concerns such aspects, as:
Analysis of stiffness of girder under applied proof In case of existing prestressed concrete bridge,
load, based on load-deflection relationship for generally we may have two possibilities:
uncracked element. The load-deflection curve is they are no cracks in all of essential elements of
generated using experimental influence surface of bridge,
deflection for prestressed element. cracks exist in some of its elements.
Moment of crack formation in prestressed element,
when we are able to cause cracks under load. In that In case of uncracked structure the analysis of stiff-
case assessment is based on the comparison of the ness under applied load is the most important resource
real crack moment with calculated value. of information about prestressing force.
Comparison of crack pattern of prestressed con- In case of cracked structure comparative analysis
crete bridge element and cracks generated in its concerns the crack pattern of real structure and the
structural model, in case of cracked structure. The map of stresses in structural model of the bridge may
general analysis method corresponds to the best be used.
match between structure and its structural model.
If the prestressing force is below design value, in REFERENCES
most of cases the reason of the lack of prestressing is
the key to assessment of the safety of the structure. Ciesla, J. 1996. Assessment of the safety of post-tensioned
Improper level of prestressing may be result of some concrete bridges by comparative analysis. FIP Symposium
negligence under construction, excessive losses of pre- on post-tensioned structures, London.
Ciesla, J., Lagoda, M. & Olaszek, P. 2008. Special tests of
stressing forces and also corrosion, which is the most
two post-tensioned concrete viaducts. Proceedings of the
dangerous case for the safety. Fourth International Conference on Bridge Maintenance,
In the Figure 1 it has been suggested general scheme Safety and Management, Seoul.
of procedure for the assessment of the safety of bridge
span based on the evaluation of real prestressing force
in the structure.
121
Assessment and strengthening of prestressed damaged beams
L. Della Sala
Universit della Basilicata, Potenza, Italy
R. Cerone
Studio di Ingegneria, Melfi (PZ), Italy
M. Franzese
A.N.A.S. S.p.A., Potenza, Italy
ABSTRACT
The paper describes the evaluation of the actual con-

ditions, the repairing design criteria and procedures,
and the final validation tests carried out on high dam-
aged prestressed beams of the viaduct Fiumara of
Tito on the highway RA005 Sicignano-Potenza in
Basilicata Italy. Figure 1. Repaired beam.
The structure, built in 1970, was affected by serious
damages involving the prestressed beams. In particu- cables and in a FRC continuous reinforcement of the
lar, a state of diffused cracks and uncontrolled water bottom of the beams, has been designed with the aim
flows has induced the partial corrosion of the steel of restoring the original structural performances of the
prestressing cables, causing a reduction of the bearing beams.
capacity of the decks. The tensioning procedure of the external cables of
In a first phase several tests have been carried out on each beam was then monitored by strain-gages, load
the beams of the viaduct with the aim of assessing their cells and by measuring the vertical deflection using
structural conditions. First of all, ware carried out spe- laser optical levels.
cific measurements for the identification of the cables The reinforcing system applied to the beams
layout using electromagnetic and radar detectors. allowed the increasing of the global safety factors and
Then were performed direct and borehole visual the reduction of the risk of cracks opening, protecting,
inspection of the cables. In many cases resulted non- in this way, the existing cables from further corrosion.
corroded strands but absolute absence of injection The validation of the repairing works was finally
mortar. performed by static load acceptance tests.
For evaluating the stresses acting on concrete due
to the residual prestressing force and to the dead loads
were performed some release tests both on concrete REFERENCES
and on the steel strands. The results of this test have
pointed out very low stress values in concrete till about Bray, A. 1965. Estensimetri elettrici a Resistenza. Strain
1.5 MPa indicating an important loss of prestressing Gauge technique. Roma (IT).
force near the intrados of the beams. Lin. T.Y. & Burns N.H. 1982. Design of prestressed concrete
structures. New York (USA).
Release tests on the prestressing strands have also Suntharavadivel, T.G. & Aravithan, T. 2005. Overview of
been performed in order to estimate the residual ten- external post-tensioning in bridges. Southern Engineering
sion acting in the cables. This tests have confirmed the Conference.
loss of prestress since have been measured values till Snchez-Beltia, S. & Schueremans, L. 2008. The hole drilling
600 MPa that are about the half of the initial value of technique in site stress measurement on the pier of the
1000 MPa. Saint Jacobs church in Leuven. Sacomatis Conference.
On the basis of the in situ test results a specific Varenna (IT).
reinforcing system, consisting in external unbounded
122
Development of rapid, reliable, and economical methods for

inspection and monitoring of highway bridges
J.D. Fasl, T.A. Helwig, S.L. Wood, V.A. Samaras, A.A. Yousef & K.H. Frank
University of Texas at Austin, Austin, TX, USA
D.L. Potter
National Instruments Corporation, Austin, TX, USA
R.E. Lindenberg
Wiss, Janney, Elstner Associates, Inc., Northbrook, IL, USA
ABSTRACT The use of an advanced monitoring system will

enhance detection of distress without the need to mobi-
Highway bridges are vital links in the transportation lize an inspection crew or disturb traffic, while giving
network in the United States, providing the public with transportation officials the tools to better allocate
routes for daily commutes and businesses with the inspection resources.
infrastructure needed to supply goods and services. A LabVIEW program based on simplified rainflow
Identifying possible safety problems in the national counting was developed to assist in diagnosing and
inventory of more than 600,000 bridges is generally estimating the fatigue life of a bridge. The program
accomplished through labor-intensive, visual inspec- was implemented on a bridge for nearly two weeks
tions, which are required at least once every two years. and compared to the results of the onboard rainflow
Pending legislation would increase the inspections for algorithm on a CR5000 data acquisition system. Data
fracture-critical bridges to once a year and likely strain was further analyzed using the LabVIEW program to
DOT resources that are already stretched to the limit. understand the fatigue implications using Palmgren-
This paper outlines ongoing research sponsored by Miners rule. This paper will present the results of that
NIST to improve inspection practices by providing the field study.
technology and methodology for real-time monitoring Understanding the results from a rainflow analy-
of bridges. sis is beneficial for evaluating the health of a bridge
Past monitoring systems for bridges have been and its components. Due to the similarity between the
time-consuming to install and unreliable in operation. CR5000 bin data and the bin data from the LabVIEW-
However, with improving technology, the ability to based program, the LabVIEW-based program was
develop resilient systems that are simple to install and validated for future use. Also, changing the time period
maintain are possible. A likely candidate for long-term for outputting bin results did not significantly affect
monitoring are fatigue sensitive areas in fracture- the results for the time periods investigated (0.5 hour,
critical steel bridges. The concentration of the study 2 hours, 12 hours, and 24 hours). Part of the small
is on developing a long-term monitoring system that variation between bin results for the different periods is
is economical and ruggedable to accommodate the due to the negligible influence of temperature-induced
severe temperature ranges, humidity fluctuations, and drift for the installed sensors. However, the similar-
natural weathering processes that bridges experience. ity between the results also suggests that 0.5 hour
To make installation easier, the focus is on wireless (30 minute) segments can be used for rainflow anal-
sensors designed to continuously monitor fracture- yses without counting too many cycles in the close
critical bridges with a targeted 10-year battery life. out period and skewing the results. Finally, having
The sensor nodes will be capable of supporting multi- 30 minute segments reduces the memory requirements
ple sensors with sufficient computing power to process of the wireless devices while minimizing the effect of
raw sensor data and send notifications off-site when a temperature-induced sensor drift.
threshold level of damage occurs.
123
Wireless sensor performance monitoring of an innovative bridge design

in New York State
M.V. Gangone, M.J. Whelan, K.D. Janoyan, L. Minnetyan & T. Qiu

Clarkson University Department of Civil and Environmental Engineering, Potsdam, New York, USA
ABSTRACT
With many of todays bridge superstructures deterio-

rating at such a rapid rate, there is a need to develop
new structures that can last longer and stand up to
the effects of both greater loading and harsh environ-
mental conditions. As innovative materials continue to
find their way into civil engineering design, monitoring
their performance is of utmost importance. Discussed
in this paper is the deployment of a custom designed
wireless sensor system on an innovative bridge super-
structure in NewYork State. The structure utilizes glass
fiber reinforced polymer (GFRP) bridge deck panels
Figure 1. Typical strain time series for an exterior girder.
as a replacement to a deteriorated steel grated bridge
deck. A wireless sensor system developed in the Lab-
oratory for Intelligent Infrastructure and Transporta-
tion Technologies (LIITT) providing acceleration and
strain measurements at critical locations of the super-
structures is deployed on both bridges. The wireless
system provides independent conditioning of acceler-
ation, strain and temperature measurements with high
rate, real-time lossless transmission capabilities.These
measurements serve to provide static and dynamic load
testing parameters and modal characteristics of the Figure 2. Mode shape estimates as reconstructed from
output-only system identification.
superstructure under both ambient excitation and ser-
vice loads. The results offer performance evaluation as
well as verification of design methodologies employed Multiple tests were completed in each lane to vali-
for the newly constructed bridges. date the data. A strong correlation is seen among the
Testing was completed over a two day period where results.
both strain and acceleration measurements were cap- Modal analysis from ambient loading conditions
tured under forced and ambient loading conditions. was completed on day two. Peak excitations from
Strain transducers were deployed at the midspan of vehicular traffic rarely exceeded 10 mg. Nonetheless,
each girder in addition to the locations near the abut- structural poles within the average power spectrum
ment of the center girder. Neutral axis locations, were clearly identified. Output-only system identi-
transverse load distribution factors and end fixity were fication was performed using Stochastic Subspace
detected based on loading from a calibrated H-truck Identification. The first 8 lower order modes are pre-
in three separate loading lanes. The results indicated sented in the paper, with the first 2 shown in figure 2
minimal composite action between the FRP panels and above.
steel girders. A typical strain time history of top and The results from the deployment provide indication
bottom flange measurements at an exterior girder can of the performance of the bridge superstructure. With
be seen in figure 1. Strain readings at the abutments testing completed near the beginning of the bridge ser-
indicated a strong level of fixity present in the integral vice life, the data obtained can be compared to future
abutment design. Load distribution factors suggest that testing results for monitoring the long term health and
the vehicle demand shed within the superstructure. performance of the FRP design.
124
The benefit of monitoring for bridge maintenance
R. Geier
Schimetta Consult ZT GmbH, Vienna, Austria
ABSTRACT
Apart from conventional bridge control, more and

more measurements and monitoring concepts for
structures have been carried out in the last years meant
at giving information on the way the condition of bear-
ing structures changes with the passing of time and at
supporting bridge investigation by means of recording
objective data. Still, these technologies did not reach in
the past the desired degree of application as they tried
to position them primarily on the basis of vibration Figure 1. View of the Erdberger Bridge.
measurements with only few sensors. This approach
however could not find any acceptance with bridge During a recent bridge inspection a relatively poor
owners as influences, like temperature, on the reso- condition of the bearing structures with lots of damage
nance frequency are far greater than the influences of was revealed. The bridge authority decided prior to the
damage. general rehabilitation of the structure to monitor the
Recent experience has shown that monitoring con- evolution of the bridge condition.
cepts need specific change if they are to be used more The system observes automatically horizontal
often. The following theses are at the centre of the movements of the arch foundations by laser and the
approach to applying monitoring presented here: connection of old, existing mechanical extensometer
sensors, bridge temperature, behavior of backstay ten-
(1) Global methods do not permit early diagnosis of
dons and overall changes in length. The design and
damage at acceptable cost.
results of the monitoring system are described in the
(2) Data-based investigation is ideally suited to
full paper.
observing known problems or damage and
changes in these over time.
(3) Objective data are collected as input parameters
2 SUMMARY
for further investigations.
For each thesis benefits of monitoring may be The processing of monitoring data is an efficient and
achieved. As an example the implementation of the reliable method of monitoring the safety of an struc-
second thesis is shown on a famous bridge in the full ture. Through the connection of the unit to the internal
paper. fiber optics network of ASFiNAG, information on the
load bearing behavior can be timely provided and if
need measures can be taken.
1 MONITORING SYSTEM Analysis done on the basis of performance data is
best suited for observation of known problems or spe-
The Erdberger Bridge with a total length of 147 m and cial questions like in our example. This application is
a width of 42,30 m built in the years 1968 to 1971 extremely interesting given that in case of problems
representing a milestone in bridge engineering due the bridge owners consider elaborate measuring and
to the design concept based on a reinforced concrete analysis to be very reasonable in order to keep records
shell. of the evolution of the condition of certain parameters.
125
Monitoring method for curved concrete bridge girders using long-gauge

deformation sensors
B. Gliic
Princeton University, Princeton, NJ, USA
D. Inaudi & D. Posenato

SMARTEC SA, Manno, Switzerland
ABSTRACT
As a consequence of their shapes, curved concrete

bridge girders have complex internal force distribu-
tions and it is recommended to verify the real structural
performance of curved girders upon construction and
to monitor their behavior in long-term. A structural
health monitoring method for curved concrete girders,
based on the use of various to pologies of long-gauge
deformation sensors combined with inclinometers and
temperature sensors, is presented in this paper.
The method is loosely model-driven the charac- Figure 1. Sensor network for curved beam, plan view,
teristic cross-sections for monitoring are first selected sensors not to scale.
based on a generalized structural analysis of the girder,
and then each cross-section is equipped with a sen-
sor topology or a combination of sensor topologies,
that can characterize in the best manner the expected
influences.
A parallel topology is used for the monitoring
of cells subject to bending, assuming that Bernoulli
hypothesis is valid. It consists of two sensors with
equal gauge lengths, parallel to the elastic line of the
beam and installed at different levels of the cross-
section. A crossed topology consists of two crossed
sensors installed with a predefined angle with respect
to the direction of normal strain lines. The aim of this
topology is to detect and quantify the average shear
strain in the plane of the sensors.
Finally, the results of monitoring of each cross-
section are linked together in order to assess the Figure 2. Average curvatures in cells C and E.
global structural behavior. The method allows for
global monitoring of axial strain, horizontal and ver-
tical curvature changes, torsion, average shear strain, curved post-tensioned bridge box girder equipped with
rotations in both vertical planes, and deformed shapes fiber-optic sensors. Important parts of the structure
in both vertical and horizontal planes. The sensor life such as construction and post-tensioning are reg-
network resulting from the method is given in Figure 1. istered, analyzed and presented. An example of results
The method is illustrated and its performance is is given in Figure 2, where numbers 1 to 6 represents
evaluated through a practical example: a 36-meter long various construction works.
126
Multiple sensor subsurface condition assessment of reinforced concrete

bridge decks
D.R. Huston, J. Cui, D. Burns & D. Hurley

University of Vermont, Burlington, VT, USA
R. Arndt
Federal Highway Administration, Office of Infrastructure R&D, McLean, VA, USA
ABSTRACT
This paper reports on a study aimed at accu-

rately assessing the subsurface damage conditions
of reinforced concrete bridge decks with multi-
ple non-destructive evaluation (NDE) techniques.
Information gained from such evaluations may signif-
icantly enhance the effectiveness of deck maintenance
activities. The present state-of-the-art is that multiple
NDE techniques are available, but that the accuracy
and reliability of the methods are not guaranteed.
NDE techniques generally employ electromagnetic,
electrochemical or elastic wave principles. The sophis-
tication ranges from the simple chain drag, to compli-
cated multi-channel radar and ultrasound instruments.
Recent field studies comparing NDE methods with
each other and ground truth data have indicated a Figure 1. Ground penetrating radar assessment of bridge
relatively high degree of variability and disagree- deck.
ment among the sensors. The foci of this study are
produce an automated easy-to-use multisensor bridge
threefold: 1. Compare the sensing data from multi-
deck assessment system.
ple sensor methods applied to bridge deck specimens
presently undergoing accelerated degradation in the
laboratory. The sensors include multichannel ultra- REFERENCES
sound, ground penetrating radar (GPR), anode ladder,
inductive rebar heating with infrared imaging and half- Hall, D.L. McMullen, and Sonya A.H. (2004), Mathemati-
cal Techniques in Multisensor Data Fusion, Cambridge,
cell electrochemical potential. The accelerated degra- Artech house, Inc.
dation combines salt bath cycling with mechanical Johansson, E.M. and Mast, J.E. (1994), Three-Dimensional
loading. 2. Attempt to understand the effect of differ- Ground-Penetrating Radar Imaging Using Synthetic
ent stages and type bridge degradation on the sensor Aperture Time-Domain Focusing in SPIE 2275 Adv.
signals. In particular, reinforcing corrosion, crack- Microwave and Millimeter-Wave Detectors.
ing around rebars and delamination are different, but Johnsen TH, Geiker MR, Faber MH. (2003) Quantifying
related damage patterns. These different damage types Condition Indicators for Concrete Structures Concrete
can produce different effects on the sensors. For exam- Intl, Dec
ple, corrosion and water-chloride contamination may Khossrow, B. (1986) Evaluation of Half-cell Corrosion
Detection Test For Concrete Bridge Decks Washington
produce a strong absorption of radar waves, but pro- State Department of Transportation WA-RD 95.1
duce relatively modest changes in ultrasound and chain Maser KR, Kim Roddis WM. (1990) Principles of Thermog-
drag tests. Conversely, air-filled delaminations are dif- raphy and Radar for Bridge Deck Assessment J Trans-
ficult to detect with standard GPR, but are readily portation Eng, V 116, N 5, p 583601, Sept/Oct
detected with chain drag and impact-echo. 3. Explore Scott M, Rezaizadeh A, Moore M. (2001) Phenomenology
methods of fusing the electrochemical, electromag- Study of HERMES Ground Penetrating Radar Technology
netic and elastic wave data based on degrees of belief for Detection and Identification of Common Bridge Deck
in sensor information to produce an enhanced assess- Features FHWA-RD-01-090 USDOT Federal Highway
ment of subsurface conditions. The ultimate goal is to Administration
127
Output-only substructural identification for local damage detection
C.G. Koh & T.N. Thanh

Department of Civil Engineering, National University of Singapore, Singapore
ABSTRACT identify structural parameters and input excitation

force within the substructure. This is done ingeniously
With declining state of aging infrastructures such as by adopting a predictor-corrector algorithm to correct
bridges and highway systems, the ability to detect the output responses of internal acceleration, velocity,
any damage at the early stage would help to reduce and displacement that are predicted using numerical
maintenance costs and increase safety to the public. In integration. Instead of using the predicted (or mea-
this regard, there has been increasing interest in R&D sured) internal responses, the corrected responses are
on structural health monitoring. For large and com- adopted in the numerical integration to predict the
plex structural, however, this is a difficult task. Not responses in the next time step, resulting in reducing
only a large number of sensors are needed, but also the accumulation of errors due to measurement noise.
formidable computational challenges are encountered To demonstrate the performance of the proposed
in order to achieve satisfactory results for the pur- strategy, numerical study is carried out for a long con-
pose of damage detection. The task becomes much tinuous truss bridge (as shown in Figure 1). Using
more challenging if input force measurement is not acceleration response contaminated by 10% noise in
possible and only acceleration data at certain loca- the numerical simulation, stiffness parameters of a
tions are available. To this end, this paper presents an substructure of a two-span continuous truss bridge are
output-only substructural identification for detecting identified with mean error of less than 5%. In addi-
damage within a large structure based on acceleration, simulated damages in this substructure with two
tion measurements to infer changes in its stiffness damage extents are successfully detected and quan-
parameters. tified with mean error of less than 3%. The results
The study uses an improved genetic algorithm show that the proposed strategy can be used for local
method to identify the stiffness and damping parame- damage detection. The proposed substructure strategy
ters of substructure. A key advantage of the proposed is further substantiated by an experimental study of a
substructure strategy is its ability to simultaneously laboratory-scale steel frame model.
Figure 1. Long-span truss structure: (a) full structure; (b) substructure.
128
Steel beam fatigue life prediction using acoustic emission amplitude

histograms and backpropagation neural networks
A. Korcak, J. Suleman, F.F. Barsoum & E.v.K. Hill

Multidisciplinary NDE Group, Embry-Riddle Aeronautical University, Daytona Beach, FL, USA
ABSTRACT is truly powerful is that even signals as small as

microstructure slipping or cracking would be picked
This paper applies nondestructive evaluation (NDE) up by using a piezoelectric transducer. The transducer
methodology to predict fatigue life in structural steel. senses the vibrations on the surface where it is mounted
Specifically, it applies acoustic emission (AE) nonde- as result of the acoustic source. These signals serve as
structive testing (NDT) to monitor the development of parameters for the Neural Network.
fatigue crack growth in bridge structures simulated in
this research as transversely loaded beams. A back-
propagation neural network (BPNN) is then utilized 1.2 Artificial neural network
to perform fatigue life prediction. Tests were done on The backpropagation neural network (BPNN) is
fourteen transversely fatigued S4 7.7 I-beams made widely used in predictive learning, and in cases in
out of A572-G50 steel. Although the I-beams were which a substantial amount of data is available with
tested under the same conditions, the material and complex relations.
slight geometry differences between the specimens
still produced significant variations in the fatigue life
of 13.6% coefficient of variation. The networks predic- 2 CONCLUSIONS
tion based on the data for first, second, and third quarter
of the fatigue lives was 18% and 16% and 5% error, The research presented here is a continuation of
respectively for training on five specimens and test on research for prediction of fatigue lives for tension
five. This trend shows good promise as the prediction members where even with variability of different stress
accuracy increases with the stage of fatigue life. levels the networks still predicted with less than 12%
error. For the I-beams although the loading conditions
were kept the same slight material differences created
1 INRODUCTION a 13.6% coefficient of variation in the fatigue lives.
The networks prediction based on the data for first
As the structures age and funding for new infras- and second quarter of the fatigue lives was 18% and
tructure decreases there is a need to maximize the 16% error for training on five specimens and test on
use of the existing infrastructure. Safety however is last five respectively. The prediction based on the third
a primary factor because of this strong and reliable quarter of life had error of about 5%. This is because
evaluation and prediction techniques are necessary. the crack propagation is more predictable than crack
Fatigue failure investigation of structural steel has been initiation. For many applications such as monitoring
well researched during the last several decades. The of bridges it is actually preferable that the prediction
universal consensus is that when it comes to health is more accurate for the later part of life as that is what
monitoring of steel structures, a NDE methodology is would happen in the field.
desirable. AE is among the most suitable techniques
when used in conjunction with other NDT methods
such as visual inspection.
REFERENCES
1.1 Acoustic methodology Suleman, Jamil, Andrej Korcak, Fady F. Barsoum, and Eric
V.K Hill. Acoustic Emission Monitoring and Neural Net-
Acoustic emissions function on the basis that sounds work Fatigue Analysis of Steel Bridges. Proc. of FCAAP
can be transmitted throughout a material. Where AE Annual Technical Symposium, Orlando, 2009.
129
Health monitoring of concrete bridges: model simulations of pre-stressed

beams under static environmental loading based on experimental data
F. Lanata
Department of Civil, Environmental and Architectural Engineering, University of Genoa, Italy
Institute de recherche en Gnie Civil et Mcanique (GeM), Universit de Nantes, France
ABSTRACT Work supported by the European Community and FEDER

founds within the duratiNet Interreg (Atlantic space, project
As recent literature has highlighted, the effects of N N 2008-1/049).
environmental parameters on the sensitivity of the
proposed damage detection techniques have to be
deeper investigated. For this reason an experimen-
tal monitoring program has been performed in order
to follow the long-term strain measurements of two
pre-stressed concrete beams (one undamaged and the
other one progressively damaged) placed not in a
controlled laboratory environment but outdoor under
environmental conditions. This work is focused on
the development of the numerical simulations based
on these experimental strain measurements. Particu-
lar attention has been focused on the comprehension
of the reliability of the performed laboratory experi-
ment. Damages of different intensities and locations
have been introduced into the model.
Simulated strain histories show that the models well
catch the temperature-induced variations in strain, and
they are very well correlated to the temperature varia- Figure 1. Simulated and measured strains (Sn11) and cor-
tions (Fig. 1). By the way, a deviation can be observed responding measured temperature (T5).
between the simulated and the measured responses.
The strain versus temperature plot for both responses catching the behavior of a simply supported beam
shows that the slope of the relationship is the same under environmental conditions. A model really rep-
for the simulated and measured responses, so to con- resentative of the mechanical system is hard to realize,
firm the fact that the model well catches the thermal in particular when complex structures have to be mod-
behavior of the beam. However, the relationship for the eled. A non model-based approach could give better
real structure shifts in the first months of the monitor- performances when a long reference period is avail-
ing before reaching a stable trend. As these shifts are able. Furthermore, the computation of the mechanical
usually associated to changes in the mechanical sys- strain after removing the thermal strains seems to be
tem, that means that something not controlled occurred promising because it could help to detrend the strain
in both beams in the first period of the monitoring, histories from environmental events.
probably a change in the support conditions or a gen-
eral settlement of the structure. Furthermore, the plot
shows a thermal hysteresis in the measured structural REFERENCES
response indicating that the relationship between strain Del Grosso, A., Lanata, F., Posenato, D. & Mercalli, A. 2010.
and temperature is non linear, as the structure has a Validation of an SHM procedure for concrete bridges
thermal inertia that the current model is not able to based on static strain records. IABMAS; Proc. 5th Int.
describe. Conf., Philadelphia, 1115 July 2010.
The observed discrepancy highlights the doubts Kim, J.T., Ryu, Y.S., Cho, H.M. & Stubbs, N. 2003. Dam-
related to the use of model-based approaches in dis- age identification in beam-type structures: Frequency-
closing damage. This work shows the difficulties based method vs. mode-shape-based method. Engineering
encountered in the construction of a numerical model Structures, 25: 5767.
130
Jointless bridge: Determination of fracture mechanical parameters values

for nonlinear analysis
D. Lehk, Z. Kerner & D. Novk

Faculty of Civil Engineering, Brno University of Technology, Brno, Czech Republic
ABSTRACT
For complex stochastic nonlinear analyses of bridges

(old or new ones) the following steps are required:
1. Laboratory or in-situ tests to determine values
of basic fracture mechanical parameters and their
statistics.
2. Inverse analysis to identify values of other param-
eters needed in numerical model.
3. In case of old bridges the damage detection (posi-
tion and magnitude) using structural health moni-
toring data. Figure 1. Selected specimen after three-point bending test;
4. The development of advanced numerical computa- detail of crack path (right) [photo: B. Kucharczykov].
tional model based on stochastic nonlinear fracture
mechanics and knowledge of possible damaged neural network based method (Novk & Lehk 2006).
parts and material parameters from laboratory tests Both techniques provided results which are close to
and identification. each other including basic information on variability
5. Stochastic nonlinear analysis to assess reliability, (COV). Inverse analysis provided additionally values
risk and lifetime of a structure. of tensile strength of concrete. Results efficiently serve
as input data for stochastic nonlinear simulation of
The proposed paper is focused on first two steps studied bridge (Podrouek et al., 2010).
of the whole procedure on the determination of
fracture mechanical parameters values of concrete
used for casting of newly built Jointless Bridge in ACKNOWLEDGEMENT
Austria. For that purpose six specimens of nominal size
100 100 400 mm with the edge notch of the depth This outcome has been achieved with the financial
about 1/3 of the depth of the specimen in the centre support of the project No. 1M0579 (research centre
of the beam were tested in three-point bending (3PB) CIDEAS) from the Ministry of Education of the Czech
configuration. Loading span was equal to 300 mm. Republic and by the research grant Eurostars E!4351
Example of the tested specimen is in Figure 1. After RLACS.
three-point bending test, two broken parts of each spec-
imen were cut to nominal size 50 50 50 mm using
diamond saw and than subjected to compression tests. REFERENCES
Subject of interest were following parameters of
concrete: specific fracture energy, modulus of elastic- Karihaloo, B.L. 1995. Fracture mechanics of concrete. Long-
ity, tensile and compressive strength, effective crack man Scientific & Technical, New York.
Novk, D. & Lehk, D. 2006. ANN Inverse Analysis Based
elongation, effective fracture toughness and effec-
on Stochastic Small-SampleTraining Set Simulation. Eng.
tive toughness. Determination of parameters was Application of Artificial Intelligence, 19, 731740.
done using two techniques (i) direct evaluation Podrouek, J., Novk, D., Strauss, A. 2010. Jointless Bridge:
from experimental loaddeflection (ld diagram) by Reliability Assessment. The Fifth International Confer-
effective crack model and work-of-fracture method ence on Bridge Maintenance, Safety and Management
(Karihaloo 1995); (ii) inverse analysis using artificial IABMAS 2010, Philadelphia, USA.
131
Coupled field monitoring and structural analysis to assess scour conditions
J. McConnell
University of Delaware
M. Cann
Gannett Fleming, Inc.
ABSTRACT
The Indian River Inlet Bridge (IRIB) is a steel I-girder

bridge subjected to strong tidal flows, causing signifi-
cant scour holes to form on both sides of two in-water
piers (in the direction of flow). Because of this, a moni-
toring system consisting of tilt sensors placed at the
top of each of the in-water piers has been installed.
The usefulness of this system in evaluating bridge
strength and stability through comparison of the mon-
itoring data to results of structural analysis models is
the primary focus of this paper.
Two different types of structural analysis tools are
developed in this work. These consist of a pushover Figure 1. Substructure analysis results.
capacity analysis of the pier to evaluate stability of the
substructure and finite element analysis to determine resulting stresses are examined. Support movements in
stress conditions in the superstructure as a result of various directions are studied. Assuming linear-elastic
pier movements. behavior, these results can be scaled to determine the
The displacements obtained from the monitoring stress conditions under other magnitudes of displace-
are compared to the results of the pushover capacity ment. Minor modifications to these analyses allow
analyses to ascertain the potential for instability of for considering larger displacements causing inelastic
the piers. Due to uncertainties in both soil properties behavior.
and potential scour conditions, a parametric study on The finite elements results give insight into the loca-
the influence of these variables on stability is neces- tions of highest stress as result of pier displacements,
sitated. The resulting yield deflection is taken as a the corresponding magnitudes of stresses, and the type
point where system instability occurs. These results of displacements that may be most severe. It was found
for five load cases and three soil strengths are shown that the highest stresses typically occur in the girder
in Figure 1. webs over the outer piers. In general, the stress in the
Because Cases B and C were deemed to be overly web at the outer piers appears to be more influenced
conservative, it is suggested that a 4 in. displace- by deflections in the direction parallel to traffic flow
ment at the tip of the pier could be an appropriate than the direction parallel to channel flow. It can be
benchmark for signaling instability of the IRIB sub- inferred from this data that a pier tipping in the traffic
structure. While this is a conservative recommenda- direction would generally be of greater concern than
tion, the unknown material properties at the site and a tilt in the channel direction for the IRIB. The most
other simplifications in the analysis are compelling critical case increases the average web stress by 5 ksi
reasons to further increase the conservatism of this compared to the control case. The data from the analy-
recommendation if desired by the owner. sis can be combined with future information from the
In the finite element analysis, prescribed support tilt sensors to estimate how the scour conditions are
movements are introduced into the model and the affecting the stresses in the bridge.
132
Proof load testing supported by acoustic emission. An example

of application
P. Olaszek
Road and Bridge Research Institute, Warsaw, Poland
G. Swit
The Kielce University of Technology, Kielce, Poland
J.R. Casas
Universitat Politcnica de Catalunya, Barcelona, Spain
ABSTRACT
The use of load testing in the assessment of existing

bridges is justified by the fact that in many experimen-
tal loading processes in bridges, the structure responds
with a higher capacity than expected based on the the-
oretical results from available numerical models. In
situ tests show that sometimes bridges have a reserve
strength that is not accounted for in design codes or
standard assessment methods.
Three main possibilities exist based on the load level
imposed to the bridge during the test execution: soft,
diagnostic and proof load testing. In the present paper,
proof load testing advantages and disadvantages are
discussed and a practical way of bridge monitoring
during the test execution is proposed. Figure 1. The technique of placing concrete slabs and steel
The aim of the proof load test is to discover hid- blocks was used during test.
den mechanisms of response that can not appear under
normal levels of load, but that develop at higher Thanks to the AE signals it was possible to evaluate
ratios of load and may increase the bridge load capa- the cracking limits without introducing any signif-
city. For this reason, in such test, the load introduced icant damage to the girders. The simple follow up
in the bridge is relatively high and due to the risks of of the deflection-load diagram or strain-load diagram
damaging the structure, this type of tests is restricted to as incremental loading is introduced in the bridge,
bridges that have failed to pass the most advanced theo- stopping the test when some sign of non-linearity is
retical assessment or when such theoretical assessment detected, does not guarantee the possibility of not cre-
is not possible due to the lack of bridge documenta- ating any damage to the bridge. In fact, in the case of
tion. The objective of this test is to directly obtain the Barcza bridge even after the detection of the cracking
maximum allowable load in the bridge with a required by visual inspection, the load-deflection diagram con-
safety level. tinued to be linear and no sign of change in the slope
Acoustic emission has been identified as a useful was detected.
technique in the follow up of the loading process in
proof load tests in order to stop the load increase before
REFERENCES
any damage can be inflicted to the bridge.
In the paper the results of field-test of Barcza bridge, Casas, J.R., Znidaric, A., Olaszek, P. 2009. Recommendation
a three span concrete bridge made of pre-stressed pre- on the use of soft, diagnostic and proof load testing. Deliv-
casted beams, are presented. erable D16. EU project ARCHES. Brussels. (available on
The technique of incremental placing concrete slab line at http://arches.fehrl.org).
layers and steel block layers was used during test Swit G., 2009. Diagnostics of prestressed concrete structures
(Fig. 1). The investigation range contained deflection, by means of acoustic emission. In Proceedings of 2009
support displacement, strain, temperature and acoustic 8th ICRMS, IEEE, Chengdu, China, 2009, p. 958963.
emission measurements and visual examination.
133
Jointless bridge: Reliability assessment
J. Podrouek & D. Novk

Faculty of Civil Engineering, Brno University of Technology, Brno, Czech Republic
A. Strauss
BOKU University, Vienna, Austria
ABSTRACT
The performance of the jointless bridge, recently

constructed in Austria, has been evaluated using
stochastic nonlinear analysis incorporating the mate-
rial properties obtained from experimental testing and
artificial neural network based identification (Lehk
et al. 2010). The computational model of bridge
(Fig. 1) is developed using FEM software ATENA.
Monitoring based analysis (Strauss et al. 2010)
served for fine tuning of the FEM model. Stochastic
model of the bridge is created considering material
parameters and loading as random variables. Ran-
domization is done using small-sample Monte Carlo
Figure 3. Safety index for the serviceability limit state.
simulation implemented in FReET software (Novk
et al. 2009). Limit state functions for both ultimate
Table 1. Selected randomized parameters of concrete.
capacity and serviceability are defined, theoretical fail-
ure probabilities and safety indexes are calculated, Parameter Unit Mean COV PDF
resulting plots in figures 2 and 3.
Elastic Modulus* [GPa] 39,5 0.10 N
Poissons ratio [] 0,20 0.05 LN
Tensile strength* [MPa] 2,90 0.09 Weibull
Compressive strength [MPa] 28,9 0.10 LN
Specific fracture energy* [N/m] 178 0.13 Weibull
ACKNOWLEDGEMENT
Figure 1. Half-symmetrical topology with constraints and
continuous load (ATENA FEM model). This outcome has been achieved with the financial sup-
port of the project No. MSM0021630519 from the
Ministry of Education of the Czech Republic and the
research project Eurostars E!4351 RLACS.
REFERENCES
Lehk, D., Kerner, Z. & Novk, D. 2010. Jointless Bridge:
Determination of Fracture Mechanical Parameters Values
for Nonlinear Analysis. IABMASS, Philadelphia, USA.
Novk, D., Vorechovsk, M. & Rusina, R. 2009. FReET
Feasible Reliability Engineering Efficient Tool, Users and
Theory guides. Version 1.5, Brno/Cervenka Consulting,
Czech Republic, http://www.freet.cz.
Strauss, A., Wender, R. & Bergmeister, K. 2010. Monitor-
ing based analysis of time dependent processes. IABMAS,
Figure 2. Probability of failure for the ultimate limit state. Philadelphia, USA.
134
An innovative approach for dynamic damage detection in bridge girders
H.A. Rasheed & Sh. Nayyeri Amiri

Kansas State University, Manhattan, KS, USA
ABSTRACT solution steps, and it considers the crack morphology

in a detailed way. The results obtained so far can only
In recent years, the conditions of aging transportation prove the assumptions made to be correct. The method
infrastructure have drawn great attention to the main- is being extended at the present time, to more involved
tenance and inspection of highway bridges. Therefore, applications. The authors are planning to conduct their
a clear need for structural health monitoring exists own experimental investigations.
for various types of bridges. In this paper, the effects
of cracking damage on the dynamic characteristics
of reinforced concrete bridge girders are investigated
using a newly proposed method. The possibility of REFERENCES
simulating equivalent single degree of freedom sys-
Bolton, R., Sikorsky, C., Park, S., Choi, S., and Stubbs, N.
tem through dynamic testing for in situ bridges is 2005. Modal Property Changes of a Seismically Damaged
evaluated. Damage is considered as a reduction in the Concrete Bridge, J. Bridge Engineering, pp. 415428.
flexural stiffness with increasing degree of cracking. A Farrar, C.R., W.E. Baker, T.M. Bell, K.M. Cone, T.W. Darling,
vibrating motor with varying eccentricities is used for T.A. Duffey, A. Eklund, and A. Migliori. 1994. Dynamic
producing the dynamic harmonic loads. The variation Characterization and Damage Detection in the I-40 Bridge
in natural frequency, amplitude of vibration, damping over the Rio Grande, Los Alamos National Laboratory
and bending stiffness with increasing eccentric mass report LA, 12767-MS.
and for increasing degree of cracking are evaluated Kam, T.Y., Lee, T.Y. 1992. Detection of cracks in structures
using modal test data, Engineering Fracture Mechanics,
through steady state vibration near resonance at mid-
v 42, n 2, p 3817.
span. The changes in the bending stiffness of girders Krawczuk, Marek, 2002. Application of spectral beam finite
in the mid-span with increasing degree of cracking element with a crack and iterative search technique for
are obtained. Damping values are calculated from free damage detection, v 38, n 6, p 537548.
vibration decay function using logarithmic decrement Kato, M. and S. Shimada. 1986. Vibration of PC Bridge
method. Characteristics of cracked girders are simu- during Failure Process, ASCE Journal of Structural Engi-
lated by using a beam model that consists of uncrack neering, 112, pp. 16921703.
segments are connected by elastic rotational springs Masoud, S., Jarrah, M.A., Al-Maamory, M. 1998. Effect of
at crack locations. Analytical expressions for stiffness crack depth on the natural frequency of a prestressed fixed-
fixed beam, Journal of Sound and Vibration, v 214, n 2,
variations in girders are found as a function of cracks
p 201212.
locations and damage intensity. The bridge girder itself Narkis, Y. 1994. Identification of crack location in vibrating
is modeled by using the finite element method too. simply supported beams, Journal of Sound and Vibration,
Excellent agreement is observed between the analyti- v 172, n 4, p 54958.
cal expressions, and the numerical results. In summery, Pandey, A.K., Biswas, M., Samman, M.M. 1991. Damage
the paper presents a method for the calculation of detection from changes in curvature mode shapes, Journal
stiffness of the cracked girders in the mid-span and of Sound and Vibration, v 145, n 2, p 321332.
then damage functions have been proposed in order Qian, G.-L., Gu, S.-N., Jiang, J.-S. 1990. The dynamic
to describe the possible damage pattern. By updat- behaviour and crack detection of a beam with a crack,
Journal of Sound and Vibration, v 138, n 2, p 23343, 22.
ing three parameters of n, w, and d, (i.e. the number
Salawu, O.S. 1997. Detection of structural damage through
of cracks, width of crack and depth of maximum changes in frequency, A review: Engineering Structures,
crack) the damage patterns and their magnitude can v 19, n 9, p 718723.
be successfully determined. The authorsmethod is the Toksoy, T. , Aktan, A.E. 1994. Bridge-condition assessment
alternative approach which is simply applicable to any by modal flexibility. Experimental Mechanics, v 34, n 3,
kinds of simple girders, it does not require complicated p 271278.
135
Definition of structural parameters span structures of bridges by results

of their tests by mobile loading
V. Redchenko
DerzdorNDI, Dnipropetrovsk, Ukraine
ABSTRACT
The operation of bridges is connected to periodic reali-

zation of difficult repair work, which are necessary for
maintenance them in an efficient condition. The total
cost of such repair work directly depends on exist-
ing opportunities on early revealing of all undesirable Figure 1. Response (deflections) of girders of the bridge on
changes in designs. The important role in planning movement of the truck.
sequence of realization of repair work is played moni-
toring a condition of a structure during the certain time. accounts FE-model, it is possible to define static and
Thus the definition of parameters of designs and ten- dynamic parameters of span structure. All this param-
dencies in their change is of the important compound eters further are used for monitoring of the bridge
general process of diagnostics. Operational loading for or additional information deviation from the project.
the bridge is the mobile loading, therefore diagnostics The technique is well fulfilled in practice and is sim-
of a condition of designs on their reactions just on ple in performance. The real example of realization
mobile loading and is by the most rational approach. of test according to the specified recommendations is
The reaction of span structure of the bridge on travel considered.
of the automobile can be divided into two sites: first,
when loading is on the bridge (active reaction AR)
and second, when it already has moved down (free REFERENCES
reaction FR) a Figure 1.
On a site of active reaction fluctuations is the sum Fryba, L., Pirner, M. 2001: Load tests and modal analysis of
of two components: quasi-static reaction (QSR) and bridges, Engineering Structures, 23, pp. 102109.
dynamic reaction to driven loading. On a site of free Luchko, J., Koval, P. et al. 2005. The Bridges: constructions
fluctuations there is only one component dynamic and reliability. Lviv: Kamenjar.
reaction of free fluctuations. Allocating QSR it is pos- Masato Abe, Makoto Shimamura, Masaaki Matsunuma.
2007. Bridge Substructure Monitoring Using Live Load
sible to receive lines of influence of the given reaction, Induced Vibration. TRB 2007 Annual Meeting.
distribution of efforts between girders of span struc- Nowak, A.S., Eom, J. & Sanli, A. 2000: Control of Live Load
ture, dynamic load factor, and at known weight of on Bridges. Transportation Research Record, No. 1696,
the truck experimental stiffness of span structure. Vol. 2, pp. 136143.
Analyzing reaction on a site FR, we receive dynamic Redchenko, V. 2006. Dynamic factor of road bridges. Prob-
parameters of span structure: natural frequencies and lems of definition. The collection Roads & Bridges 2006
them damping ratio. Having natural frequency of first No. 5: pp. 122133. Kiev: DerzdorNDI.
mode and the stiffness of span structure is possible to Redchenko, V. 2007. Diagnostics of a condition span struc-
define its mass. tures of bridges by results of their tests by mobile
loading. The collection Roads & Bridges 2007 No. 7:
The tests of bridges by mobile loading and anal- pp. 143146. Kiev: DerzdorNDI.
ysis of its results according to the submitted rec- Redchenko, V. 2009. Features of use of the spectral analysis
ommendations are the effective approach at research at research of vibrations of building designs. The Bulletin
of bridges. Analyzing the compelled and free fluc- of Donbas National academy of Civil Engineering and
tuations, and also comparing them with results of Architecture 2009-4(78): 4448.
136
Repair monitoring and experimental work associated with

Ferrycarrig Bridge
P.C. Ryan & A.J. OConnor

Trinity College Dublin, Ireland
L. Duffy & A. Daly

National Roads Authority, Dublin, Ireland
G. Jones
Structural Healthcare Limited, Manchester, United Kingdom
ABSTRACT of each repair strategy cannot be determined due to the

fact that the responses observed may be to a greater
This paper presents details of an onsite bridge man- or lesser degree as a result of the extensive disruption
agement project and an experimental study aimed at to the crossheads during the repairs. However, the pat-
examining the relative efficiencies of various repair terns of the probe data can be observed and interpreted
methods in resisting the ingress of chloride ions, which to give an indication as to what is happening within the
lead to reinforcement corrosion. Details of the repair crosshead beams.
and instrumentation of a marine bridge in the South A notable rise has been observed in the corrosion
East of Ireland are presented. The latest results from potential probe readings over the past 22 months. This
the onsite monitoring system are also presented. has resulted in a shift in the corrosion risk classification
The bridge in question is Ferrycarrig bridge which from medium risk to low risk. This would indicate that
carries the N11 single carriageway over the River the crossheads are stabilising with time.
Slaney in Wexford. Built in 1980, the 125.6 m long Four of the six corrosion rate probes are displaying
structure consists of 8 spans of precast, prestressed higher rates of corrosion than would be expected at
beams with a reinforced in-situ concrete infill deck. low or medium risk of reinforcement corrosion. These
In August 2002 an inspection of Ferrycarrig bridge readings are likely to be due to the initial corrosion
identified extensive cracking on the bridges crosshead which was occurring in the steel when the concrete for
beams and South abutment. Following a structural the crosshead beams was placed. In general the read-
analysis and a more detailed inspection a decision ings show a drop in the rate of corrosion penetration
was made in 2007 to repair Ferrycarrig bridge. The over the period of December 2007 to October 2009.
rehabilitation works afforded a unique opportunity to It is likely that this drop is due to the progression of
gather information regarding the efficiency of typical the re-passivation process. The chloride depth probe
alternative concrete repair options in Irish marine envi- readings show very low chloride contents at the level
ronments. It was therefore decided to utilise five differ- of the reinforcement as would be expected following
ent concrete repair strategies for the seven crosshead the crosshead beams extensive repairs.
beams. Six crossheads would be instrumented and The paper also presents details of an experimen-
remotely monitored so that the relative efficiency of tal study which is being undertaken at Trinity College
the various methods could be studied over time. The Dublin in parallel with the Ferrycarrig bridge moni-
repair methods employed were as follows: toring project. The initial phase of testing, which is
concerned with the time to initiation of corrosion,

OPC with standard 50 mm cover
is currently underway. The testing is aimed at estab-

OPC with increased cover (70 mm)
lishing the ability of six repair methods to prevent

60% GGBS as partial replacement in OPC mix
the ingress of chlorides into concrete. The six repair

OPC with mixed in corrosion inhibitors
methods being examined are those used at Ferrycarrig

OPC with silane treatment
bridge, with the addition of a 30% pulverised fuel ash
Each of the crossheads was instrumented with cor- partial replacement mix.
rosion potential probes, corrosion rate probes and In the coming months the laboratory aspect of
chloride ion penetration depth probes. Detailed results the project will provide comparative information on
of the output from the monitoring system to date are the relative merits of the concrete repair strategies in
presented in the paper. At this early stage in the probe resisting the ingress of chlorides.
monitoring (just 22 months in) the long-term benefits
137
Analysis of in-service data collected during biennial inspections

on typical bridges
H.W. Shenton III, K. Connor & M.J. Chajes

Department of Civil and Environmental Engineering, University of Delaware, Newark, Delaware, USA
M. Rakowski
Pennoni Associates, Wilmington, Delaware, USA
B. Brookes
Formerly, Department of Civil and Environmental Engineering, University of Delaware, Newark, Delaware, USA
ABSTRACT
A program has been underway in recent years to

collect in-service strain response data from a num-
ber of typical highway bridges in Delaware. This has
been done in conjunction with the regularly sched-
uled biennial inspection. Strain data is captured using
the In-Service Bridge Monitoring System (ISBMS),
which is usually mounted near mid-span of the most
heavily loaded girder. The system monitors 24/7 for a
period of about two weeks, and records only the peak
strain induced by heavy vehicles. A simple histogram
or timeline plot summarizes the results and provides
a quantitative record of how the bridge is performing
under load. ISBMS data has been collected from a sam- Figure 2. Histogram of recorded strain for bridge 1826.
ple of twelve slab-on-steel girder bridges in the state.
The first data sets were obtained in 2006 and 2007; new estimated, assessment of changes in traffic, and steps
data sets will be collected in 2009 and 2010. Presented to be taken if changes can be attributed to changes in
is a summary of the results of the data collected thus the condition of the bride. It is this type of data that
far. Once the new data sets are collected comparisons can become a part of the permanent bridge inspection
will be made between the first and second sets. Dif- record, and can be used to better manage and maintain
ferences could be due to normal variability, changes the states bridge inventory.
in traffic, or changes in the condition of the bridge.
Also discussed is how the normal variability will be
REFERENCES
Chajes, M.J., and Shenton III, H.W., 2006, Using Diagnostic
Load Tests for Accurate Load Rating of Typical Bridges,
Journal of Bridge Structures, 2(1), 1323.
Howell, D.A., and Shenton III, H.W., 2006, System
for In-Service Strain Monitoring of Ordinary Bridges,
ASCE Journal of Bridge Engineering, Vol. 11, No. 6,
pp. 673680.
Rakowski, M. D. 2008, Bridge Evaluation Using In-Service
and Weigh-In-Motion Data, Masters thesis, University of
Delaware.
Figure 1. Timeline plot of recorded strain for bridge 1826.
138
Determination of concrete bridge ageing by structural health monitoring
S. Soyoz
Bogazici University, Istanbul, Turkey
M.Q. Feng
University of California, Irvine, CA, US
ABSTRACT
Change in the stiffness of a concrete bridge was deter-

mined by vibration-based monitoring over 5 years.
Jamboree Bridge is located in Orange County, CA.
This three span 111-m long bridge is instrumented
with 13 acceleration sensors at both the superstructure
and the columns. The sensor data are transmitted to a
server computer wirelessly.The modal frequencies and
the shapes were identified by processing traffic induced
excitations. Bridge structural parameters, stiffness and
mass, and the soil spring values were identified utiliz-
ing the neural network technique. The identified modal
frequencies vary within +/10%. The identified stiff-
ness of the bridge deck varies within +/3%. These
results are shown in the following two figures. Based
on the statistical analysis of the collected data for each
year, 5% decrease in the first modal frequency and
2% decrease in the stiffness of the bridge deck were
observed over the 5-year monitoring period. Proba-
bility density functions were obtained for stiffness
values each year. Stiffness threshold values for the
design life of the bridge deck under the operational
loading can be determined. Therefore, the information
obtained in this study is valuable for studying age-
ing and long-term performance assessment of similar
bridges.
139
Monitoring based performance checking of the jointless

Marktwasser Bridge S33.24
A. Strauss, R. Wendner & K. Bergmeister

University BOKU, Vienna, Austria
D.M. Frangopol
Lehigh University, Bethlehem, Pennsylvania,USA
ABSTRACT
In bridge engineering maintenance strategies are

highly influenced by construction type and quality
of design. Nowadays bridge designer and owner tend
to include life-cycle cost analyses in their decision
processes regarding the overall design by trying to
optimize structural reliability and durability within
financial constraints. However efforts to reduce main-
tenance costs over the expected lifetime by adopting
well established design principles leads to unknown
risks. Monitoring solutions can reduce this associated
risk of new designs by a permanent observation of per-
formance of structural components during prescribed
time periods.
In this research monitoring is applied to the jointless
Marktwasser Bridge, with emphasis on assessing the
effects of time dependent processes including creep,
shrinkage and temperature. The Marktwasser Bridge
S33.24 is a three span bridge over the Danube in
Lower-Austria, Austria, with 19.50 m, 28.05 m and
19.50 m spans, see Fig. 1. The bridge is designed for
the five lane highway S33 according to the national Figure 1. Marktwasser Bridge, S33.24.
standard B-EN1992 for reinforced concrete bridges.
In order to reduce construction and maintenance cost,
the Austrian Highway Financing Company ASFINAG strength and its associated fracture mechanical prop-
instructed the engineering offices to convert the orig- erties and provided essential information for the initial
inal design to a jointless bridge structure by omitting inverse analysis of the static system. Finally, a prob-
expansion joints and bearings. abilistic approach was used to assess the influence
A fiber optic monitoring network has been com- of uncertainties on structural response (load-carrying
missioned in the first lateral south western field of capacity).
the Bridge S33.24, in order to capture effects due The current investigations demonstrate a very
to (a) the consecutive concrete pouring of the bridge good identification of time dependent mechanical and
fields, (b) the removal of the formworks, (c) the pave- chemical (e.g. creep and shrinkage) processes and
ment loading, (d) the variability in the environmental a continuous observation of the structural loading
temperature, (e) the creep and shrinkage processes in response by the installed permanent monitoring sys-
the first three years, and (f) a proof loading proce- tem that consists of several different sensor types.
dure which is planned in Spring 2010. In addition, Further research in this area will include the appli-
numerous concrete samples were extracted during cation of the methodology to the evaluation of the
the casting procedures in January 2009. These sam- soil structure interaction, and the calibration of finite
ples served for the determination of the concrete element models among others.
140
Monitoring based verification of the soil structure interaction

of the Markwasser Bridge S33.24
R. Wendner, A. Strauss & K. Bergmeister

University BOKU, Vienna, Austria ands
ABSTRACT
In structural bridge engineering maintenance strate-

gies and thus budgetary demands are highly influenced
by the quality of design in general as well as the cho-
sen construction type in particular. Nowadays bridge
owners and planers tend to include life-cycle cost
analyses in their decision processes regarding the over-
all design trying to optimize structural detailing with
respect to robustness, durability, exchangeability of
certain structural members and reliability (Frangopol
et al. 2008, Strauss et al. 2009). However efforts to
reduce maintenance costs over the expected lifetime
by adopting well established design principles leads
to unknown risks concerning for instance boundary
conditions.
Monitoring solutions can reduce the associated risk Figure 1. Installation of (a) top fibre-optic sensor layer in
of new design concepts by constant supervision of sand bed, and (b) sensor principle, and geometry of geotextile
layer.
critical structural characteristics and in combination
with data analyses allows for the verification of new
numerical and analytical models. This paper especially
focuses on the layout and analysis of monitoring data
obtained by an integrative multi-sensor-monitoring
system targeting the soil-structure interaction of frame
bridges.
Based on an application example of a three-span
frame bridge with a total length of 67 m different
aspects of the design such as build up of earth pres-
sure against the abutment wall, the bedding of the Figure 2. Total horizontal deformation in the dilatation area
drilling piles, the functionality of the chosen slab between abutment and earth dam due to contraction of the
detail to accommodate changes in length caused by structure by 20 mm (winter).
temperature loads in combination with the respec-
tive regulations in national and international standards serve for the analysis and further optimization of struc-
were to be evaluated. To that end four different sensor tural details and thus ensure efficient and safe design.
systems were installed behind the southern abutment,
consisting of 2 extensometers, 20 electrical strain
gages, 10 fibre-optic strain sensors and 3 vertical incli- REFERENCES
nometers. Figure 1 shows the installation of the top
Frangopol, D.M., A. Strauss & S. Kim 2008 Bridge Reliabi-
fibre-optic sensor layer. Additionally 20 fibre optic
lity Assessment Based on Monitoring. Journal of Bridge
strain and temperature sensors were placed in the deck Engineering, 13, 258270.
of the southern span. Strauss, A., S. Hoffmann & R. Wendner 2009 Structural
Finally the installed monitoring system will allow assessment and reliability analysis for existing engineer-
the calibration of suitable finite element models such as ing structures, applications for real structures. Structure
shown for instance in Figure 2. Those models then can and Infrastructure Engineering, 5, 277286.
141
Experimental trials on the detection of reinforcement breaks

with the magnetic flux leakage method
T. Wolf & T. Vogel

Institute of Structural Engineering, Swiss Federal Institute of Technology (ETH) Zurich, Switzerland
ABSTRACT predefined reinforcement layouts including breaks in

reinforcing bars have been executed.
Hardly any fatigue breaks of reinforcing bars have The measurements of the magnetic flux density
been reported so far. Reasons may be that fatigue were recorded and compared with theoretical con-
breaks are not actively searched and that they are hard siderations. It could be found that the data correlate
to detect by non-destructive testing methods. The mag- well.
netic flux leakage (MFL) method can fulfill this gap. In a second step, measurements with the MFL
The applicability and the limits of the MFL method method are prepared for application on a reinforced
will be investigated in connection with large-scale concrete frame which will be loaded cyclically. Breaks
fatigue tests. of the reinforcement are expected to occur in the frame
The MFL method is based on the physical fact that corners. Acoustic emission is used to determine the
a break in a steel bar leads to a change of sign in the time of break. Once a break is assumed, loading is
magnetic flux density, which can be measured using halted and the measurements with the MFL method
hall sensors. Prior to the measurement, the steel bar are performed.
has to be magnetized with an external magnet that After the experiment has been finished the rein-
will be moved alongside the steel bar. Figure 1 shows forcement will be uncovered and the positions of the
4 aligned disc magnets besides a reinforcing bar. The measured breaks will be compared with the effective
sensor (Fig. 2) has also to be moved alongside the ones. Due to the much denser reinforcement layout of
reinforcing bar to measure the longitudinal and radial this specimen the data interpretation of the measured
component of the magnetic field. magnetic flux values will be more complicated as in
In preliminary tests first experimental trials on sin- the preliminary tests. Evaluation of the data will give a
gle unbonded steels bars and concrete specimens with first impression of the capability of the MFL method.
Figure 1. Permanent magnet besides a reinforcing bar. Figure 2. 3-axis sensor module besides a reinforcing bar.
142
Innovative structural health monitoring for Tamar Suspension Bridge

by automated total positioning system
K.Y. Koo & J.M.W. Brownjohn

University of Sheffield, Sheffield, UK
D. List & R. Cole

Tamar Bridge and Torpoint Ferry Joint Committee, UK
T. Wood
Cornwall Country Council, UK
ABSTRACT
This paper presents an innovative structural health

monitoring system using an automated Total Posi-
tioning System (TPS) and its application to Tamar
Suspension Bridge. Conventional TPS also known as
Total Station used for surveying have evolved to have
a functionality of automated and unmanned opera-
tions. For SHM purposes, an automated TPS can be
utilized for monitoring 3D deflections of a structure.
Figure 3. Temperature vs. Easting deflections.

Tamar Bridge, Plymouth, UK has been equipped by
an automated TPS on the roof of the bridge office for
monitoring of the bridge deck and tower to understand
complex behavior of the suspension bridge system
under environmental loadings such as temperature,
wind and traffic loads. From Sept 2009, 15 points
around the bridge deck and towers have been moni-
tored and the global deflections were recorded. Based
on the three months monitoring records, temperature
change was found to be the dominant contributor
Figure 1. Automated TPS: Leica TCA1201M and Reflector. of bridge deflections in the longitudinal and vertical
directions while the transverse deflections are not
understood yet. Further analysis on the deflections is
underway. Limitations of the current technology were
discussed.
REFERENCES
Fish R., Gill J., Tamar suspension Bridge strengthening and
capacity enhancement. In Bridge Modification 2: Stronger
and Safer Bridges Thomas Telford, London, 1997, ed B
Pritchard.
Brownjohn, J.M.W. and Carden EP, Real-time Operational
Modal analysis of Tamar Bridge. IMACXXVI, Orlando,
USA, 47 Feb 2008.
Brownjohn, J.M.W., Pavic A., Carden P., Middleton, C.,
Modal Testing of Tamar Suspension Bridge, IMAC XXV,
Figure 2. Easting deflections. Orlando, USA, 1922 Februray 2007.
143
MS3: Present & future of bridge inspection & evaluation
Organizers: S. Alampalli, A.K. Agrawal & M. Ettouney
Inspection needs of deteriorating bridge components
A.K. Agrawal
City College of New York, New York, NY, USA
S. Alampalli
New York State Department of Transportation, Albany, NY, USA
ABSTRACT the New York State highway bridge inspection pro-

cess and the second study is the development of
Currently, about 25% of highway bridges in the United deterioration curves for bridge components based on
States are considered either structurally deficient or inspection data.
functionally obsolete. Even though these deficient A Federal Highway Administration (FHWA) study
bridges are safe for the legal loads allowed on them, completed in 2001 reported that bridge inspection
they are in need of corrective maintenance, repair, ratings on a national level have high variability and
rehabilitation, or replacement to bring them to current questioned the quality and reliability of visual based
standards with no load restrictions or inconvenience inspection procedures. Due to the methodology used,
to the travelling public. This number is expected to these results may not be applicable to reliability of
increase due to limited capital funds available to recon- the bridge inspection program of individual states.
struct these aging bridges. In such a situation, an For example, the NYS bridge inspection program is
inspection program that can effectively address both more robust and detailed than what is mandated by
safety and performance needs is required. FHWA. Thus, a research project has been initiated to
The deterioration of bridge elements is influenced quantitatively document the variability associated with
by the combined effects of many complex phenom- the bridge inspection program, suggesting improve-
ena (e.g., corrosion, concrete degradation, creep, ments to policy and procedures and areas requiring
shrinkage, cracking, and fatigue). In the absence of further training, if needed, to reduce this variability
mechanistic-based models that require quantitative and improve the reliability/consistency of the program.
contribution of these complex phenomena, a model This paper gives some background details and the
using bridge inspection data may be used to estimate scope of the on-going project.
deterioration of bridge elements. A new approach using Weibull distribution has
This paper at first presents a review of national been presented to calculate deterioration rates based
bridge inspection standards and data collected through on the historical bridge inspection data for highway
bridge inspections. This is followed by the review of bridges in New York. Advantages and capabilities
gaps and needs presented by the recent ASCE/SEI- of the Weibull-based approach are demonstrated by
AASHTO ad hoc group, formed in the aftermath of the three cases comparing Markov Chains and Weibull
recent Minnesota bridge collapse, that discussed the distribution-based approaches. These rates can be used
bridge inspection and rating practices to identify the for estimating remaining life, comparing alternatives
gaps, needs, and issues associated with current prac- for repair and retrofits, and analyzing effects of factors
tices and policies for the condition assessment of on deterioration of bridge elements. These can also be
the bridges. Finally, the paper gives a brief descrip- used to select or customize advanced assessment tech-
tion of two studies, authors are involved with, meant niques during the life-cycle of the structure to augment
to improve bridge inspection and management prac- visual inspection and to collect data useful for bridge
tices. The first study is to evaluate the reliability of management functions.
147
Periodic NDE in support of structural health monitoring of bridges
R.W. Arndt & F. Jalinoos

Federal Highway Administration, Office of Infrastructure R&D, McLean, VA, USA
J. Cui & D. Huston

University of Vermont, Burlington, VT, USA
ABSTRACT
In the last two decades, the U.S. Federal Highway

Administration (FHWA) invested significant effort
into the development of nondestructive evaluation
(NDE) tools that may be suitable for an objective
condition assessment of bridge structures. Parallel
developments explored the concept of structural health
monitoring (SHM) and its application to constructed
systems especially bridges. Generally there is a wish
to have a wider utilization of the various technologies Figure 1. Ground penetrating radar measurements with
2.3 GHz handheld antenna on concrete slab specimen with
that have been developed in the last two decades and frozen and thawed cupcake pattern water layer.
to make a transition from the current practice of high-
way and bridge engineering to one that is based on benchmark approach project (Arndt & Jalinoos 2009)
objective measurements and the lifecycle benefit/cost using and evaluating state of the art NDE methods
of the system. Accordingly there is a need for strategies for SHM of bridge decks is outlined and first results
to incentivize and facilitate the application of these documented. These are already very promising in the
technologies. laboratory but its final application in the field is very
The strategy presented in this contribution is sup- challenging; a periodic multi sensor NDE for SHM
plementary to the global SHM; it can be regarded as a requires a high standard of repeatability for data com-
practical and less costly implementation for tracking parability and advanced processing, which means a
mission critical performance for much of the nations high level of automation is required (Huston et al.
600,000 bridges. It focuses research and development 2008).
on techniques to facilitate quality control and periodic
lifecycle applications of measurements including peri-
odic and multi-sensor NDE measurements with the REFERENCES
following combined strategy:
Arndt, R. & Jalinoos, F. 2009. NDE for corrosion detec-
Multi-sensor NDE to improve reliability of data
tion in reinforced concrete structures A benchmark
NDE for structural condition assessment to the approach. In Nondestructive Testing in Civil Engineering
level of detail to perform any bridge maintenance, (NDTCE); Proc. intern. symp., Nantes, June/July 2009.
preservation and rehabilitation type actions use Nantes: France.
of periodic NDE as a strategy for bridge preserva- Huston D., Cui J., Burns D. & Jalinoos F. 2008. Con-
tion, as opposed to the standard concept of NDE crete Bridge Deck Condition Assessment with Automated
inspection for safety improvement (Jalinoos et al. Mul-tisensor Techniques. In International Association for
2009). Bridge Maintenance and Safety (IABMAS); Proc. intern.
symp., 2008. Seoul: South Korea.
Thus periodic inspection can be a part of the reg- Jalinoos, F., Arndt, R., Huston, D. & Cui, J. 2009. Struc-
ular biennial inspection or performed on-demand in tural Health Monitoring by Periodic NDE: NDE for Bridge
response to an extreme event such as hurricane, storm, Maintenance. In Material Evaluation 67(11): in print.
earthquake, blast, collisions, fire, and superload.
Examples using periodic GPR for SHM of bridge
deck deterioration, detection of ice (compare Fig. 1),
and rebar corrosion are presented. Furthermore a
148
To see is to know: visualization in bridge inspection and management
J. Bien & M. Kuzawa

Wroclaw University of Technology, Poland
B. Bien
SPIN s.c., Research & Development, Poland
ABSTRACT
A naturally designed user interface, especially in the

engineering software systems, innovates the way peo-
ple work, improving their efficiency and having a
positive impact on the whole branch. Today, a rapid
development of computer graphics hardware allows
to model software systems as interactive real-time
visualizations, even when using affordable mid-range
machines.
Among many software genres that this fact makes
evolve quickly, there are Bridge Management Sys-
tems (BMS). A demand for better user experience in
BMS yields migration from traditional data forms into
model-centric interfaces. In model-centric systems,
most user operations are based on real-time interaction
with a visual model of the infrastructure, represented in Figure 1. The geometry model matrix (Bien 2002).
a convenient level of detail, both spatial and temporal.
The same structure, or a part of it, can be represented
in a variety of aspects, depending on the current con- over time. The static category includes invariable ele-
text in which the user is working at the moment. All ments such as maps, photograph, and non-adaptive
geometrical, physical, and chemical properties of the drawings and models, and the dynamic category
structure, as well the extents, intensity, and specifics of includes elements that change over time, such as films,
existing or predicted damages, can be shown exactly TV, adaptive drawings and models, and variable traffic
where they exist in a real world or where they can be signs.
expected, accordingly. Proposed classification of the geometry models
The paper presents diverse types of interactive is presented in Figure 1. The columns represent
visualizations that can be applied into certain areas the dimensionality of space used for visualization,
of BMS, in diverse phases within the lifespan of a from a single point to the full three dimensions,
structure. For example, when a bridge is already in accordingly. The rows indicate the dimensionality of
operation, the used representations and models often models that can be represented in the correspond-
do not need to be as detailed as during the design, ing spaces. A notation for the dimensional level of
which reduces complexity and requirements for the detail has been defined as ex sy , where x is the dimen-
computational power. The discussed solutions include: sionality of elements, and y is the dimensionality
component-based modeling of bridge inventory infor- of space.
mation, visual recording of damage data during an
inspection, and representing predictions of technical
state changes in dynamic environments. Visualization
REFERENCES
examples from the finalized projects are presented, as
well as several future prospects in the field. Bien, J. 2002. Modeling of Bridge Structures During Oper-
For the BMS, we propose a classification of visual- ation (in Polish). Wroclaw: Publishing House of the
ization techniques based on a variability of the image Wroclaw University of Technology.
149
Knowledge-based expert tools in bridge management
J. Bien, A. Banakiewicz, M. Gadysz, T. Kaminski & P. Rawa

Wrocaw University of Technology, Poland
B. Bien
SPIN s.c., Research & Development, Poland
ABSTRACT
The paper presents a procedure of expert tools cre-

ation by means of multi-level hybrid networks together
with some practical applications of the pilot versions
supporting bridge management. An expert tool can
be based on various components dedicated to specific
techniques of symbolic or non-symbolic knowledge
representation. Selection of a representation tech-
nique depends on the form and content of the avail-
able knowledge. In the presented applications, the
techniques of neural networks, fuzzy logic and multi-
dimensional mathematical functions are involved in
the process of knowledge representation.
For creation of each individual component and for
configuration of the final network, the specialized soft-
ware system NEURITIS was developed. Details of the
system were presented by Bien & Rawa (2004). In this
system the hybrid network can be created depending
on the problem that should be solved and on the type Figure 1. Technical condition rating in the RBMS SMOK
supported by the Bridge Evaluation Expert Function (BEEF).
of available information by means of the following
components:
neural component, based on non-linear multi-layer time-consuming process, engaging civil engineers,
artificial neural networks (ANN) trained using the knowledge engineers, computer scientists, etc.
supervised back-propagation method; The proposed expert tools enable easy composi-
fuzzy component, based on the fuzzy logic with the tion of data and knowledge accumulated in the BMS
ability of fuzzy inference; and ensure uniformity of the decision procedures
functional component that enables implementation within the whole management system. Utilization of
of the analytical functions. the expert tools is user-friendly and enables even the
users with limited computer experience to analyze
Experience in design, testing and preliminary appli-
and successfully solve the complex bridge engineering
cations of the following expert tools is presented
problems arriving in bridge management.
and discussed: Bridge Evaluation Expert Function
(BEEF) supporting assessment of railway bridge con-
dition (Fig. 1), expert system Masonry Bridge Dam-
age Evaluator (MyBriDE) for evaluation of load REFERENCES
capacity of damaged masonry arch bridges proposed
by Kaminski (2008) and conception of the expert Bien, J. & Rawa, P. 2004. Hybrid Knowledge Representation
in BMS, Archives of Civil and Mechanical Engineering
system Concrete Bridge Degradation (CoBriDe) IV(1): 4155
supporting forecast of the degradation processes of Kaminski, T. 2008. The ultimate load of masonry arch bridge
concrete bridge structures. spans taking into account influence of defects (PhD The-
Preparation of network components and the cre- sis), Wrocaw University of Technology, Institute of Civil
ation of practically effective networks is a very Engineering.
150
Enhanced bridge management via integrated remote sensing
S. Chen, E. Hauser, K. Dai, W. Liu, B. Ribarsky, S. Lee & B. Tolone

University of North Charlotte at Charlotte, Charlotte, NC, USA
C. Boyle
Boyle Consulting Inc., Charlotte, NC, USA
ABSTRACT Issues that limited the application of CRS-GSI tech-

nology to conventional bridge monitoring have been
Commercial remote sensing (CRS) as robust health identified as: 1) limited acceptable bridge inspection
monitoring techniques offer unique features that are procedure; 2) inconsistency in bridge management
lacking from current embedded structural health mon- styles; 3) lack of in-depth understanding of CRS-GSI
itoring systems, including geo-referencing, spatial technologies; and finally 4) complexity in multivari-
view and high-resolution top view. Beginning in 2007, ate data integration. Because CRS data exist in image
a research partnership (University of North Carolina at format and bridge data exist in PDF or text-file for-
Charlotte, ImageCat Inc., Boyle Consulting, Charlotte mats, there is the challenge in integrating both data into
DOT and North Carolina DOT) has completed a proof- BMS, so that the bridge managers can visually study
of-concept project to develop ground-based LiDAR the combined or fused data and other information.
scan and sub-inch resolution aerial photography into An IRSV system is proposed that would allow the
the IRSV (Integrated Remote Sensing and Visualiza- fusion of different remote sensing data with exist-
tion) bridge data diagnostic platform. CRS technology ing bridge data into a multi-variate viewing platform.
for structural health monitoring differs from conven- IRSV represents a critical juncture towards the Total
tional technologies in that the measurements are from View Integration (TVi) infrastructure monitoring
a distance away from the object without physically concept, which is defined in this paper, and provides
touching the object. further incentives for CRS development. This paper
The terrestrial 3D laser scanners, operating in the presents an overview of some technologies adopted
narrow near infrared wavelengths and with scanning in IRSV and discusses the potentials of CRS tools to
capability, can generate dense point clouds of bridge enhance bridge inspection and data management.
geometric data. The manipulation of the 3D geomet-
ric data allows damage detection and quantification
of surgical damages to a structure. Applications of REFERENCES
the scanning LiDAR also include deflection mea-
surements during load testing. A 3 mm resolution of Chen, S.E., E. Hauser, R. Eguchi, W. Liu, C. Rice, Z. Hu,
C. Boyle and H. Chung. 2009. Bridge Health Monitor-
displacement measurements can be achieved using
ing Using Commercial Remote Sensing, Proceedings, 7th
LiDAR. The results are also ideal for FE model val- International Workshop on Structural Health Monitoring,
idation, which has been demonstrated using a bridge Stanford, CA.
model for a steel-girder bridge. AASHTO. 2008. Nationwide Bridge Management Survey,
The use of geo-referenced, high-resolution (sub- Dec 9th to Dec 19th 2008.
inch) aerial photography and automated imaging Ribarsky, W., E. Hauser, S. E. Chen, W. Tolone, S. W. Lee,
allows surface mapping of the wear surface conditions. R. Chang, W. Liu, R. Vatcha, and X. Wang. 2009. Inte-
To achieve the high-resolution imagery, the Small For- grated Remote Sensing and Visualization (IRSV) System
mat Aerial Photography (SFAP) technology is adopted for Transportation Infrastructure Operations and Manage-
ment, Poster Presentation,Transportation Research Board
along with onboard GPS system to establish geo-
Annual Meeting, Washington D.C.
referencing and geo-stamping of all imageries taken.
Application of SFAP is demonstrated in a bridge rating
technique solely based on crack detection capabilities
of the high-resolution SFAP.
151
Full field mapping of bridge deformation using digital speckle

photography
F.P. Chiang & J. Yu

Department of Mechanical Engineering, Stony Brook University, Stony Brook, NY, USA
ABSTRACT
Many of our countrys bridges are aging structures that

require periodic and frequent inspection. Currently the
prevailing mode of inspection is visual inspection by
human. Not only this is a tedious and time consum-
ing process, inspectors with different background and
training may render different conclusions. Inspection
using instrument such as thermal imaging can only
inspect a small region at time. In this paper, we present
a new technique whereby the entire bridge deformation
pattern can be mapped simultaneously in a single step.
The new technique called digital speckle photography
[Chiang, 2003] does not need expansive equipment
Figure 1. Bridge main beam deflection from CASI and
and is easy to set up. And the data can be obtained in FEM.
semi-real time.
A speckle pattern is a collection of random dots
or particles that show up distinctively when pho-
tographed. They can be used for quantitative measure- the entire bridge. The deflection of the bridge along the
ment of displacement and strain. In this study, we built entire span is compared with finite element result as
a 10 ft by 1.25 ft truss bridge using commercially avail- shown in Fig. 1. If local strain distribution is needed,
able pine wood planks. One surface of the wood bridge the camera can zoom into a small region and calculate
is painted with a thin coat of retro-reflective paint the detailed strain distribution. The technique can also
which has imbedded minute glass particles. As a result be used in dynamic situation in which the bridge is
when a light beam illuminates the paint surface most of oscillating (Chiang, 2009).
the light is reflected back to the light source. Two 200 w
flood light are used to illuminate the bridge. They are
situated at 36 ft away from the bridge and on either side
of a digital camera. The bridge is loaded evenly with REFERENCES
50 lb sandbags totaling 600 lbs. The painted surface
Chiang, F.P. 2003. The evolution of white light speckle
of the bridge was photographied before and after the method and its applications to micro/nano technology and
application of load. Each digital image of the bridge is heart mechanics. Optical Engineering 42: 1288.
subdivided into subimages of 128 128 pixel arrays Chiang, F.P. 2009. Super-resolution digital speckle photogra-
and then compared using a specially design software phy for micro/nano measurements. Optics and Lasers in
call CASI. The result is a displacement vector map of Engineering 47:274279.
152
Design for inspection one way towards durable infrastructures
M. Drissi-Habti
Laboratoire Central des Ponts & Chausses, France
ABSTRACT goals. To achieve this, an appropriate design procedure

must be implemented, including several complemen-
New materials developed in the last three decades tary research topics : Materials science Mechanical
as well as the significant advances recorded in engi- Engineering Instrumentation (NDT and SHM, com-
neering and computering are currently enabling us to putering). In addition, given the new rules governing
figure out new infrastructures based on new design, the politics worldwide, new infrastructures must meet
new materials and challenging smart structures. sustainable development requirements.
In the field of Civil Engineering, the design of new This work will address the points listed above in the
infrastructures like bridges and roads must allow ease case of bridge infrastructures.
of structural health monitoring, as one of the main
153
Non-homogeneous Markov Chain for bridge deterioration modeling
G. Fu
Wayne State University, Detroit, MI, US; Tongji University, Shanghai, China
D. Devaraj
Wayne State University, Detroit, MI, US
ABSTRACT model the entire life span of the bridge elements dete-
rioration process. Given inspection data of condition
Bridge components or elements deteriorate with time. state evolution, a fitting scheme is also suggested here
Deterioration modeling is an important step in bridge to estimate the non-homogeneous transition probabil-
management operation. In the popular bridge manage- ities depending on age. In addition, a new screening
ment system Pontis, a Markov Chain model is used to scheme is proposed in this paper for initial data screen-
describe bridge element deterioration. The core part of ing to filter inconsistent raw data. Application of the
this model is its transition probability matrix, which proposed model to Michigan bridge inspection data
describes the deterioration process using a probability shows improved modeling results. This approach was
measurement for the likelihood that the element dete- also applied to the bridge condition rating data in
riorates (transits) from a condition state to another. the National Bridge Inventory format and thereby
Namely, those elements that deteriorate faster are produced more realistic predictions.
assigned higher transition probabilities from a bet-
ter condition state to a worse one. It is interesting
though to point out that this approach does not take REFERENCES
into account the age of the element. In other word, an
AASHTO Pontis Technical Manual, Release 4.4, 2004
older bridge element is assigned the same transition Fu, G. Modeling of Lifetime Structural System Reliabi-
probabilities from better conditions to worse ones as lity, Ph.D. Dissertation, Report No. 87-9, Department
a much younger one. This model is thus referred to of Civil Engineering, Case Western Reserve University,
as a homogeneous Markov Chain. In addition, due to Cleveland, OH, May 1987.
typically long life span of bridge elements in decades, Fu, G., and Devaraj, D. Methodology of Homogeneous and
small inaccuracy in transition probability matrix esti- Non-homogeneous Markov Chains for Modeling Bridge
mation for short time intervals (a few years) can cause Element Deterioration, Final Report to Michigan Depart-
large errors in the final result of service life prediction. ment of Transportation, Center for Advanced Bridge Engi-
This paper proposes a new model of non- neering, Department of Civil and Environmental Engi-
neering, Wayne State University, August 2008, MDOT
homogeneous Markov Chain to take into account the Research Report 1520.
elements age, with detailed elaboration on the issues Fu, G. and Moses, F. Application of Lifetime System Reli-
discussed above. Namely, the transition probability ability, Preprint No. 52-1, American Society of Civil
matrix is now age-dependent not constant for all ages. Engineers Structures Congress86, New Orleans, LA,
Therefore, multiple such matrices will be needed to Sept. 1518, 1986, p. 281.
154
New tools for inspection and evaluation of steel truss bridge gusset plates
C. Higgins, Q.D. Nguyen & O.T. Turan

School of Civil and Construction Engineering, Oregon State University, Corvallis, OR, USA
ABSTRACT
Gusset plate connection performance has become an

important issue for many transportation agencies after
the collapse of the I-35W Bridge in Minnesota, USA.
Due to this failure, steel truss bridges are undergo-
ing additional scrutiny focused on the gusset plate
connections. Connection evaluations require accu-
rate as-built drawings and typically use simplified
specification-based design methods for analysis.
A new methodology has been developed that per-
mits collection of accurate field measurements, extrac-
tion of data to production of as-built CAD connection
drawings and direct finite element (FE) modeling and
analysis of as-built connections. The method uses Figure 1. Original image of mock gusset plate.
close-range photogrammetry techniques to rectify
field-collected digital images taken with consumer-
grade cameras to produce scaled orthographic pho-
tographs (orthophotos) of the bridge connections as
illustrated in Figs. 1 and 2. From these orthophotos,
the user collects true-scale measurements and exports
the plate and fastener geometric information. The geo-
metric data are used directly to create CAD drawings
thereby limiting data entry errors. These as-built draw-
ings can be compared with design drawings or used
in simplified analysis and ranking methods. The same
geometric data are ported to ABAQUS using scripts
that generate finite element models of the gusset plates.
When combined with member forces applied at the
fastener locations, the FE analysis can be performed
(including buckling and nonlinear analyses), and the
results can be used to rate the connections.
The approach enables rapid and accurate collec-
tion of field measurements and allows direct assess- Figure 2. Orthophoto that allows metrification of connec-
tion geometry.
ment of the connection behavior as compared to
traditional simplified methods. The implementation
procedure is straightforward and does not require data flow and sophisticated evaluation of the com-
specialized knowledge of photogrammetry or finite plex plate stress interactions to enable bridge ratings
element methods. It can be practicably employed under of existing steel gusset connections and represents a
field conditions using current technology and per- new era for quantitative inspection of transportation
sonnel. Integration of the techniques allows seamless infrastructure.
155
The computerized cables stay inspection of the Cooper River

Bridge (Arthur Ravanel, Jr)
B. Kroely & G. Hovhanessian

Advitam, Sterling, VA, USA
J. Stieb
Freyssinet Inc., Sterling, VA, USA
ABSTRACT FLLC provided a recommended inspection & mainte-

nance manual defined for the Ravenel Bridge, however,
The Ravenel Bridge is a 4km cable stayed bridge it was written at a high comprehensive level. Advitam
which opened in July 2005. It connects downtown worked with FLLC to simplify the questions given to
Charleston, SC to Mount Pleasant, SC. Freyssinet, the inspectors. Finally, lists of defects with attributes
LLC (FLLC), supplied and installed 128 parallel were entered.
strand stay cables supporting the structure. The orig- After brief training sessions off and on-site, each
inal contract included a warranty inspection every 2 team of inspectors were given a Tablet PC loaded
years for 10 years after the bridge opening, the first with ScanPrint software. They were also equipped with
inspection occurred in October 2007. digital cameras and measuring tools. Following the
A main objective was to reduce time spent on the inspection plan, the teams then visually inspected the
inspection process especially on report generation, bridge structure filling out the forms along the way;
Advitams Inspection software: ScanPrint was used by and if necessary drew defects and attached pictures
FLLC for this purpose. Another important objective directly on the Tablet PC.
was to make the information easy to access and to fur- Once the on-site inspection was performed, the
ther facilitate efficient and accurate inspections in the Tablet PC databases and files were merged together.
future. Advitam worked with FLLC to define the inspection
ScanPrint was initially developed in 1999 for the report. Two report templates were implemented:
Inspection and Maintenance management of the Vasco
de Gama Bridge in Portugal. It is presently in use on Word report: Containing the information gathered
many large cable stay bridges, large suspension bridges on the field: AutoCAD drawings with the defects,
and several landmark structures. defect information, inspection forms, results and
The first step of the project for Advitam was setting pictures.
up the inventory of the bridge and computerizing the Excel Report: Containing only the results of the
inspection manual. inspection forms, it allows quick and easy compar-
The inventory is entered by breaking down the ison of results between elements.
bridge into a tree view of components. The break down
is made according to the bridge structural types and A draft report was first generated in November allow-
in a manner consistent with the way inspections are ing SCDOT to send comments prior to issuing, the
planned to be performed. This concept is fundamental final report in December.
to ScanPrint. Once the inventory has been entered, no The Ravenel Bridge inspection proved to be a per-
additional data entry will be required for future inspec- fect project to use the ScanPrint software. The visual
tion and only a few steps are necessary to start the next inspection of multiple similar elements makes the data
inspection. even more relevant to compare to each other. Beside
After entering the inventory, the inspection manual the time saved during the report generation, full value
can be computerized. For this type of unique struc- of the computerized inspection will be unveiled dur-
ture, the standard NBI inspection forms or AASTHO ing the second inspection, and so on. More and more
components do not adequately address the particu- data will be entered, and therefore can be compared
lar issues associated to cable stayed systems. A more and analyzed; each cycle is adding more value to the
detailed and specific inspection manual is required. database.
156
Load ratings of a concrete-encased steel 3-hinge arch bridge
L.-Y. Lai, W.-S. Chang & M. Majd

Specialty Engineering, Inc., U.S.A.
ABSTRACT
A concrete-encased steel arch bridge was rated using

both the ACI and AISC criteria because the steel ratios
are within the applicable range of both criteria. Fig-
ure 1 shows the P-M interaction strength curves and
their relationships with the factored loads following
both ACI and AISC procedures. The AISC procedure
results in more conservative rating factors because of
high dead load and live load axial force to moment
ratios. The HS20 loading inventory ratings are 3.15
and 3.94 based on the AISC and ACI procedures,
respectively. Figure 2. ACI and AISC P-M strength curves.
A parametric study based on the steel ratios shows
that ACI strength interactive curves typically have
higher strength than the AISC curves in the compres- REFERENCES
sion zone as shown in Figure 2. When the axial force
ratios drop below Pu /c Pn = 0.2, the AISC procedure, AASHTO 2008. AASHTO LRFD Bridge Design Spec-
however, results in higher strength. ifications. 4th Edition with 2008 interim revisions,
Based on the information from the original testing Washington, D.C.: American Association of State High-
data and studies used in developing the AISC com- way and Transportation Officials.
posite requirements, the concrete encased arch ribs ACI 2008, Building Code Requirements for Structural Con-
with multiple steel shapes connected by lacing bars crete (ACI 318-08), Farmington Hills, MI: American
Concrete Institute.
are considered to be composite members. AISC 2001. Manual of Steel construction, Load and Resis-
tance Factor Design. 3rd Ed. Chicago: American Institute
of Steel Construction.
Bridge, P. Q. & Roderick, J. W. 1978. Behavior of built-up
composite columns. Journal of the Structural Division,
ASCE, Vol 104, No. ST7, July, 11411165. ASCE.
Furlong, R. W. 2000. Should I use ACI 318 or LRFD for
designing composite columns? ACI SP 196-7, 125140.
Farmington Hills, MI: American Concrete Institute.
Galambos, T. V. (ed.) 1998. Guide to Stability Design Criteria
for Metal Structures, 5th Ed. NewYork, NY: John Wiley &
Sons.
Mirza, S.A., Lacroix, E.A. 2004. Comparative strength analy-
ses of concrete-encased steel composite columns. Journal
of the Structural Engineering, ASCE, Vol 130, No. 12,
Figure 1. Rib B P-M interaction curves and load ratings. December, 19411953. ASCE.
157
Rope access bridge inspections
B. Leshko
HDR Engineering, Inc., Pittsburgh, PA, USA
ABSTRACT Table 1. Representative Rope Access Bridge Inspections.
Bridge Inspection has received the spotlight since the Client Bridges
tragic collapse of the I-35W Bridge in Minneapolis,
MN. Proper bridge inspection is dependent upon ade- AKDOT&PF 12 Fracture Critical Highway Bridges
throughout Alaska (statewide)
quate access to the structure. Since you cannot inspect
AKDOT&PF 36 Fracture Critical Transfer Bridges
what you cannot see, visual inspections rely upon the throughout Alaska (statewide)
inspector having the requisite access to all components Mn/DOT 5 Truss Bridges throughout Central
of a bridge structure. This paper focuses on the vari- Minnesota
ous rope access methods currently available to bridge NHDOT Memorial Bridge over the Piscataqua
inspectors to supplement conventional means. River in Portsmouth, NH
HDR Engineering, Inc. (HDR) has developed rope NHDOT Sarah Long Bridge over the Piscataqua
access techniques for evaluating structures in remote River in Portsmouth, NH
locations as well as urban settings. Engineers are chal- ODOT Steel Bridge over the Willamette River in
Portland, OR
lenged to evaluate structural components within arms
reach on designated fracture critical member bridges.
Rope access is the application of specialized rope tech-
niques to place inspectors in hard-to-reach locations
in the vertical environment, especially when conven- an independent certification program. HDRs speciali-
tional man-lift, bucket truck or under bridge inspection zed bridge inspectors have received Rope Access I
vehicle access is prohibited due to insufficient deck training and annual refresher training from Skala,
width (functionally obsolete), desire not to impede Inc. (Skala), and have hundreds of hours of logged
normal travel lanes, or load posting of the structure time on ropes, exclusively performing bridge inspec-
(structurally deficient). Rope access is a work system tions for numerous owners and maintainers. All rope
using ropes and specialized hardware as the primary access inspections are performed under the supervi-
means of supporting inspectors. Rope access inspec- sion of a qualified SPRAT-certified Level III, Safety
tors descend, ascend and traverse ropes to access the Supervisor, with the training, skills, experience and
structure to perform a hands-on inspection. The sup- qualifications necessary to assume responsibility for
port of the rope completely eliminates the likelihood the entire rope access work site to include the design,
of a fall. Rope access inspectors use a back-up fall pro- analysis, evaluation and implementation of the rope
tection system in the unlikely failure of their primary access system. HDRs teaming partner, Skala, has pro-
means of support. This redundant system is achieved vided safety supervision for all rope access bridge
by using two ropes, a working line and a safety line. inspections completed to date.
HDRs rope access bridge inspectors are certified to This paper describes the planning, coordination,
Level I, Worker, by the Society of Professional Rope logistics and overall process of conducting successful
Access Technicians (SPRAT), a member-based orga- rope access field inspection of fracture critical mem-
nization that serves the rope-access industry by devel- ber bridges for representative owners and maintainers
oping and maintaining standards and administering located throughout the United States (see Table 1).
158
California bridge management
B. Newton
California Department of Transportation, Structure Maintenance & Investigations
ABSTRACT service of the states bridges. The preservation effort

is comprised of three main components: Caltrans
The Golden State of California boasts a growing, inspection and maintenance crews, major maintenance
diverse population of 38 million people. For more than contract projects and the State Highway Operation and
50 years, California has embodied a spirit of innova- Protection Program (SHOPP)
tion and entrepreneurship that has given birth to new Beginning in 2004, the California Legislature rec-
technologies and industries while building the eighth ognized the need to provide more investment in pre-
largest economy in the world. ventative activities for pavement, bridges and drainage
The backbone of the economic miracle that is assets. Caltrans prepared the first Five Year Main-
California has been the construction, operation and tenance Plan in 2005 and has since submitted two
maintenance of the finest highway system in the world. additional plans, the most recent in 2009. This plan
Today, that system of highways and bridges safely is required to be updated every two years. The primary
and reliably carries millions of people and dependably goal of the plan is to reduce or defer the costly, time-
delivers commerce to and from every corner of the consuming, major works such as rehabilitation and/or
state, the nation and the world. replacement projects.
The California Department of Transportation (Cal- Bridges like all physical assets will deteriorate
trans) is responsible for maintaining more than 50,000 over time and require maintenance, rehabilitation or
lanes miles of pavement and 12,189 bridges on the replacement.
state highway system. Caltrans bridge preservation approach is a three
As part of the Caltrans Division of Maintenance, prong attack that strives to counteract deterioration:
the Structure Maintenance & Investigations (SM&I)
1. Bridge crews employed by Caltrans address minor
unit is charged with ensuring the structural integrity of
preservation very quickly to keep bridges in the
the state highway bridge inventory and 11,790 bridges
green.
owned by local government agencies.
2. Major maintenance contracts are required when the
SM&Is staff of 179 bridge engineers, structural
scope of the work exceeds what the crews can do.
steel technicians and support staff perform bridge
3. Major maintenance contracts are designed to delay
inspections and engineering investigations in accor-
or prevent the progression into the costly red
dance with federal regulations on the state highway
rehabilitation/replacement area.
and local government agency bridges.
SM&I makes repair work recommendations, deter-
mines the safe load capacity of all bridges, reviews and REFERENCES
approves encroachment permits and air space lease
proposals involving structures, manages the bridge Caltrans, Structure Maintenance & Investigations website,
programs, delivers plans specifications and estimates http://www.dot.ca.gov/hq/structur/strmaint/
for maintenance and rehabilitation on bridge projects Caltrans, 2009, Draft 2010 State Highway Operation and
and coordinates the protective coating work on more Protection Plan
than 800 steel state highway bridges. Caltrans, 2009 Five Year Maintenance Plan, Legislative
California has an ambitious bridge preservation Report
program designed to extend the reliable, productive
159
Post-earthquake bridge inspection guidelines for New York state
J.S. OConnor
University at Buffalo, Buffalo, NY, U.S.A.
S. Alampalli
New York State Department of Transportation, Albany, NY, U.S.A.
ABSTRACT
The New York State Department of Transportation

(NYSDOT) employs a rigorous bridge evaluation pro-
gram to assess the condition of its bridges and uses this
information to formulate its inspection, maintenance
and capital programs to improve safety of the bridge
network.
In general, the bridge failure history in New York
State shows that the greatest potential for serious
bridge damage is not due to earthquakes; it is from
hydraulic forces imposed on the substructure and
corrosion and/or fatigue in superstructure elements.
Though earthquakes are not the number one concern
in NYS, the NYSDOT is taking steps to prepare for
this low probability, but possibly high consequence
scenario.
In the absence of nationally accepted standards for
post-earthquake bridge inspection procedures, NYS-
DOT initiated a research project to develop the guide-
lines. These guidelines are under development and
are briefly described in this paper. This earthquake
response plan (see Figure 1) is for any earthquake
of intensity equal to or greater than M3.5. The plan
gives guidance on initial damage assessments, man-
aging subsequent bridge inspections, follow-up action,
and repairs that may be necessary. Though this proce-
dure is tailored to NYSDOT needs, it may be useful
to other highway transportation agencies with some
modifications.
Figure 1. Proposed process flowchart for post-earthquake

damage assessment program.
160
A framework for evaluating the impact of structural health monitoring

on bridge management
M. Pozzi & D. Zonta

DIMS, University of Trento, Trento, Italy
W. Wang
Weidlinger Associates Inc., Broadway, Cambridge, MA, USA
G. Chen
Department of Civil, Architectural and Environmental Engineering, Missouri University of Science and
Technology, Rolla, MO, USA
ABSTRACT condition of the existing bridge. The neural network

based technique can not only detect each damage sce-
Although structural monitoring has been recognized nario successfully, but also identify the location of
as a powerful information tool, bridge managers often changes correctly. Indeed, the maximum identification
make decisions based on their experience or on com- error for the stiffness ratios is less than 5 percent for
mon sense, somehow regardless of the action sug- all the conditions used in the training of the networks.
gested by instrumental damage detection algorithms. To illustrate how the procedure for estimating the
The first reason for this is that damage detection is VoI applies to the Bill Emerson Memorial Bridge, we
normally affected by modeling errors and environmen- consider the case of a single interrogation of the system
tal noise: damage features are not deterministically after a seismic event. Keeping the problem as simple
related to the state and decision makers will weight as possible, we arbitrarily consider in this exercise only
differently the outcomes of the detection based on two possible states (damaged and undamaged) and two
their prior perception of the state of the structure. The decision options (do-nothing and inspection). Using a
second point is that infrastructure owners are very con- finite element model (FEM), we estimate the proba-
cerned with the consequences of wrong action, and so bility distributions of the response of the bridge for a
will decide keeping in mind the possible effects of the possible damage scenario involving formation of plas-
action they can undertake. In this paper we propose a tic hinges at the intersection of tower columns and cap
rational framework to include the impact of the men- beams. A Monte Carlo analysis has been performed to
tioned issues on decision making. The former issue estimate the empirical likelihood distributions in the
can be addressed formally by using Bayesian logic two states: numerical simulations have been carried
(MacKay 2003, Beck & Katafygiotis 1998), while out on the FEM in the damaged and intact situa-
the latter point suggests the problem be recast in the tions, introducing various sources of disturbances. The
more general framework of decision theory (Russell & example shows how it is possible to estimate the eco-
Wefald 1991). To quantitatively rate the benefit of nomic benefit of a monitoring system for any event
health monitoring on bridge management, we will which requires a damage assessment.
use the concept of Value of Information (VoI), which
represents the money saved every time the manager REFERENCES
interrogates the monitoring system after a potentially
damaging event. Beck, J.L. & Katafygiotis, L.S. 1998. Updating Models and
The methodology is applied to the Bill Emerson Their Uncertainties, I: Bayesian Statistical Framework,
Memorial Bridge, a new 1206 meter long cable-stayed Journal of Engineering Mechanics, 124(2), 455461.
structure across the Mississippi River, instrumented MacKay, D.J.C. 2003. Information Theory, Inference and
with an 84-channel seismic instrumentation system. Learning Algorithms, Cambridge University Press, Cam-
The approach developed for processing the monitoring bridge, UK.
Russell, S.J. & Wefald, E. 1991, Do the Right Thing, MIT
data recorded on the bridge is based on two sequen-
Press.
tial neural networks (Wang et al. 2007), respectively Wang, W., Chen G. & Hartnagel, B.A. 2007. Real-time con-
trained for modeling i) the healthy response of the dition assessment of the Bill Emerson Cable-stayed bridge
bridge and ii) the relation between damage and vari- using artificial neural networks, Proceedings of SPIE,
ation in the response. After the training process, the the International Society for Optical Engineering ISSN
two networks are applied to evaluation of the current 0277786X.
161
Bridge management and inspection for the County of Baltimore, Maryland
J.K. Shaffer & M.C. Schellhase

InspectTech Systems Inc.
ABSTRACT significantly improve both the inspection and man-

agement processes. In the latest inspection cycle an
Baltimore County owns over 350 bridges covering a inspection team consisting of five different consulting
wide range of types and lengths. Its prior methods of companies was selected. Their experiences using the
utilizing bridge data for its management and inspec- software along with how it was enabled to support mul-
tion processes were typical of that of many other cities tiple field teams and offices to work together will be
and counties: largely paper based with a handful of presented. The companies used the software on a vari-
spreadsheets and small disjoint databases. For its lat- ety of hardware ranging from pen-enabled tablet com-
est inspection cycle the county adopted an advanced puters to web-based office clients. Electronic forms
bridge inspection and management system to organize were created to meet the countys requirements and
all of its bridge information covering a wide variety of allow for entry of all information from the inspection.
structures. The countys usage and goals of its new manage-
This paper will cover an overview of the countys ment software which utilizes the electronic inspection
needs and the solutions that have been developed to data will also be discussed. Baltimore County person-
nel are now able to review all reports via a secure web-
based browser and utilize powerful tools to instantly
track changes occurring, highlight problem areas, and
generate a wide range of needed reports. This new
approach has allowed the county to improve its anal-
ysis and accuracy of inspection data while assisting
in identifying problem early allowing a more proac-
tive approach toward bridge management instead of a
reactive approach.
REFERENCES
Federal Highway Administration., 1995. Recording and Cod-
ing Guide for Structure Inventory and Appraisal of the
Nations Bridges, December 1995, FHWA-PD-96-001,
and Errata Sheet, March 2004, Washington, D.C.
Federal Highway Administration (FHWA), Bridge Inspec-
tors Reference Manual, December 2006, Washington,
D.C.
Maryland Bridge Inspection Coding Manual and Inspection
Figure 1. Software components by functions (Montgomery Forms: Maryland SHA Office of Bridge Development:
Implementation). Baltimore, MD 2004.
162
Overview of remote structural health monitoring: Focus on end

user demands
T. Spuler, G. Moor & R. Berger

mageba SA, Blach, Switzerland
Monitoring of bridge structures has developed from a

time-consuming, labour-intensive manual exercise to
highly automatic systems, benefitting from the tech-
nologies of the information age. The systems are now
highly automated, independent and versatile. They can
be tailored to provide almost any type of information
that may be required on the condition of a bridge and its
environment and loadings. This paper describes some
of the types of monitoring system available and the
purposes they can serve:
BASIC systems with a low measurement fre-
quency (15 minutes). This system is fully automatic
and independent, offering on-line analysis and alert
functions, with low initial investment and low run-
ning costs. Suitable for any applications which
require slow movement observation.
High-end and bespoke monitoring systems
Figure 2. ADVANCED monitoring system ROBO
ADVANCED, offering individual and integrated CONTROL installed at Weyermannshaus Viaduct in Bern
system design solutions and on-line analysis for Switzerland.
highest demands. Measurement frequencies of
1001000 Hz are possible. Uses of such elabo- bridge. It offers off-line data analysis with compre-
rate systems would include monitoring of complete hensive results. Particularly suitable for assessment
bridge behavior, or any specific needs of the bridge of the condition of a bridge or individual bridge
owner which cannot be fulfilled by the simpler parts (e.g. cables).
systems available.
Temporary-use systems with high measurement An overview of executed projects is presented, to
frequency (100 Hz). This type of system offers illustrate the diversity of modern monitoring solu-
an easy and quick health check of a bridge. It is tions. In addition end user demands are discussed, and
portable and therefore can be used on more than one the gained experience and challenges from practical
applications are described.
REFERENCES
Wenzel H. 2009, Health Monitoring of Bridges: Wiley, a John
Wiley and Sons, Ltd, Publication.
Wenzel H., Pichler 2005, Ambient Vibrations Monitoring:
Wiley, a John Wiley and Sons, Ltd, Publication.
Figure 1. BASIC monitoring system ROBO CONTROL

installed at Incheon Grand Bridge in Seoul Korea.
163
Methods of quality control and quality assurance for highway bridge

inspection in the United States
G.A. Washer
University of Missouri, Columbia, MO, U.S.A.
ABSTRACT Methods for ensuring the quality of bridge inspec-

tions are typically described as QC, procedures
Highway bridge inspections in the United States rely intended to assure quality is maintained at a cer-
primarily on visual assessments to determine and tain level, and QA, methods intended to assure the
record the condition of a bridge. These assessments effectiveness of QC and verify or measure the overall
typically result in a numerical rating for components or quality of the program.
elements of a bridge. The National Bridge Inspection Due to the diversity of bridge inspection programs
Standards (NBIS) require that States report the condi- in the U.S., there has been a variety of approaches
tion of the bridge deck, superstructure and substructure implemented to provide QA within the respective pro-
on a scale of 0 throught 9 (CFR, 2004). These condition gram. As a result, there was a need to develop generic
ratings are subjective in nature, generally describing models that described different approaches to QA in
the state of observed deterioration or damage. a generalized way. Using these models, it would then
States meet the requirements of the NBIS through be possible to efficiently characterize the methodology
a variety of inspection schemes and processes, includ- used within a specific state, evaluate different options
ing the use of element-level inspections and other or methods that could be implemented in the future,
approaches to meet the needs and requirements of indi- and identify methodologies that best met individual
vidual states. In addition to the variability of inspec- needs. A review of the existing QA programs in the
tion schemes, organizational structures of states vary US was conducted through a comprehensive literature
widely. Because of the diversity in approaches to review, assembly of codes and guidelines from vari-
bridge inspection in the U.S., there have been dif- ous states, and interviews with key personnel within
ferent approaches implemented by states to ensure states. Based on this review, the characteristics of
the quality of bridge inspection practices. Revisions various bridge inspection programs were summarized
to the NBIS in 2004 required states to implement into four general models: the Independent Oversight
systematic quality control (QC) and quality assur- Model (IOM), the Control Bridge Model (CBM), Col-
ance (QA) measures. These measures are intended laborative Peer Review Model (CPR), and the Field
to address variability in inspection results, ensuring Verification Model (FVM). Each of these models has
consistency and accuracy of inspection results within distinct characteristics in terms of how QA reviews
states. To assist States with implementing the new are conducted and quality evaluated. The models are
requirements for bridge inspection QC/QA, the FHWA described in more detail in the body of the paper.
provided a recommended framework for QC/QA for Approaches common for the implementation of QC
bridge inspection. To enhance this framework and pro- are also briefly described.
vide further guidance to states implementing QC/QA
programs, a guideline was developed that summa-
rized the methods of QC/QA currently implemented REFERENCES
in the U.S., and provides generalize or generic descrip-
tions of the different approaches such that states could CFR, 2004. National Bridge Inspection Standards. 23 CFR
part 650, USA, 7441974439.
implement methodologies to meet their programmatic
Washer, G. A., and Chang, C. A. 2009. Guideline for Imple-
needs (Washer 2009). This guideline provides infor- menting Quality Control and QualityAssurance for Bridge
mation on the typical processes and methods for bridge Inspection. NCHRP/AASHTO, Washington, D.C.
inspection QC/QA, methods of measurement, and
characteristics of effective programs.
164
Studies on the use of KEEL software for intelligent analyzing of bridge

load-bearing capacity
P. Wjtowicz
Faculty of Computer Science and Management, Wrocaw University of Technology, Poland
D. Krl
Institute of Informatics, Wrocaw University of Technology, Poland
ABSTRACT Table 1. RMSE values of experiments second stage for

tested methods in comparison with results of the expert tool,
The goal of this paper is to work out an effective com- normalized and not normalized input data.
puter method to assess the bridge structure condition
that is expressed by means of load-bearing parame- Normalized data Non-normalized data
ter considering typical bridge damages. In order to Methods Avg. Max Avg. Max
find the proper algorithm a survey of available regres-
sion methods within the KEEL software was made. RBFN 50.4 84.5 50.4 84.5
KEEL software is used to assess evolutionary algo- COR_GA 69.1 100 19.5 54.4
rithms for common data mining problems. Selected Fuzzy-GAP 26.4 47 12.4 12.9
Fuzzy-SAP 24.7 43.3 9.2 19.1
methods were tested using real data that were pro-
Thrift 48.4 67.1 7.1 8.6
vided by experts and obtained results were compared to MyBriDE 3.5 3.9 3.5 3.9
results assessed by the MyBriDE expert tool. All con-
sidered algorithms base either on the neural network
or fuzzy rule learning approach.
First stage of the experiment focused on selecting than default algorithms parameter values. Presented
a group of best methods taking into account mean results given by KEEL software are very useful for
squared error (MSE), standard deviation (STDEV) and planned implementation of intelligent tools for the
mean response time (MRT) for each algorithm. Ver- Bridge Management Systems.
ified methods use normalized input data where the
considered range is the interval [0, 1]. After this initial
selection RBFN, COR_GA, Fuzzy-GAP, Fuzzy-SAP
and Thrift algorithms were promoted for further tests. REFERENCES
The aim of experiments second stage was to com-
pare previously selected methods to the MyBriDE Alcala-Fdez, J., Sanchez, L., Garcia, S., del Jesus, M. J., Ven-
expert tool. In order this comparison was possible the tura, S., Garrell, J. M., Otero, J., Romero, C., Bacardit, J.,
structure of input data sets was reorganized to use the Rivas, V. M., Fernandez, J. C. & Herrera, F. 2009. KEEL: A
same datasets as the expert tool was tested on.Also root Software Tool to Assess Evolutionary Algorithms to Data
mean squared error (RMSE) and mean absolute error Mining Problems. Soft Computing 13(3): 307318.
Bien, J., Kaminski, T. & Rawa P. 2006. Technology and pilot
(MAE) are used to determine the quality of possible version of export tool supporting evaluation of degrada-
predictions. tion level for masonry bridges with damages. SPR 78.
The comparison between tested methods and the Wrocaw: Wrocaw University of Technology.
expert tool shows that only two out of five consid- Casillas, J., Cordon, O. & Herrera F. 2002. COR: a methodol-
ered algorithms using not normalized input data give ogy to improve ad hoc data-driven linguistic rule learning
results that are similar to expert tool predictions. These methods by inducing cooperation among rules. IEEE
methods are Fuzzy-GAP and Thrift with mean RMSE Transactions on System, Man and Cybernetics, Part B:
for all data sets respectively 12.4% and 7.1% whereas Cybernetics 32(4): 526537.
the reference RMSE value is 3.5% (see Table 1 for Cordon, O. & Herrera, F. 1997. A three-stage evolutionary
process for learning descriptive and approximate fuzzy
details). Mainly because of a long execution time of the logic controller knowledge bases. International Journal
single algorithm only default parameter values were of Approximate Reasoning 17(4): 369407.
applied so it can be assumed that previously mentioned Garcia-Pedrajas, N., Garcia-Osorio, C. & Fyfe C. 2007. Non-
Thrift and Fuzzy-GAP methods are able to provide linear Boosting Projections for Ensemble Construction.
even more accurate predictions when applying other The Journal of Machine Learning Research 8: 133.
165
MS4: Research & applications in bridge health monitoring
Organizers: F.N. Catbas, J.R. Casas, H. Furuta & D.M. Frangopol
Design and implementation of load cell bearings to measure dead and live
load effects in an aged long span bridge

Drexel University, Philadelphia, PA, United States
ABSTRACT validate the selected design, a prototype was devel-

oped and tested in the laboratory under various axial
In the spring of 2010, the expansion bearings of the forces, unidirectional and bidirectional moments, and
Burlington-Bristol Bridge were replaced due to a poor shear forces. Once validated, 14 smart bearings
condition rating (that caused the bridge to be clas- were fabricated and installed on the Burlington-Bristol
sified as structurally deficient). The replacement of Bridge. To verify their proper operation, a series of
these bearings represented a unique opportunity to load tests were carried out following installation. This
configure the new bearings to monitor the dead load paper and presentation will detail the development
and live load actions as well as their variation with and validation activities for the smart bearings as
environmental conditions. Towards that end, a series well as the results from the first few months of long-
of trial designs were developed with various bearing term monitoring. In addition, non-technical challenges
types, load cell configurations, etc. These candidate associated with developing appropriate design and
designs were then evaluated through a series of finite performance specifications for the bearings will be
element analyses and a single design was selected. To discussed.
169
Maintenance monitoring system design of a movable bridge
F.N. Catbas, H.B. Gokce, T. Dumlupinar, R. Zaurin, M. Gul, T. Terrell & D. Harper
University of Central Florida, Orlando, Florida, United States
D.M. Frangopol
Lehigh University, Bethlehem, Pennsylvania, United States
ABSTRACT challenges, d) analysis and information generation for

safety, reliability and maintenance.
Movable bridges are important intersection points The instrumentation plan is designed to monitor
for both highway and marine traffic. Federal agen- the most critical electrical, mechanical, and structural
cies and movable bridge owners in the United States components. The current installation consists of an
broadly address the need to effectively manage and array of 162 sensors which add up to 200+ channels.
maintain these complex structures. Movable bridges The monitored structural components include main
are unique structures due to the complex interaction girders, floor beams, stringers and live load shoes. As
between their structural, mechanical and electrical for the mechanical and electrical components; elec-
systems and mechanisms. These mechanisms pro- trical motor, gear box, shafts, open gear, rack and
vide versatility to movable bridges; however, their pinion and trunnions are monitored with various sen-
intricate interrelation also produces some operation sors. Figure 2 shows some examples of the monitored
and maintenance challenges. Problems related to their locations. In addition, a weather station is also installed
locations (e.g. proximity to waterways) and fatigue to monitor and correlate the environmental factors.
(stress fluctuations during the operation) also increase Since many different components of the bridge
the maintenance cost of the movable bridges which are monitored, it is clear that different methodologies
is significantly higher than that of the fixed bridges. should be used to obtain maximum information about
Operation, maintenance and repair are particular chal- the safety, reliability and maintenance of the bridge.
lenges for movable bridges because of their unique The data from structural, mechanical and electrical
design. Therefore, continuous monitoring of these parts are analyzed by using different methods. For
structures is essential to improve maintenance oper- example, high speed strain and acceleration data from
ation, to decrease the maintenance costs, to track structural elements are analyzed and correlated with
and evaluate performance and to provide solution traffic video stream to detect unusual behavior (Fig-
alternatives to these issues. ure 2). The final aim of the SHM system is to mitigate
In this study, the authors first provide an overview both problems and maintenance costs by providing
of the recent experiences from monitoring design and useful information to the bridge owners and bridge
implementation on a representative movable bridge engineers.
in Florida. The paper reviews a) issues and monitor-
ing needs for the maintenance and safety of movable
bridges, b) design of the sensing, data acquisition
and communications, c) field implementation and
Figure 2. Example strain data from the main girder during

Figure 1. Some of the monitoring items. regular traffic.
170
Exploring indirect vehicle-bridge interaction for bridge SHM
F. Cerda1 , J. Garrett, J. Bielak, R. Bhagavatula & J. Kovacevic

Carnegie Mellon University, Pittsburgh, Pennsylvania, USA
ABSTRACT locations. We then applied the multiresolution clas-

sification system to the simulated responses to these
In this paper, we explore an indirect measurement damaged states in order to determine how accurately
approach for bridge structural health monitoring the damage level could be classified. We conducted
(SHM) that collects sensed information from the two experiments.
dynamic responses of many vehicles travelling over Experiment 1: Existence of damage classification.
a bridge, and then makes extensive use of advanced In this experiment, we attempted to classify the bridge
signal processing techniques to determine information into one of two categories: Undamaged; or Dam-
about the state of the bridge. aged. Results indicate that this approach is able
Most of the existing literature addresses direct mea- to achieve very high average accuracy of 88.4% in
surement approaches, in which sensors are placed on determining the existence of damage.
the structural elements from which one wishes to Experiment 2: Severity of damage classification.
collect information to be used for the damage iden- This experiment considers five classes/labels: an
tification. Yang et al. presented an indirect approach Undamaged class and four Damaged levels. Over-
in 2004, with the sole objective of extracting bridge all, the multiresolution approach achieved an average
frequencies from the dynamic response of a moving accuracy of 71.2% in determining the severity of the
vehicle. This was later validated in a real case exam- damage.
ple (Lin et al. 2005). We have extended the use of this The results of these two experiments, while limited
indirect approach to SHM of bridges. and very preliminary, seem promising. We are encour-
We expect that using an indirect approach to SHM aged to further pursue the refinement and evaluation
will have a number of potential advantages. In this of this approach.
paper we discuss some issues such as: powering sen-
sors, the lifespan of structures is much greater than
the current reliable lifespan of most sensors, protec- REFERENCES
tion from environmental conditions and the threat of
vandalism, traffic interruption and the potential for a Chebira, A., Barbotin, Y., Jackson, C., Merryman, T.,
broader coverage of the entire bridge population. Srinivasa, G., Murphy, R.F., & Kovacevic, J. 2007. A mul-
This approach is based on the collection of vehicle tiresolution approach to automated classification of pro-
dynamic response data and the use of multiresolu- tein subcellular location images. BMC Bioinformatics,
tion image classification techniques on the collected vol. 8, no. 210.
data. In particular, we make use of a new multires- Lin, C.W. & Yang Y.B. 2005. Use of a passing vehicle to scan
the fundamental bridge frequencies: An experimental ver-
olution classifier (Chebira et al. 2007) developed by
ification. Engineering Structures 27, no. 13 (November):
Kovaevis lab for image processing. 18651878.
To test the validity of this approach, we created a Yang, Y.B., Lin C.W., & Yau J.D. 2004. Extracting bridge fre-
numerical model of the interaction between a simple quencies from the dynamic response of a passing vehicle.
oscillator and a simple beam. We then subjected the Journal of Sound and Vibration 272, no. 35 (May 6):
beam to different levels of section loss at different 471493.
1
Also affiliated with the Universidad de Concepcin, Con-
cepcin, Chile.
171
Novelty detection based on symbolic data analysis applied to structural

health monitoring
A. Cury
Paris-Est University, Laboratoire Central des Ponts et Chausses, Paris, France
C. Cremona
Direction for Research and Innovation, MEEDDM, La Dfense, France
ABSTRACT
Structural health monitoring is a problem that can be

addressed at many levels. Generally, the objective is to
determine whether damage is present or not based on
the analysis of measured dynamic characteristics of a
monitored system.
Dynamic measurements can easily contain over
thousands of values making an analysis process exten-
sive and prohibitive. Nevertheless, several damage
detection methods exist in the literature based on
signature principles, but they usually fail when mak-
ing them practical. In this sense, despite the current
processing power of computers, the necessary compu-
tational effort to manipulate large datasets remains a Figure 1. Strengthening system and procedure.
problem. Table 1. Description of tests.
In this paper, the concept of Symbolic Data Anal-
ysis is used in order to classify different structural State Phase Tests
behaviors (Billard et al. 2006). For this purpose, raw
information (acceleration measurements) but also pro- Before TGV1A, TGV2A, . . ., TGV15A
cessed information (modal data) are used for feature During 1 TGV1R, TGV2R, TGV3R
During 2 TGV4R, TGV5R, TGV6R
extraction.
During 3 TGV7R, TGV8R, TGV9R
This paper proposes an original approach in which During 4 TGV10R, . . ., TGV13R
the SDA is applied to three well know classification After TGV1B, TGV2B,, TGV13B
techniques: Bayesian Decision Trees, Neural Net-
works and Support Vector Machines (SVM). Results Table 2. Classification obtained by the SVM method.
regarding experimental tests performed on a railway
bridge in France are presented to highlight advan- Detected class
tages and drawbacks of the described methodology. A
dynamic monitoring of this railway bridge was per- Test True class Signals Frequencies Modes
formed to characterize and quantify the effect of a
strengthening procedure (Fig. 1). In summary, three TGV3R 1 1 1 1
sets of dynamic tests were performed: before strength- TGV4R 1 1 1 1
TGV5R 1 2 1 1
ening (15 tests represented by the letter A), during (13
TGV9R 2 1 1 2
tests represented by the letter R) and after (13 tests rep- TGV10R 2 2 2 2
resented by the letter B). Also, the 13 tests performed
during the strengthening were divided into 4 phases, achieved. In this case, only testTGV5R was incorrectly
as Table 1 shows. classified using the signals. Moreover, testTGV9R was
In general, the SDA combined to the cited methods incorrectly classed by signals and frequencies.
are efficient to classify and to discriminate structural
modifications, either considering the vibration data or REFERENCES
the modal parameters.
Table 2 presents some of the results obtained when Billard, N. & Diday, E. (ed.1) 2006. Symbolic Data Analysis.
SVM is used. It can be seen that reasonable results are Chichester: Wiley.
172
A temporal multi-scale analysis based bridge health divide-and-conquer

monitoring and diagnosis strategy
D. Dan & M. Zhihua

S. Limin
State Key Laboratory for Disaster Reduction in Civil Engineering, Tongji University, Shanghai, China
ABSTRACT
In present paper, a temporal multi-scale monitoring

data process and interpretation scheme is proposed for
bridge health monitoring and condition diagnosis. The
signal time domain decomposition skills are applied
to obtain the different components corresponded to
different temporal scale information. For each tempo-
ral scale, the decomposed component is related and
interpreted with a dominant physical or mechanical
environment monitoring quantity.After that, the differ-
ent features which act as a token of structure behaviors
and operation conditions in different temporal scales
are extracted by proposed with a divide-and-conquer
algorithms collection. The proposed strategy is illus-
trated with two case studies on strain monitoring data
process of Donghai Bridge Health Monitoring System
and a freeway Bridge Health Monitoring System.
For a given civil engineering structures, both its Figure 1. Divide-and-conquer strategy for interpretation of
behaviors and characteristics takes on random, time- monitoring data and status assessment and diagnosis.
variant and multi-scale features. As a reflection of
structural behaviors and characteristics, the measur- damage index can be found and therefore a more effec-
ing data also takes on these features. The Physical tive monitoring, data interpreting and health status
and mechanical environment and structural response evaluation can be achieved.
have a significant temporal multi-scale nature. This This paper is supported by NSFC project
information has different variation rules on different 50978196; Free research project of Nation laboratory
temporal scale. The temporal multi-scale phenomenon of civil engineering anti-disaster SLDRCE09-D-01;
discussed above can be illustrated in figure 1. Provice-Ministry joint project of Guangdong Provin-
The long term acceleration signals monitoring from cial Science and Technology Department & Ministry
Donghai health monitoring system are interpreted in a of Education, PRC, 0812226100023
viewpoint of temporal multi-scale. Second case study
concerned of the dynamical strain data which col-
lected from a freeway bridge in north China. One REFERENCES
way decomposition tactics are adopted to give a
divide-and-conquer treatment on these strain data. Zhihua Min, Limin Sun, Danhui Dan. Effect analysis of envi-
The real monitoring case investigations shows that ronmental factors on structural modal parameters of a
under the guidance of divide-and-conquer strategy, the cable-stayed bridge. Journal of vibration and shock (in
Chinese). 2009, V10:99105.
information contained in the different time scales can
Sandeep T. Vohra, Bryan Althouse, and Gregg Johnson.
be discovered one by one. And due to the decom- Quasi-Static Strain Monitoring During the Push Phase
posing, the interference among the numerous factors of a Box-Girder Bridge Using Fiber Bragg Grating Sen-
can be separated distinctly. More important is that by sors. European Workshop on Optical Fibre Sensors, Pee-
decomposing original signals into its temporal scale bls Hydro, 07.1998, Scotland. Proc. SPIE, Vol. 2721,
counterpart, high quantity, featured performance or 149 (1996).
173
Validation of an SHM procedure for concrete bridges based

on static strain records
A. Del Grosso & F. Lanata

Department of Civil, Environmental and Architectural Engineering, University of Genoa, Italy
D. Posenato
Smartec S.A., Manno, Switzerland
A. Mercalli
Autostrade per lItalia, Romagnano Sesia, Italy
ABSTRACT
The paper is aimed at presenting the results of a

research activity conducted jointly by the University
of Genoa, Smartec S.A., and Autostrade per lItalia
S.p.A. The experiment consists in the continuous
monitoring (for more than one year) of strains and
temperatures in two pre-stressed concrete beam spec-
imens, placed in open air and one of which has been
subjected to increasing damages (Fig. 1).
Data interpretation is dedicated to find a relation-
ship between the measured static structural response
and features related to damage and/or degradation.
The research focuses on the experimental validation
of different damage identification algorithms; in this
work Proper Orthogonal Decomposition (POD) and
correlation analysis will be presented.
The first attempts to use real field measurements Figure 2. Complete time histories of temperature T1 and
(Fig. 2), without any data pre-processing, to identify strains at sensor locations Sn1 and Sn5.
the already induced damages have apparently failed, seems to be effective in reducing the strain variations
producing false positives or unclear evidence of dis- and the dead load positioning can be clearly identi-
turbances in the features of the signals associable to fied from all extracted eigenvectors (POD analysis).
damage. This difference between numerical simula- Some doubts still remain regarding the identification
tions and field experiments is most likely referable to of damages. Its in the Authors opinion that the intro-
one or both of the following reasons: a) the damages duced damage cannot be detected as a punctual event
have been induced too early, when the reference period due to the pretension force but they could be visible
was not long enough and the signals were not stabi- after a certain period as a degradation of the beams
lized; b) the intensity of the effect of damages was too performance with time. The consideration related to
low and hidden by the effects of irregular or anoma- damages introduced too early before the signal stabi-
lous temperature changes. In fact, the work shows that lization seems to be also confirmed by the eigenvectors
environmental variations have a great effect on mea- obtained after the strain pre-processing.
surements and have to be removed from the strain time
histories before a reliable damage detection procedure REFERENCES
can be successfully applied.
Further analysis aimed at removing the environ- Gliic, B. & Inaudi, D. 2007. Fibre Optic Methods for Struc-
mental changes has been presented. The procedure tural Health Monitoring. John Wiley & Sons, England,
262 p.
Lanata, F. 2010. Health monitoring of concrete bridges:
model simulations of prestressed beams under static envi-
ronmental loading based on experimental data. IABMAS;
Proc. 5th Int. Conf., Philadelphia, 1115 July 2010.
Figure 1. Sensors location on the Beam A (Sn: SOFO Wenzel, H. & Pickler, D. 2005.AmbientVibration Monitoring.
sensor; T: thermocouple). John Wiley & Sons, England, 291 p.
174
Uncertainty and reliability analysis using monitoring data and artificial

neural network (ANN) calibration
T. Dumlupinar, H.B. Gokce, K.R. Mackie & F.N. Catbas

University of Central Florida, Orlando, FL, USA
D.M. Frangopol
Lehigh University, Bethlehem, PA, USA
ABSTRACT
Probabilistic techniques in engineering problems pro-

vide a deeper understanding of the aleatory and epis-
temic uncertainties inherent to the structures being
analyzed. Complex engineering structures are usually
analyzed with finite element techniques to incorpo-
rate all critical details with geometric representation.
The prediction of structural reliability based on a pre-
defined limit state can be obtained with a finite element
model and can be updated using Bayesian methods
with the monitoring data. Another common approach
is to calibrate a finite element model with the monitor-
ing data by minimizing the error between the analysis
and the measurements, which requires more time and
user interaction.
The objective of this paper is to explore the compar-
ison of the model responses and predictions between Figure 2. Framework of the proposed approach.
these two approaches where uncertainties are incorpo-
rated in a different manner. For this study, a test set-up
which is a simplified series-parallel physical model
with four structural elements (Double-H-Frame-DHF)
is designed and extensively instrumented with vari-
ous sensors, and monitored over time with different
structural boundary conditions.A large number of sim-
ulations using the finite element model are performed
under uncertainties associated with material proper-
ties, geometry, loading and boundary conditions. The
boundary conditions are changed gradually and the Figure 3. Reliability index change under support deteriora-
two approaches are executed to obtain the reliability tions.
distribution at each structural state and also to pre-
dict future performance. The findings from the two
approaches are compared.
Figure 1. DHF and reliability block diagram. Figure 4. Test setup and data acquisition.
175
Effects of environmental changes on the dynamic characteristic

of reinforced concrete beams
M.R. Esfahani & G. Behnam

Ferdowsi University of Mashhaad, Mashhad, Iran
ABSTRACT
A modal test program comprised ten reinforced con-

crete beam specimens, with the dimensions of 150 mm
width, 200 mm height and 2200 mm long, was con-
ducted in the structural laboratory to study the cor-
relation between the environmental changes with the
variation of the dynamic characteristics of concrete
beams. The compressive strengths of 21 MPa and
40 MPa for concrete were considered for different
specimens. Tensile and compressive reinforcement
ratios varied in different specimens. The dynamic
response of the beams was recorded using one
accelerometer. To eliminate the problem of support
effects on dynamic test results, the specimens were
suspended by elastic cables (Fig. 1). In the dynamic Figure 1. A suspended beam and elastic cables, and different
tests, the impact hammer provided the excitation input degrees of freedom of beams.
at every degree of freedom, and the frequency response
functions (FRF) were derived by measuring the input specimens. The damping values of the specimen kept
force (of impact hammer) and response (of the sta- in hot and dry conditions were considerably larger
tionary accelerometer) at each measurement point. than those of other specimens. It may be concluded
For each point, test procedure was repeated ten times that the damping value increases with the increase of
and the average of FRFs was finally stored as the temperature and concrete shrinkage.
input-output transfer function for that particular point.
Frequencies and damping values of specimens were
recorded in the model tests. From the test results it was REFERENCES
found that the frequencies of all specimens decreased
Askegaard V, Mossing P. 1998. Long term observations of
with the age of specimens. The frequencies of the RC-bridge using changes in natural frequency, Nordic
specimen kept outdoors under hot and dry conditions Conference Research, 7:207.
(during 1139 days after casting) decreased, signif- Peeters B, De Roeck, G. 2000. One year monitoring of
icantly. However, when it was brought back to the the Z24-bridge using vibration monitoring analysis, In:
laboratory after 39 days of casting, the frequencies Proceedings of European CST F3 Conference on Sys-
changed mainly with the temperature of the specimen tem Identification and Structural Health Monitoring, p.
at the test time. The average values of the frequencies 23342.
of the specimens with 40 MPa concrete strength were Xia Y, Hao H, Zanardo G, Deeks A. 2006. Long term vibra-
tion monitoring of an RC slab: temperature and humidity
approximately 18 percent larger than those of the spec-
effect, Engineering Structures, 28: 44152.
imens with 21 MPa concrete strength. The frequencies Rohmann RG, Baessler M, Said S, Schmid W, Rucker WF.
of the specimens with the larger tensile reinforcement 1997. Structural causes of temperature affected modal
ratio were larger than those of the specimens with the data of civil structures obtained by long term moni-
smaller one. No meaningful trend was seen in the rela- toring large structures, In: Proceedings of DAMAS 97
tionship between the damping values and the age of (Euromech 365), UK: Sheffeild Academic Press, p. 117.
176
Damage detection using a novel time series methodology: Application

to the Z24 Bridge data
M. Gul & F.N. Catbas

University of Central Florida, Orlando, Florida, United States
ABSTRACT
In this study, a novel time series analysis methodology

is used for detection, localization, and quantification
of damage. The methodology is based on creating
ARX models (Auto-Regressive models with eXoge-
nous input) for different sensor clusters. The output of
each sensor in a cluster is used as an input to the ARX
model to predict the output of the reference channel
of that sensor cluster. After the ARX models for the
healthy structure at each DOF are created, the same
models are used for predicting the data from the dam-
aged structure. The difference between the fit ratios is
used as damage indicating feature. The methodology is
applied to the experimental data coming from the Z24
Bridge. It is shown that the approach is successfully Figure 2. The geometry of the bridge; top and plan view
used for identification, localization, and quantification (modified from the Z24 benchmark documents).
of different damage cases. The potential and advan-
tages of the methodology are discussed along with the
analysis results. The limitations and shortcomings of
the methodology are also addressed.
The Z24 Bridge benchmark is one of the most
commonly used benchmark problems by researchers
in SHM and experimental dynamics. Therefore, the
methodology is applied to the Z24 Benchmark data
for verification. In this study, three levels of pier set-
tlement are investigated. The levels of the settlements
are 40 mm, 80 mm and 95 mm. The damaged pier
is referred as the Koppigen pier and it is shown in
Figure 2.
The Damage Features (DFs) computed for base-
line and damaged cases are shown in Figure 3. It is
observed that the DFs are higher for Setups 69 where Figure 3. The damage features for each setup.
the pier settlement occurred. This figure represents a
good illustration of the increase in the DFs for the an idea about the severity of the damage, as the DFs
damaged locations. Furthermore, the plots also give get higher with increased settlement. It is observed
that the DFs for 95 mm settlement are higher than DFs
for 80 and 40 mm settlement (except two points) and
the DFs for 80 mm settlement are higher than that of
40 mm settlement.
Figure 1. Creating different ARX models for each sensor

cluster (first sensor cluster).
177
Development of a bridge health monitoring approach using train-bridge

interaction analysis and GA optimization
X. He, T. Hayashikawa & T. Matsumoto

Hokkaido University, Sapporo, Japan
M. Kawatani
Kobe University, Kobe, Japan
H. Furuta
Kansai University, Osaka, Japan
1 INTRODUCTION actual structures with enormous possible damage pat-

terns, the soft computing methods of Neural Networks
In the maintenance scheme of Shinkansen viaducts (NN) and Genetic Algorithm (GA) are introduced and
in Japan, the health monitoring and diagnosis of the applied as follows.
structures becomes especially important because of
a) In this proposed approach, the traffic-induced
the huge number of structures. Currently, the over-
dynamic responses of the structure used to compare
all health condition of Shinkansen viaducts is mainly
with measured ones will be simulated by a devel-
examined by visual inspections, which demand a large
oped vehicle-bridge interaction analysis computer
number of technicians and also a considerable cost.
program. However, for a large-scale structure even
With the proceeding of a decreasing birthrate as well
one time of such an analysis will demand consider-
as an aging society in Japan, it is significant to develop
able computational capacities, which will leads to
more effective and economical health monitoring and
the infeasibility in an actual identification process
diagnosis approaches to the civil infrastructures. If the
that needs a great number of interaction analyses.
dynamic bridge response induced by the running trains
Therefore in this research, the NN techniques are
can be effectively used for the health monitoring and
planned to be used to simulate the running train-
diagnosis process, it will be an economical and con-
induced bridge response, which, once established,
venient way because the train-induced vibrations have
can shorten the computational time of the identi-
been continually recorded.
fication process to an acceptable degree in actual
To identify the structural characteristics of the
applications. To establish such a NN tool, it is
bridge with traffic-induced vibration data, a currently
impossible to use measured results to carry out
conventional way is to perform the inverse analy-
the supervised learning process, thus the results
sis of vehicle-bridge interaction. However, such an
from the train-bridge interaction analysis program
approach can encounter considerable difficulties in
have to be used as the sample data. Not to men-
actual applications because of numerical errors caused
tion, adequate demanded accuracy for such an
by the inverse analysis due to the large number of
analytical program is indispensable in the actual
members. In this research, a structural identification
applications.
approach is developed employing only direct analyses
b) The calculated traffic-induced bridge responses
of train-induced bridge vibration by means of intro-
under certain damage patterns described above are
ducing soft computing methods to avoid the numerical
then used for the identification process. However,
error mentioned above.
even only the possible damage patterns based on
In the actual railway viaducts, the possible dam-
engineering facts are assumed, the number can still
age patterns of the structures are comprehended by
be considerable large and difficult to identify. This
the bridge engineers based on theoretical and empir-
is a typical combinatorial optimization problem
ical facts. In this approach, the damage patterns of
and can be solved by some metaheuristic search
the bridge members are assumed in advance and
algorithms. In this research, the GA technique is
used as the input information. Then, the train-induced
adopted to find the exact damage pattern. In the GA
dynamic responses of the bridge under a certain dam-
algorithm, the assumed damage patterns are set as
age pattern are calculated by a developed analytical
the population and the difference between the cal-
procedure. In the assumed damage patterns, the one
culated results and the recorded ones is defined as
identical or nearest to the actual damage condition
the object function.
will give the most similar dynamic responses to the
recorded ones, through which the exact solution can The basic concepts and process are represented
be identified. To make this approach applicable to using simple numerical models in this paper.
178
Assessment of remaining fatigue life of aging orthotropic steel deck bridges
K. Kashefi, M. Zeinoddini & A.P. Zandi

Faculty of Civil Engineering, K.N. Toosi University of Technology, Tehran, Iran
ABSTRACT
The maintenance cost of old steel bridges is rapidly

increasing since many of them are deteriorating due to
accumulative stress cycles and reaching their design
life. Moreover, selection of orthotropic steel decks for
long-span bridges has become more popular, consi-
dering their longitudinal stiffness and low dead weight.
Despite these advantages, containing several types of
welded joints, these decks are prone to fatigue. There-
fore fatigue life estimation of these structures seems
to be an inevitable task in order to make cost effective
decisions for rehabilitation solutions.Although fatigue Figure 1. Stress histogram after filtering stress ranges below
fatigue threshold, for girder No.3.
analysis based on specification loads and distribution
factors is completely reliable for new bridge designs,
it usually underestimates the remaining fatigue life of
existing bridges. Instead fatigue evaluation using field As a classic approach, S-N curves are used in the
measurement data under actual traffic load has proven current study to estimate the fatigue life of the bridge.
to be more accurate for existing bridges. Stress cycles/ranges were extracted using a program
This paper presents a procedure using this method which was developed in MATLAB based on rain flow
for fatigue life evaluation, applied for one of aging counting algorithm proposed by ASTM (2004). Since
orthotropic steel deck bridges in Iran. Hafez-Taleghani large number of very low strains will decrease the
Fly-over is a 31-simple-span Orthotropic Deck Steel effective stress range obtained from strain histogram,
Bridge in Tehran. It carries three traffic lanes and was there should be a criterion for eliminating strain cycles
constructed in 1970. As maintained by the city urban below a defined threshold. In the current study truck
plan, this bridge and three other identical bridges were weighing 10 tons was used to define the fatigue strain
primarily a temporary solution to growing traffic of thresholds. Stress histogram after filtering the strain
the city, but they are still under service. After deter- ranges below fatigue threshold for one of the girders
mination of stress ranges using specification loads, is shown in Figure 1. Accordingly, as per AASHTO
the fatigue life of the bridges obtained insufficient for (1990) and based on the existing traffic regime the
their service life. So in order to assess the level of fatigue life of the girders is estimated to be infinite.
fatigue damages and to find out feasible mitigations
measures, an extensive plan for structural health mon-
itoring including field strain measurement, has been
accomplished. The most vulnerable details in this case REFERENCES
are the longitudinal weld between the web and bottom AASHTO, 1990. Guide specifications for fatigue evaluation
flange which experiences tension cycles. The in-field of existing steel bridges. Interim specifications in 1993
data which has been collected due to normal traffic and 1995, Washington, D. C.
and during a seven day period, processed using the ASTM, 2004. Standard practices for cycle counting in fatigue
procedures outlined in AASHTO specifications. analysis. E1049-85, ASTM, West Conshohocken, Pa.
179
Identifying bending stiffness change of a beam under

a moving vehicle
C.W. Kim
Department of Civil and Earth Resource Engineering, Kyoto University, Kyoto, Japan
M. Kawatani
T. Fujimoto
Kajima Corporation, Tokyo, Japan
ABSTRACT
Usually, to define damage for bridge structures is

difficult differently from other structures such as auto-
mobiles, aerial vehicles, etc. This is one reason why
most precedent studies particularly addressing bridge
health monitoring (BHM) has specifically examined
global change of modal properties and quantities of
bridge structures. Those modal parameters are usually
identified using the ambient vibration (e.g. Wenzel &
Pichler 2005).
For bridges with a long span, wind-induced vibra-
tions are important dynamic sources. On the other Figure 1. Identified damage location and severity of the
hand, for short span bridges which are insensitive to bridge with damage at ELEM. No.2 according to vehicle type.
the wind load, how to excite the bridge economically,
reliably and rapidly is an important task.
This study is an attempt to use traffic-induced
vibrations for the damage identification of short span
bridges using an algorithm derived from a bridge-
vehicle interactive system (Kim et al. 2008), because
a moving vehicle is able to actuate short span bridges
economically, reliably and rapidly. Feasibility inves-
tigation through a moving vehicle laboratory experi-
ment is the major goal of this study.
In experiment three vehicle models called as VT-A,
VT-B and VT-C of which the natural frequencies for
the bounce motion are respectively 2.93 Hz, 3.76 Hz
and 3.03 Hz are considered with six loading scenarios:
SCN1 of VT-A vehicle traveling with v = 0.93 m/s; Figure 2. Identified damage location and severity of the
bridge with damage at ELEM. No.2 according to vehicle
SCN2 of VT-A vehicle traveling with v = 1.63 m/s; speed.
SCN3 of VT-B vehicle traveling with v = 0.93 m/s;
SCN4 of VT-B vehicle traveling with v = 1.63 m/s; without great variation according to vehicle type and
SCN5 of VT-C vehicle traveling with v = 0.93 m/s; speed.
and SCN6: VT-C vehicle traveling with v = 1.63 m/s.
The experimental study demonstrates that better
chance to detect damage is observed by using the REFERENCES
vehicle with the frequency which is close to the nat-
ural frequency of the experimental bridge (2.69 Hz, Kim, C. W. & Kawatani, M. 2008. Pseudo-static approach for
2.59 Hz and 2.54 Hz respectively for the intact and damage identification based on coupling vibration with a
two damage scenarios.) as shown in Figure 1 as moving vehicle, Structure and Infrastructure Engineering,
4(5): 37179.
well as adopting higher vehicle speed as shown in
Wenzel, H. & Pichler, D. 2005.AmbientVibration Monitoring,
Figure 2. It can be concluded, however, that loca- John Wiley & Sons, NY.
tions and severities of damage are constantly identified
180
Low-cost wireless sensor node for vibration monitoring

of infrastructures
C.W. Kim
Department of Civil & Earth Resource Engineering, Kyoto University, Kyoto, Japan
M. Kawatani & R. Ozaki

N. Makihata & M. Kano

JIP Techno Science Corporation, Osaka, Japan
ABSTRACT
Maintaining and improving condition of civil infras-

tructures is critical to the structural integrity. Structural
health monitoring (SHM) from vibration data of the
structure is recognized as an important approach for
maintaining the structures. To activate health monitor-
ing of civil structures using vibration measurements,
developing a cost-effective monitoring method and
sensor system have been technical issues.
Experience with field-deployed systems has shown
that the wired sensor can be costly to maintain because Figure 1. Wireless sensor node.
of environmental degradation (Lynch 2004). In addi-
tion, the deployment of such system can be challenging
with a great portion of the installation time attributed
to the installation of system wires and cables for large
scale structures (Farrar et al. 2006). The integration of
wireless communication technologies into SHM meth-
ods has been widely investigated in order to overcome
the limitations of wired sensing networks. Wireless Figure 2. Acceleration and Fourier amplitude before and
after recovering missing data taken from wireless sensor.
communication can remedy the cabling problems of
the traditional monitoring system and significantly
obtained, even though noisy signals on the measured
reduce the maintenance cost.
time history taken by wireless sensors are observed.
Considering real situation, the available wireless
This study also investigates the feasibility of miss-
sensors are not necessarily designed to meet specific
ing data recovery by means of the Kalman filter.
requirements for the vibration-based SHM such as
Recovered signals (see Figure 2) show comparable
high sensitivity, power supply, cost effectiveness, etc.,
Fourier spectra profile with those of cabled sensors.
and are very limited.
However the recovered amplitude are relatively small
This paper discusses a performance experiment
in comparison with cabled sensor, which is one of the
for a wireless sensor node (see Figure 1) equipped
next challenges of this study.
MEMS accelerometer and wireless device for data
transmitting, which is aiming to design a low-cost
wireless sensor node for monitoring civil structures. REFERENCES
Performance of the prototype of the sensor node is
Farrar, C.R., Park, G., Allen, D.W. & Todd, M.D. 2006. Sen-
investigated through vibration experiments on a newly
sor network paradigms for structural health monitoring,
constructed bridge and a real pedestrian bridge. In Struct. Control Health Monit., 13(1): 210225.
the experiments, a relay node and a high performance Lynch, J.P. 2004. Overview of wireless sensors for real-time
antenna are used to improve radio transmission. Com- health monitoring of civil structures, Proc. of the 4th Int.
parable results between data taken from cabled and Workshop on Structural Control and Monitoring, New
wireless sensors such as acceleration and FFT are York City, NY, USA, June 1011.
181
Bridge retrofit design optimization for fatigue based on monitoring

and FE analysis
K. Kwon & D.M. Frangopol

ABSTRACT performed at the identified critical locations based on

field monitoring data and FEA.The proposed approach
Over the past decades, aging steel tied-arch bridges, is illustrated on an existing bridge.
which have been built with the floorbeams connected
into the tie girder with web shear connections, have
been deteriorated due to fatigue cracks caused by rel- REFERENCES
ative movements. Several retrofit strategies have been
proposed to remove structural deficiencies caused by AASHTO Guidelines 2002.AASHTO Standard Specification
for Highway Bridges. Washington, D.C.
fatigue and to preserve bridge safely up to its remain-
Connor, R.J. & Fisher, J.W. 2002. Report on Field Inspec-
ing fatigue life. As a retrofit, the softening method tion, Assessment, and Analysis of Floorbeam Connec-
by partial removal of the connection angles can be tion Cracking on the Birmingham Bridge Pittsburgh PA.
employed in order to prevent stresses within the web Lehigh Universitys Center for Advanced Technology for
gap from concentrating. However, it is not simple to Large Structural Systems (ATLSS), Bethlehem, PA.
find relevant retrofit sizes since fatigue cracks after Connor, R.J. & Fisher, J.W. 2006. Identifying Effective and
retrofit may move into other regions or reinitiate near Ineffective Retrofits for Distortion Fatigue Cracking in
the retrofit regions. For this reason, optimal retrofit Steel Bridges Using Field Instrumentation. Journal of
solutions have to be found. Bridge Engineering, ASCE, 11(6): 745752.
Connor, R.J., Fisher, J.W., Hodgson, I.C. & Bowman, C.A.
For this purpose, cut-out area for retrofit is defined
2004. Results of Field Monitoring Prototype Floorbeam
as objective, while the AASHTO constant amplitude Connection Retrofit Details on the Birmingham Bridge.
fatigue limit (CAFL) is imposed as upper and lower Lehigh Universitys Center for Advanced Technology for
bounds of stress constraints. Finite element analysis Large Structural Systems (ATLSS), Bethlehem, PA.
(FEA) is performed to investigate stress distributions Fisher, J.W. 1984. Fatigue and Fracture in Steel Bridges. John
in structural details associated with different cut-out Willey & Sons.
areas. Fatigue reliability evaluation at potential critical Fisher, J.W., Jin, J., Wagner, D.C. & Yen, B.T. 1990. NCHRP
locations is performed to estimate remaining fatigue Report 336: Distortion-Induced Fatigue Cracking in Steel
life associated with the optimal retrofit solutions. Bridges. TRB, National Research Council, Washington,
D.C.
The fatigue reliability approach is based on
Keating, P.B. & Fisher, J.W. 1986. Evaluation of Fatigue Tests
AASHTO Specifications (2002). All necessary infor- and Design Criteria on Welded Details. National Coop-
mation is obtained from both finite element modeling erative Highway Research Program (NCHRP) Report,
(FEM) and field monitoring data from the origi- 286, Transportation Research Board, National Research
nal retrofit size. The developed FEM is validated Council, Washington, D.C.
by comparing the analytical results from FEA with Kwon, K. & Frangopol, D. M. 2009. Fatigue Reliability
field measurements. Critical locations, where poten- Assessment of Steel Bridges Based on Monitoring and
tial distortion-induced fatigue cracking can occur, are FE Modeling. Safety, Reliability and Risk of Struc-
identified from FEA. The optimization problem is for- tures, Infrastructures and Engineering Systems, edited
by Furuta, Frangopol and Shinozuka, Taylor & Francis
mulated as the cut-out area minimization subjected
Group, London, CD-ROM, 10561061.
to stress constraints associated with the CAFL cor- Liu, M., Frangopol, D. M. & Kwon, K. 2010. Fatigue Relia-
responding to each category. The optimal solutions bility Assessment of Retrofitted Steel Bridges Integrating
are computed by linking the FEM software ABAQUS Monitored Data. Structural Safety, Esevier, 32(1): 7789.
(version 6.7.1, 2007) with the optimization software Miner, M. A. 1945. Cumulative Damage in Fatigue. Journal
VisualDoc (2006). Fatigue reliability assessment is of Applied Mechanics 12(3): 159164.
182
Updating the finite element model of bridge structures by an improved

Taguchi updating method
Y. Liu
School of Transportation Science and Engineering, Harbin Institute of Technology, Harbin, China
H. Li, Z. Duan & Y. Yao

School of Civil Engineering, Harbin Institute of Technology, Harbin, China
ABSTRACT (SN) ratio. But Taguchi method is only applied to solve

the single objective optimization problem of model
Structural finite element (FE) model updating is the updating. Therefore, this paper proposed improved
core technique of vibration-based structural health Taguchi updating method to deal with the problem
monitoring (SHM) of bridge structures since this tech- of model updating using multi-objective optimization
nique can provide an accurate baseline model for technique. The classification of the searching popu-
damage detection, structural control and safety assess- lation and the new objective function (the new SN
ment of bridge structures. In the procedure of model ratio) are introduced into the Taguchi updating method,
updating, the objective function is usually set as the which makes the improved Taguchi method have the
weighted sum of the difference between analytical ability to search the Pareto-optimal set of FE model
and experimental dynamic characteristics of structures updating.
(M.I. Friswell, 1995). But it is difficult to select the Finally, a parameterizing scheme for structural
weighting factors since the relative importance of each boundary condition is proposed, and the improved
parameter to updated results is not obvious but specific Taguchi updating method is employed to update the
for different problem. boundary conditions of a truss bridge structure. The
To overcome above problem, multi-objective updated results show that the proposed method is
genetic algorithm (GA) (Gyeong-Ho Kim, 2004) is promising for FE model updating of bridge structures.
introduced into FE model updating since there is no
need for selecting weighting values. In practical situ-
ations, however, it is difficult to update the FE model
REFERENCES
of bridge structures by GA owing to the relative low
efficiency. Taguchi updating method, deemed as an M.I. Friswell, J.E. Mottershead, 1995. Finite Element Model
efficient and robust updating method, is a good alter- Updating in Structural Dynamics. Kluwer Academic Pub-
native to GA for updating the FE model of large-scaled lishers.
structures. Gyeong-Ho Kim, Youn-Sik Park, 2004. An Improved Updat-
Taguchi method (Kye-Si Kwon, 2005) is a good ing Parameter Selection Method and Finite Element
replacer of GA to update the FE model of the large Model Update Using Multi-objective Optimization Tech-
structures. On one hand, Taguchi method has more nique. Mechanical Systems and Signal Processing, Vol.
18, p: 5978.
efficient than GA since orthogonal arrays (OA) (A. S.
Kye-Si Kwon, Rong-Ming Lin, 2005. Robust finite element
HEDAYAT, 1999) are applied to screen the main effect model updating using Taguchi method. Journal of Sound
of parameters to objective function rather than stochas- and Vibration, Vol. 167, p: 347374.
tic search. On the other hand, the updated results of A. S. Hedayat, N. J. A. Slaone and J. Stufken, 1999. Orthogo-
Taguchi method are robust against various noises since nal Arrays-theory and Applications. New York: Springer,
parameters are updated to maximize the signal to noise Verlag.
183
High-order local vibration properties of RC Viaduct

under the passing high speed train
K. Matsuoka & K. Kaito

Osaka University, Suita, Osaka, Japan
T. Watanabe & M. Sogabe

Railway Technical Research Institute, Kunitachi, Tokyo, Japan
ABSTRACT Table 1. A result of identification by the experiment with a

passing train (central slab).
In order to sustain the development of high-speed
trains, it is necessary for railway bridges to grasp the
dynamic behavior of bridge members precisely, by
identifying not only whole bridge vibration proper-
ties but also local ones, and high-order mode vibra-
tion properties. Actually, there are common concerns
noises caused by the vibration of bridge members
due to high-speed trains. However, there have been
few vibration measurement experiments. In this cir-
cumstance, authors conducted the passing train exper-
iment, measuring vibration concurrently at several
points on the intermediate and projecting slabs of a
RC viaduct for high-speed trains. Through the vibra-
tion measurement experiment, the authors verified
the possibility of detection of high-order vibration
modes of members and identified the outstanding
vibration mode when a high-speed train passes such
like table 1. In this table shows a result of vibration
measurement experiment and identification analysis
in case of central slab. As a result, the authors could
observe the modes in which the vibration of each
member is outstanding, and found that the vibra-
tion of the RC bridge due to a passing train is
mainly caused by the vibration of the central slab, REFERENCES
although there are low-order vibrations of the over-
hanging slab. It was also found that the second- Fryba, L. 1972. Vibration of solids and structures under
order mode like table 1 contributes to outstanding moving load, Noordhoff International,
components. HARA Tsuneo, et al. 2004. Development of a New Method
The future mission would be to reflect these results to Reduce SHINKANSEN-Induced Wayside Vibrations
in actual bridges. It is considered necessary to identify Applicable to RigidI Frame Bridges, Journal of Structure
Mechanics and Earthquake Engineering, Vol. 766, I-68,
the cause of such vibration problems and clarify influ- pp. 325338. (in Japanese)
ential modes and vibration modes, by discussing the SOGABE Masamichi. 2003. Dynamic response and design
relations with noise and ground vibration. In this study, of continuous concrete railway bridge in resonance area,
the authors adopted the most fundamental method for Journal of Structure Mechanics and Earthquake Engi-
identification. neering, Vol. 724, I-62, pp. 83102. (in Japanese)
184
Condition assessment of bridge deck truss using in-service monitoring

data of strain
Y.Q. Ni, H.W. Xia & J.M. Ko

Department of Civil and Structural Engineering, The Hong Kong Polytechnic University, Hong Kong
K.Y. Wong
Bridges & Structures Division, Highways Department of HKSAR Government, Hong Kong
ABSTRACT
Structural condition assessment is in great demand for

efficient management of aging in-service infrastruc-
ture. A strategy which integrates structural condition
assessment with structural health monitoring (SHM)
system is expected to be viable because a SHM system
can offer plentiful information on structural condition Figure 1. Tsing Ma Bridge and deck sections instrumented
with strain gauges.
by use of various sensors and analysis techniques.
In practice, in-service monitoring of strain response
plays an important role in structural condition assess-
ment via monitoring because strain measurement can of structural assessment at two levels: (i) deck truss
offer information about stresses experienced by the components, and (ii) deck cross-sections. Since the
structure under operational conditions. Strain moni- in-service monitoring data of strain is a result of multi-
toring and testing in the field can be classified into source combinations of live loads and environmental
two categories: long-term continuous monitoring and effect, wavelet-based multi-resolution analysis is first
short-term portable testing. During temporary strain carried out to decompose the measured strain data
testing, such as live load testing, some long-term effect sequences for signal extraction in recognizing that
on strain response can be ignored since this kind of some ingredient induced by temperature is released
testing is usually completed in short time and purposed through boundary conditions and contributes little to
to record short-term structural behavior. On the other stress quantities. Thereafter, the time-history of stress
hand, long-term continuous monitoring of strain can response in deck truss components due to external load
capture real-time structural behavior not readily appar- effects is derived from the extracted strain, and the
ent from short-term testing, for example, daily and peak values are counted to infer their statistical char-
seasonal changes. Continuous monitoring of real-time acteristics for the component-level assessment. As for
streaming of dynamic strain is emerging as a criti- the cross section-level assessment, the dynamic inter-
cal strategy in the assessment, inspection and decision nal forces such as bending moments, axial forces and
making on maintenance of bridges. shear forces on the deck cross-section are evaluated
In this study, an approach for structural condition by synthesizing the resultant internal forces of truss
assessment of bridge deck truss making use of in- members from the derived stress-time history. Struc-
service monitoring data of strain from a SHM system tural condition assessment of the deck cross-section is
is proposed and applied to the instrumented Tsing then carried out under various load conditions such as
Ma Bridge (Figure 1). The proposed method consists monsoon, typhoon, with and without railway traffic.
185
Simultaneous monitoring of the coupled vibration between a bridge

and moving trains
Y. Oshima, K. Yamamoto & K. Sugiura

Civil and Earth Resources Eng., Kyoto University, Kyoto, Japan
A. Tanaka
Chuo Fukken Consatants Co., Ltd Osaka Japan
M. Hori
Traffic Service Ltd., Osaka, Japan
ABSTRACT that both vibration systems are surely correlated to

each other and the train vibration must be affected
The purpose of this study is to evaluate the coupled by the bridge vibration. Thus this opens up possibil-
vibration between train and bridge experimentally. ity that the bridge vibration can be estimated in the
Moreover, numerical simulation on a simple beam and train vibration, and the estimation theory should be
vehicles was also conducted to obtain the fundamental developed.
information of the coupled vibration.
The monitored bridge is located in Osaka city and
belongs to Osaka city transportation authority. This
bridge is a skew steel bridge with double decks, and
the length is 39.8 m. On the basis of the actual monitor-
ing by the bridge sensors, this bridge has, in the vertical
direction, the first bending mode at 2.36 Hz, the first
torsion mode at 3.89 Hz and the second torsion mode
at 5.00 Hz, respectively. As for the horizontal direc-
tion, the peaks at 3.47 Hz, 3.89 Hz and 4.72 Hz were
recognized. The train can have the eigenfrequency of
1.40 Hz.
As a result of the monitoring, it is found that the
train eigenfrequency of 1.5 Hz can be recognized in the
power spectrum of the train response, but it is difficult
Figure 2. Power spectrum of the first car of the train.
to find the bridge frequency in the power spectrum.
However, the peak close to the bridge frequency of
2.36 Hz was found in the measured cross power spec-
trum between the bridge and train. This fact indicates
Figure 1. Picture of the monitored bridge and the train. Figure 3. Cross spectrum of train and bridge.
186
Challenges and uncertainty mitigation in structural identification of long

span bridges
J.B. Prader, J. Zhang, F.L. Moon & A.E. Aktan

Drexel Intelligent Infrastructure Institute, Drexel University, Philadelphia, Pennsylvania
ABSTRACT of possible approaches for coping with the challenges

presented by uncertainty in assuring practical identi-
Using full scale vibration testing as a means for fication of the dynamic characteristics of large-scale
structural identification (St-Id) of long span bridges constructed systems. This paper examines uncertainty
represents an efficient technique for quantitatively mechanisms in the following areas: (1) experiment
characterizing the in-service mechanical attributes and design and implementation, (2) data pre-processing,
behaviors of these complex systems. The resulting (3) data post processing and modal parameter estima-
characterization can serve as an effective baseline for tion. Statistical analysis of St-Id results from several
structural health monitoring, designing more reliable datasets is also performed, which provides an effective
and cost-effective structural retrofits, or developing means of investigating data reliability and its effect on
and implementing timely and efficient maintenance St-Id. The identification results obtained for the bridge
procedures. However, there are various uncertainties, spans and towers demonstrate that the developed field
known and unknown, involved in the experimental testing and data processing methods may provide a
and identification processes that impact the relia- reliable solution for bridge characterization.
bility of St-Id and serve as a barrier to its more
widespread acceptance and application in engineering
practice. The prevailing excitations (wind and traf-
REFERENCES
fic), environmental conditions (radiation and ambient
temperature), experimental hardware (sensors, cabling ASCE. (2009). Structural Identification of Constructed Facil-
and data acquisition system), and the execution of ities:Approaches, Methods and Technologies for Effective
the experiment (array density and distribution, data Practice of St-Id. A State-of-the-Art Report. ASCE SEI
acquisition parameters, on-site quality control, etc) Committee on Structural Identification of Constructed
have significant influence on data quality. Develop- Systems.
ing analytical models of complex structural system Conte JP, He X, Moaveni B, Masri SF, Caffrey JP, Wahbeh M,
also introduces significant uncertainty in the St-Id pro- Tasbihgoo F, Whang DH, and ElgamalA. (2008). Dynamic
testing of Alfred Zampa Memorial Bridge. Journal of
cess. Recognizing the impact of various uncertainty
Structural Engineering, 134(6), 10061015.
mechanisms and employing appropriate techniques to Grimmelsman KA, and Aktan AE. (2005). Impacts and miti-
quantify, bound and mitigate their impacts will greatly gation of uncertainty for improving the reliability of field
benefit bridge owners and engineers. measurements. Proceedings of the 2nd International Con-
A full-scale vibration test on a long-span suspension ference on Structural Health Monitoring of Intelligent
bridge is taken as an example to illustrate a number Infrastructure (SHMII2), Shenzhen, China.
187
Bridge instrumentation for long term structural health monitoring
M. Sanayei, J.D. Sipple & J.E. Phelps

Department of Civil and Environmental Engineering, Tufts University, Medford, MA, USA
E. Santini-Bell & P.J. Lefebvre

Department of Civil Engineering, University of New Hampshire, Durham NH, USA
B.R. Brenner
Fay, Spofford, & Thorndike, LLC, Burlington, MA, USA; Tufts University, USA
ABSTRACT The structural models created for the Vernon

Avenue Bridge are a meaningful base for the evalua-
Advances in structural analysis, instrumentation, data tion of the collected data. Each phase of data was used
management, and reporting make it feasible to recon- to refine the structural sub-system prior to combining
sider alternate approaches for bridge design. A new them for the overall system behavior. These models
procedure can incorporate development of a base- are more comprehensive than a structural design ana-
line bridge model that can be used for structural health lytical model. Not only must the model capture the
monitoring (SHM) of the bridge over its lifetime. Reg- geometric and section properties, connection char-
ular and effective use of SHM for bridges can provide acteristics and boundary conditions but also all of
more objective data on bridge conditions over time and the loading influences on the bridge. Typical meth-
lead to improved maintenance for more efficient use ods for bridge design lead to well designed structures,
of limited resources. This approach has great promise however fail to accurately capture actual global struc-
at a time of aging infrastructure and limited funds for tural behavior. When measured structural response is
maintenance and repair. required to match a predictive model, typical structural
This paper presents a new procedure for bridge modeling methods fall short. Changing from a design
design and load rating that incorporates the develop- model to a condition assessment model for use with a
ment of a long term SHM system which will lead to structural health monitoring programs would provide
more efficient and effective bridge maintenance and bridge managers an in-depth understanding of struc-
management. A baseline finite element model that rep- tural behavior through the in-service life of the bridge
resents the actual 3D system behavior of the bridge is with relatively little effort.
presented as a part of the designers submittal require- The SHM and condition assessment program being
ment. Such a calibrated model captures the design developed for the Vernon Avenue Bridge will be used
intelligence envisioned by the original designers. by the research team as a benchmark example showing
For this project, a continuous three span composite the process can provide useful information for asset
steel stringer bridge in Barre, Massachusetts was used allocation, and be fairly simple using an established
as a pilot bridge. During construction of the Vernon framework. There are several benefits to the bridge
Avenue Bridge, a monitoring system was installed in owner of having a structural model that reflects the
order to capture the locked in stresses that occur in the actual bridge 3D system behavior that can be used for
girders during construction, prior to being fully loaded. load rating and overload permitting.
Once construction was completed a static load test was This paper presents a collaborative research project
conducted to provide data that could be compared to currently funded by the Partnership for Innovation and
the baseline model for verification. It is intended to the CAREER programs at the National Science Foun-
use the collected test data during and after construc- dation to develop a framework for bridge condition
tion for finite element model updating using PARIS, assessment integrating instrumentation and structural
a parameter estimation program that was developed at modeling for highway bridge decision-making and
Tufts University. management.
188
Modeling and instrumentation of the Tobin Memorial Bridge
M. Sanayei & E.J. Pheifer

Department of Civil and Environmental Engineering, Tufts University, Medford, MA, USA
B.R. Brenner
Fay, Spofford, & Thorndike, LLC, Burlington, MA, USA; Tufts University, USA
E. Santini-Bell & W.L. Durack

Department of Civil Engineering, University of New Hampshire, Durham, NH, USA
ABSTRACT rotational stiffness of the truss connections, boundary

conditions, and built-up member properties.
The Maurice J. Tobin Memorial Bridge carries US To date, the team has developed a set of three dimen-
Route 1 across the Mystic River, connecting the city of sional finite element models representing components
Chelsea and the Charlestown section of Boston, Mas- of the bridge. The combination of a global structural
sachusetts. Construction on the bridge began in 1948 model and several detailed finite element models, cre-
and it was opened to traffic in 1950. The Tobin Memo- ated through a series of special studies, more closely
rial Bridge is a critical link in the network of Boston represents the bridge structures true system behavior.
area highways, connecting metropolitan Boston with The global structural model incorporates the geometry
business and residential areas to the north. The of the truss members, the floor system, the concrete
approximately 2 1/4 mile long structure includes 32 road decks, and end supports. The models from the
approach spans on the Chelsea side, 36 approach spans special studies focus on the structural truss connec-
on the Boston side, the Little Mystic Truss Span, the tions that are typically assumed to function as either
Big Mystic Cantilever Truss Span, and the Toll Plaza. a frictionless pin or fully-fixed joint in conventional
The Massachusetts Port Authority has contracted structural design and analysis.
consultant services for the structural modeling instru- An instrumentation plan that included strain gages,
mentation and monitoring of the Maurice J. Tobin accelerometers, tiltmeters, and temperature sensors
Memorial Bridge. The consultant team, led by Fay, was developed and deployed for the Little Mystic
Spofford, & Thorndike, LLC, and supported by Tufts Truss during the winter of 2009. The instrumentation
University, the University of New Hampshire, and plan serves the purpose of verifying the finite element
Geocomp, Inc., began work on this project in January models and may eventually be used as part of a long
of 2009. term Structural Health Monitoring System. In early
This paper describes the analytical structural mod- fall 2009 a load test was performed to help verify the
els created for this project, including a global struc- models.
tural finite element model of the Little Mystic Truss Work proceeds on refining the modeling work and
and three supporting special studies focused on the evaluating the data.
189
Instrumentation for reinforced concrete durability monitoring

of Qingdao Bay Bridge
X.P. Shao & B.L. Guo

Qingdao Bay Bridge Construction Headquarters, Qingdao, Shandong Province, China
Y.Q. Ni
ABSTRACT category is about the causes to induce and affect cor-

rosion, such as temperature, relative humidity, and
The Qingdao Bay Bridge with a total length of concrete resistivity. The second category is electrode
28,047 m is a sea-crossing bridge currently under con- assemblies characterizing the corrosion extent of steel
struction in Qingdao, Shandong Province, China. As bars, such as limit values for de-passivation, potential
illustrated in Figure 1, it comprises a cable-stayed difference, and polarization resistance.
bridge (Cangkou Bridge) with a main span of 260 m, As listed in Table 1, the three types of corrosion
a single-pylon cable-stayed bridge (Hongdao Bridge) sensors have been embedded in different spans and
with two main spans of 120 m each, a single-tower heights of the bridge (mainly in tidal zone, splash zone,
suspension bridge (Daguhe Bridge) with two main and atmosphere zone). The corrosion status of rein-
spans of 190 m and 260 m respectively, and a series forcement at different service zones and the effect of
of approach bridges. The construction of the Qingdao protective coating will be investigated using the long-
Bay Bridge will be completed in 2011. As part of a term monitoring data from these sensors, which will
sophisticated structural health monitoring and safety benefit the bridge owners to optimize protective mea-
evaluation system being implemented on the bridge, a sures and prioritize maintenance activities targeting
durability monitoring system is specifically designed to prolong service life of the bridge. The data col-
to monitor corrosion-related parameters of the rein- lected from different corrosion sensing systems will
forced concrete structures, in recognition of severe be compared, and the performance of the three sensing
coastal environmental conditions around the bridge systems will be evaluated.
such as about 50-times freezing-thawing cycling per
year and 29.4 to 32.9 saline concentration. Table 1. Deployment of corrosion sensors on Qingdao Bay
The durability monitoring system consists of three Bridge.
types of corrosions sensors: (i) 13 sets of Anode
Ladder Systems (ALS), (ii) 12 sets of Expansion Monitoring position altitude
Ring Systems (ERS), and (iii) 4 sets of Embedded Construction
Corrosion Instrument-1 (ECI-1). Two categories of zones Sensor A B C
corrosion-related parameters are measured. The first
1B ALS 2.0(1) 4.0(1) /
2 ALS 2.5(2) 5.0(3) 7.0(3)
ECI-1 2.5(2) 5.0(3) /
3 ALS 0.3(1) 4.0(1) /
4 ALS 2.0(1) 4.0(1) /
5 ERS 2.0(1) 4.0(1) 7.0(1)
6 ERS 2.0(2) 4.0(3) 7.0(3)
7 ERS 2.0(1) 4.0(1) /
ALS 2.0(1) 4.0(1) /
ERS 2.0(1) 4.0(1) /
ECI-1 2.0(1) 4.0(1) /
8 ALS 2.0(1) 4.0(1) /
10 ERS 2.0(1) 4.0(1) /
(1)
Figure 1. Qingdao Bay Bridge under construction. pier column; (2) cushion cap; (3) bridge tower.
190
Structural monitoring of Lezria Bridge since its construction
H. Sousa & J. Figueiras

LABEST, Faculty of Engineering, University of Porto, Porto, Portugal
J. Bento
BRISA Auto-Estradas de Portugal S.A., Lisboa, Portugal
ABSTRACT
Currently, automatic monitoring systems are increas-

ingly used in the civil engineering structures especially
in bridges with a high socioeconomic impact. In the
last decades, the technology evolution that the world
has being subjected allow, nowadays, the implemen-
tation of monitoring systems capable of observe the
structural behavior in a programmable, remote and
real-time manner. The databases originated by the sen-
sors records are valuable information to understand
better the real behavior of the monitored structures
as well to improve the numerical models used in
the analysis and prediction. This work presents the
Lezria Bridge case, equipped with an integrated and
automatic monitoring system. This system allows the
observation of the structural behavior and the mate-
rials durability. A presentation and discussion of a
set of results obtained since the bridge construction
is performed. The principal considerations refer to:
(i) environmental effects temperature and humidity;
(ii) rheological effects of concrete creep and shrink-
age; (iii) events related with the bridge construction,
long-term behavior and short-term behavior correlated
to the bridge traffic. A final discussion is presented
which is focused the usefulness of the information
obtained by the monitoring system for the assessment
and surveillance of the structure.
In this work it is intended to present the long-term
monitoring system of Lezria Bridge by a discussion
of a set of results obtained since the construction.
The gathered results provide valuable information for
further study and for the understanding of the real
behavior of the structure. Through an adequate data
treatment and processing to a higher level of informa-
tion abstraction, it is intended to demonstrate the utility
of this type of systems to help on the assessment and Figure 1. Lezria Bridge north viaduct (top), main bridge
surveillance of the structure. (middle) and south viaduct (bottom).
191
Reliability assessment oriented monitoring system design

for Shanghai Yangtze River Bridge
Z. Sun & H.-S. Wu

State Key Laboratory on Disaster Reduction for Civil Engineering, Shanghai, China
Q.-M. Wang & Y.-C. Lu

Shanghai Municipal Engineering Design General Institute, Shanghai, China
Z.-F. Zhou
Shanghai Juyee Technology Development Ltd. Company, Shanghai, China
ABSTRACT b) Force of critical cables in the cable net system;

c) Acceleration of key sections for structural natural
The Shanghai Yangtze River Bridge is part of the frequency and mode shape estimation.
Chongming Cross River Passage Project connecting
Moreover, the environment and load should be moni-
the Pudong district and the Chongming Island in
tored to calibrate structure load model and the defor-
Shanghai. The main navigation channel bridge is a
mation and rotation should be monitored to calibrate
double pylon double cable plane cable-stayed bridge
structure geometrical alignment. Figure 1 shows
with slotted box girder. The bridge is 1430 meters long
with the following span alignment 92 m + 258 m +
730 m + 258 m + 92 m. Concerning its importance, REFERENCES
the authority made a decision to install a bridge health
monitoring system on the structure. Besides that, an FRANGOPOL, D.M., STRAUSS, A., KIM, S. 2008. Bridge
inspection and maintenance plan is carefully made. It reliability assessment based on monitoring, Journal of
will be put into use when the bridge is open to traffic. Bridge Engineering, 13(3), 258270.
To fulfill the requirement for structure reliability BRUNEAU, M. 1992. Evaluation of system-reliability meth-
assessment, the following items should be monitored: ods for cable-stayed bridge design, Journal of Structural
Engineering, 118(4), 11061120.
a) Strain at the stress concentrated points for all Chinese General Code for Highway Bridge Design. JTG D60-
critical sections of the girder and cable pylons; 2004.
Figure 1. Sensor placement for Shanghai Yangtze River Main-navigation Channel Bridge.
192
Bridge Sensor Mart: A flexible and scalable data storage and analysis
framework for structural health monitoring
N.C. Tas, C. Raileanu, M. Dejori & C. Neubauer

Siemens Corporate Research, Princeton, NJ
ABSTRACT hardware or certain vendors. The BSM architec-

ture is sensor- and provider-agnostic. The data
This paper reports on the new sensor data collec- import layer provides a generic interface that can be
tion, analysis and storage framework, Bridge Sensor extended through a plug-in mechanism to support
Mart (BSM), for bridge health monitoring data. BSM new sensor hardware.
defines a distributed data storage and data analytics 4. Flexible Data Analytics Support: BSM provides a
infrastructure in order to collect, store, analyze and generic interface for users to plug-in their own data
manage sensor data for Structural Health Monitoring analysis algorithms, and to extend the analytical
applications (Ansari 2005). capabilities without requiring any changes in the
SHM applications are gathering large amounts of system.
data that are very hard to be managed with the current
BSM is based on a very efficient hybrid database
systems; thus making the dissemination and sharing of
structure which combines and utilizes two kinds
such data very cumbersome or even impractical. BSM
of databases: a scientific database for storing large
architecture targets one of the biggest challenges in
amounts of historical sensor data and a regular row-
SHM applications: How to efficiently collect, store
oriented database for storing metadata. Simulations
and provide the accumulated data to the end users
performed showed that BSM database structure out-
and applications for further analysis? In this work, we
performs traditional databases considerably for sensor
present a new data storage and analysis framework,
related data.
Bridge Sensor Mart (BSM), designed specifically for
BSM allows to robustly manage large amount of
long term SHM.
data, by implementing a fully distributed architecture
The key features of this framework are:
for data gathering, analysis and dissemination. Thus,
1. Distributed Data Analytics: The post-processing of a single point of failure is avoided and the workload
incoming data is usually performed on a single is distributed among several computers. This architec-
computer. The BSM framework allows distribut- ture promises high scalability as it is easy to extend the
ing data analysis and data processing procedures system by simply deploying new hardware (e.g. new
across multiple machines allowing users to handle server nodes) in order to lessen the pressure on highly
large amounts of data. loaded computers.
2. Distributed Data Storage: BSM also supports
the distributed storage of accumulated data. This
distributed architecture allows the system to be REFERENCES
extended to handle larger scales of data by simply
adding additional hardware components without Ansari F. Sensing issues in civil structural health monitoring.
Springer, 2005.
changing the design or codebase.
3. Flexible Sensor Support: Current data management
systems are designed to support specific sensing
193
Full-scale measurements on buffeting response of Sutong Bridge under

typhoon Fung-Wong
H. Wang, A.Q. Li, T.Y. Zhu & R.M. Hu

College of Civil Engineering, Southeast University, Nanjing, China;
Key Laboratory of Concrete and Prestressed Concrete Structure, Ministry of Education,
Nanjing, China
ABSTRACT established to predict and assess the health condition of

SB during its construction and operation stage, espe-
The field measurement of bridge buffeting response cially under the affects of various disasters. Typhoon
is essential to the research of bridge wind engineer- Fung-Wong attacked the SB in July 2008. During the
ing. Sutong Cable-stayed Bridge (SB) is the longest typhoon period, field measurement was conducted to
cable-stayed bridge in the world. For such a long- record wind velocities; SHMS of the SB was used to
span cable-supported bridge, the structural stiffness collect the structural vibration responses.
drops significantly with the increase of the bridge In this paper, the measured data are studied using
span, which makes wind-induced vibration particu- spectral analysis and statistical method. The study
larly important for structural safety. includes the analyses of wind characteristics on bridge
Taking into account the importance of SB, site and the response characteristics of stay cable and
Structural Health Monitoring System (SHMS) was the deck, the relationship of vibrations and wind speed,
the comparison of upstream and downstream cable
vibrations, the effectiveness of cable dampers and so
forth. The spectra of the upstream cables are plotted in
Figure 1.
The results validate the stability and reliability
of the SHMS; indicate the acceleration RMS val-
ues become larger as the wind speed increases. The
calculated dynamic characteristics agree with the mea-
sured ones well, validating the reliability of the FE
model. Moreover, the comparison between the mea-
sured results and data from other studies reveals the
effectiveness of dampers installed around anchor ends
of the cables. The results can be employed to vali-
date the current buffeting response calculation method,
and provide references for wind-resistant design of the
super-long-span cable-stayed bridges.
REFERENCES
Xu,Y.L. and Zhu, L.D, 2005. Buffeting response of long-span
cable-supported bridges under skew winds: case study.
Journal of Sound and Vibration, 281, 675687.
Brownjohn, J.M.W., Bocciolone M., Curami A., et al. 1994.
Humber bridge full-scale measurement campaigns 1990
1991. Journal of Wind Engineering and Industrial Aero-
dynamics, 52, 185218.
Miyata T., Yamada H., Katsuchi H., and Kitagawa M., 2002.
Full-scale measruments of Akashi-Kaikyo Bridge dur-
ing typhoon. Journal of Wind Engineering and Industrial
Aerodynamics, 90, 15171527.
Figure 1. Acceleration spectra of upstream cables.
194
Structural health monitoring for damage detection based on integration

of computer imaging and sensor data
R. Zaurin & F.N. Catbas

University of Central Florida, Civil & Environmental Engineering Engr II, Orlando, Florida, USA
ABSTRACT
Novel structural health monitoring strategies for

better management of civil infrastructure systems
(CIS) are increasingly becoming more important as
CIS structural performance decreases due to reasons
such as damage, over loading, severe environmen-
tal conditions, and aging due to normal continued
use. Structural Health Monitoring (SHM) paradigm
is designed to provide objective information for
decision-making on safety and serviceability and it is
expected to complement the current visual and heuris-
tic based assessments. SHM utilizes advanced tech- Figure 1. UCF-4-Span Bridge.
nologies to capture the critical inputs and responses
of a structural system in order to understand the root
causes of problems as well as to track responses to pre-
dict future behavior. Very recently, some investigators
have explored the possibility of incorporating imaging
and optical devices and combining them with sensing
technology. In this paper, the use of computer vision,
sensing, damage indices and statistical analysis meth-
ods are presented and demonstrated in the UCF 4-span
bridge model for damage identification in the context
of structural health monitoring. The framework of this
study is given in Figure 1 and the test setup in Figure 2.
Two different remote controlled vehicles under var-
ious loading scenarios crawl over the bridge while a
video camera supervises the structure providing traffic
video stream. At the same time, a distributed array of
sensors collects data.
Correlation between moving vehicular load and
structural responses is determined as Unit Influence Figure 2. UCF-4-Span Bridge.
Lines (UIL) are extracted and used as an index
for monitoring the bridge behavior. Different dam-
age scenarios were considered and studied for the processing the different damage cases, influence lines
UCF 4-span bridge. These cases were chosen because were obtained. A Mahalanobis-distance-based outlier
they represent some of the most common issues detection algorithm is used to identify changes in the
affecting bridges, according to the Department of structure, and results are analyzed.
Transportation (DOT) engineers based on our private Finally, a new method to more effectively iden-
discussions. Rusted supports, stiffness reduction and tify, localize and quantify (in a relative sense) induced
loss of connectivity between composite sections are damage by using a new index called normalized dis-
experimentally simulated, processed, and analyzed by tance Nd is derived and demonstrated. Results are
means of statistical methods. After application and presented and discussed.
195
MS5: New procedures for bridge rehabilitation
Organizer: V. Popa
Sustainable and cost effective solutions to life extension of bridges
C.P. Atkins, P. Lambert, R. Brueckner, R. Merola & A.R. Foster

Mott MacDonald, Altrincham, United Kingdom
ABSTRACT the Silver Jubilee Bridge was chosen as to be the

most suitable system to withstand bridge vibrations,
In order to minimise costs and disruption associated reducing installation time and avoiding long lasting
with repair it is necessary to minimise the amount traffic management. The system designed for the M4
of concrete removed wherever practical. This also in London was governed by access and noise restric-
benefits the environment, with less material going tions. Every structure needs to be assessed for their
to landfill, and less fresh material being employed. local restriction and the most appropriate system needs
Cathodic protection (CP) can be used as a repair tech- to be identified by the engineer. Conventional sys-
nique for chloride contaminated concrete. The main tems such mesh and overlay and discrete anodes can
benefit is that it is no longer necessary to remove all be used most of the time, however, to go one step
the chloride contamination. This minimises the extent forward and reduce the environmental footprint a lit-
of repairs to the replacement of spalled material and tle thought and research is necessary. Innovative CP
can avoid the need for costly temporary support during systems are able to reduce installation time and the
the repair process. amount of materials used as well as a significant
Chlorides commonly derive from de-icing salts and/ life extension of bridges. For specific site and CP
or were formerly used as concrete admixture. Chloride system requirements traffic management, working at
contaminated reinforced concrete structures mainly height and exposure to vibration and noise can be
suffer from cover concrete delamination and loss of limited.
reinforcement cross-sections. By using CP there is no Over the years that CP systems have been designed,
need to remove all of the chloride-contaminated con- installed and subsequently monitored, it was observed
crete reducing the time for access, repair and traffic that they have generally exceeded the performance
management. criteria with ease, Lambert (2009). After the first
CP involves polarizing the reinforcement in an 12 months of operation, cathodic protection systems
electrical circuit. The electrical circuit converts the stabilize and require little intervention provided the
reinforcement into the cathode and an inert electrode monitoring probes continue to give consistent and
(e.g. mixed metal oxide coated titanium) forms the stable readings. The experience gained from each
artificial anode which can be installed discretely or on system has been used to refine the design of the fol-
the surface. lowing applications, resulting in bigger zones, less
This paper presents an outline of a number of UK anodes and reference electrodes, and the develop-
based projects where CP has formed a key part of a ment of hybrid protection systems to overcome local
maintenance strategy, highlighting the decision pro- difficulties.
cesses involved and lessons learned for the future.
These are:
M4 Elevated Freeway, London. REFERENCES
Silver Jubilee Bridge, Runcorn.
Theale Railway Bridge, London. Lambert, P., Foster, A., Atkins, C. & Brueckner, R., 2009.
Adaptive Design of Cathodic Protection for Reinforced
Repair strategies are mainly governed by access Concrete, Proceedings of 17th IBAUSIL 2009, University
restrictions at the structures. The system installed at of Weimar.
199
Electrochemical chloride extraction and electrochemical re-alkalization

Foreva Regebeton PA process
R.A. Bedn & D. Deschamps

Freyssinet, Inc., Sterling, VA, United States of America
ABSTRACT
Corrosion is a main issue of reinforced concrete

structures with regards to durability. Due to the impor-
tant porosity in concrete, corrosion protection is not
guaranteed and can fail if atmospheric carbon diox-
ide (CO2), or chloride contamination penetrate the
concrete.
Foreva Regebeton PA is a process dealing with
both carbonation and chloride ions attack. With the
same electro-chemical treatment applied to the cor-
roded reinforced concrete, chloride ions are partially Figure 1. Re-alkalization process.
extracted from concrete and concrete is re-alkalized
(means pH getting values back to 12 or 13) around the
reinforcing steel in the concrete.
Foreva Regebeton PA gives back a high alkalinity
environment close to reinforcing steel. The electrolyte
solution contains hydroxyl ions (OH ) that are trapped
by reinforcing steel during the treatment. The increase
of pH is less farther from reinforcing steel since
hydroxyl ions will go close to reinforcing steel. The
sketch below shows the principle of the re-alkalization
treatment:
Foreva Regebeton PA can be applied if chlorides are
coming from the outside and are mainly present in the
cover depth. Chlorides farther than reinforcing steel Figure 2. Chloride extraction process.
from the exposed face of the structure cannot be satis-
factory extracted. This treatment is therefore applied in
the case of sea salt or deicing salt contamination. Dur- three to four weeks. An impressed current (low amount
ing the treatment, the chloride ions move toward the of current) is applied to shorten the treatment dura-
external anode embedded in an electrolytic solution tion. Lifetime of the structure is increased. In the
and they are then trapped in the electrolytic solu- case a protective coating is applied afterwards, life-
tion. The sketch below shows the principle of chloride time is extended farther. Aesthetic is unchanged: the
extraction treatment: product is fully removed at the end of the treatment.
Foreva Regebeton PA can be applied even if the Both effects (re-alkalization and chloride extraction)
structure is corroded. The treatment duration is from are performed with one application.
200
Bridge crossings raised to provide more overhead clearance
R.A. Daniel
Ministry of transport Public Works and Water Management, Civil Engineering Division, Utrecht, The Netherlands
B. van Sinten
Heijmans Infra B.V., Roosmalen, The Netherlands
H.T. van Manen

Ministry of transport Public Works and Water Management, Civil Engineering Division, Utrecht, The Netherlands
ABSTRACT
A frequent dilemma regarding old bridges is a choice

between adaptation to new, usually higher require-
ments; and replacement by new structures. This choice
is not only a matter of structural strength and func-
tional fitness; it also has environmental, esthetical,
heritage-related and other impacts. The Netherlands
is a country of an intensive inland navigation that is
considered economically and ecologically preferable
to other modes of transport. Navigation is, therefore,
another important issue for this choice.
In order to allow high loaded barges pass the Juliana
Canal, the main waterway to Belgium and France, two
old bridge crossings Roosteren and Echt were
raised by over 3.0 m. This solution proved to be prefer-
able to new construction in all respects mentioned
above. Also its costs proved to be lower than a new
construction. A raising project of that size required,
however, a thorough investigation of both the condi-
tion and behaviour of the old bridges. Moreover, since
the bridges had a continuous 3-span system, middle
arch-span 80.0 m long, the raising had to be well syn-
chronized in order not to generate excessive loads. An
important issue was also the remaining service life. If
it was not long enough, new construction might still
have been a better choice. To assess such risks, load
tests were performed (Fig. 1 top).
The raising was performed using the technology
of hydraulic jacking-up. 12 jack-up units, 2400 kN
working load each, simultaneously lifted the bridge Figure 1. Bridge load testing (top) and jacking-up the
superstructures to the desired level (Fig. 1 bottom). superstructure (bottom).
The full paper presents the details of this operation, fol-
lowed by the substructure leveling and finishing works.
The method proved not only to be very effective and
reliable, but also environment-friendly. It consumed
only a fraction of the materials and energy which would
have been required by other solutions. Both bridges
perform well in raised position.
201
Choosing the appropriate sustainable polymer concrete material for bridge

preservation and maintenance
A.M. Dinitz & M.S. Stenko

Transpo Industries, New York
ABSTRACT Bearing Pads

Wearing Surfaces Overlays
Today engineers and owners are faced with a variety of Skid Resistance Surfaces
available specialty materials for a wide range of appli- Crack And Surface Sealing
cations on every bridge rehabilitation project. Sustain-
Each of these applications has their own unique mate-
able polymer concrete products and material systems
rial requirements that must be considered before a
developed by industry with input from engineers have
decision can be made in selecting the proper mate-
proven their ability to extend the service life of exist-
rial to be specified. It is important that an engineer
ing structures. Applications of polymer products and
not only consider the material physical properties
material systems on newly constructed structures can
requirement but they also need to consider the jobsite
reduce maintenance and increase traveler safety.
application conditions. Some of the key requirements
Both, polymer and concrete technology are an
to be considered when selecting the proper polymer
important part in choosing the appropriate polymer
material are:
product or polymer concrete material system to meet
a project requirement. Material physical properties, Compressive Strength
installation procedure, construction time, material Flexural Modulus
cost, weather limitations, performance requirements, Elongation
and field requirements as well as safety and environ- Viscosity
mental issues are all factors that must be considered Temperature Limitations
in order to select the proper material. Cure Time
Polymer concrete specified and used today for the Required Mixing and Installation Equipment
sustainability of highway structures can be broken
This paper will give engineers and owners the
down into several different resin types:
knowledge they need to make an informed decision
Epoxy on which of these products and material systems
Modified Epoxy to choose, and what variables should be taken into
Methyl Methacrylates consideration for an array of different repairs and
High Molecular Weight Methacrylates safety concerns including but not limited to Spall
Polyesters Repairs, Joint Headers, Bearing Pads, Wearing Surface
Overlays, and Crack and Surface Sealing.
These basic resin types can be formulated to develop
Several successful preservation, rehabilitation and
different physical properties which make them ver-
new construction installations will be discussed
satile when they are to be used for a variety of
including spall repair, thin bridge overlay and concrete
applications. Some of the typical areas where they are
sealing projects.
used with success are:
Spall Repair
Joint Headers
202
Shear resistances and strengthening of aged prestressed concrete bridges

considering deterioration and fatigue effects
P. Mark, M. Bender, M. Strack & Ch. Fust

Institute of concrete structures, Ruhr-University Bochum, Germany
V. Birtel
Materialprfungsanstalt Universitt Stuttgart, Germany
ABSTRACT
The major part of European road bridges was built

of prestressed concrete, about forty to twenty years
ago. Compared to nowadays standards, they lack of
resistances due to increased live loads, temperature
effects and detailing provisions. Moreover, deteriora-
tion effects weaken the structures. Often, a pronounced
diagonal shear cracking arises at the supports giving
rise to subsequent strengthenings and recalculations of
the present shear resistances in the focus of degradat-
ing influences as well as realistic live load scenarios
during the previous and remaining live time.
In the paper an elaborated shear design approach
is presented to estimate remaining shear resistances
under fatigue actions. It accounts for the at the time
of bridge erection typically used strong prestress-
ing actions in combination with low amounts of shear
reinforcements. The governing resistance of the tensile
strut is summed up from the partitions of the concrete
section, the stirrups and arch actions introduced by
the compressive axial forces (Fig. 1). Fatigue degra-
dations are modelled within the first two partitions by
time variant weighted load-cycle factors with respect
to concrete and steel strengths. The fatigue shear load-
ing is seperately assigned to the contributions, while Figure 1. Shear mechanisms of members with or without
the stirrups are activated at last. The governing equa- shear reinforcement.
tions are derived, verified to experimental data and
compared to approximations in design codes.
In the second part of the paper a shear strength-
ening method for multi-span bridges is presented that REFERENCES
uses external tendons. The tendons are vertically devi-
ated by specific steel devices near the coupling joints Birtel, V. Mark, P. & Stangenberg F. 2009. Tragverhalten
(up) and at the supports (down) achieving a stepwise von Stahlbetonbalken bei geneigter Querkraft. Beton- und
Stahlbetonbau 104(1): 2736.
linear course and pronounced relieving shear forces. Okamura, H. & Ueda, T. 1982. Fatigue Behavior of Rein-
The steel devices ensure direct couplings of horizontal forced Concrete Beams under Shear Force. In Proceedings
deviation forces and suitable vertical load applica- of the IABSE Colloqium Fatigue of Steel and Concrete
tions from underneath the bottom slab and by anchor Structures, Lausanne: 415422.
bolts near the deck slab and the transversal girders to Ponzel, U., Grebe, B. & Eisler, R. 2005. Verstrken von
account for the limited amount of vertical reinforce- Spannbetonbrcken mit externen Spanngliedern. Beton-
ment. The principles of the design calculations as well und Stahlbetonbau 100(S1): 5761.
as details of deviation devices, tendon geometry and
anchorage constructions are presented.
203
Fatigue strength and repair methods of corroded bridge wires
S. Nakamura
Tokai University, Hiratsuka, Japan
K. Suzumura
Nippon Steel Engineering, Futtsu, Japan
ABSTRACT because stresses concentrate on a notch at the sharp

edge of a triangle and initiates a fatigue crack. The
Bridge cables in suspension and cable-stayed bridges triangle shape pit also decreases the fatigue strength
are under severe corrosion environments due to but it depends on the ratio of pit length to width. This
water and high temperature inside the cables. Fatigue ratio also relates with stress concentration factor. The
strength of corroded galvanized steel wires was inves- round shape pit is safer because stress concentration
tigated. is less likely to occur.
Corroded galvanized steel wires on three corrosion The S-N data of the triangle pit specimens (group-
levels were produced at laboratories. Then, fatigue Q) and the notched triangle pit specimens (group-R)
strength of corroded wires was obtained by cycli- are on the extension S-N line of the level-3 corrosion
cally loading under dry and wet environments. Fatigue wires. This verifies that the pit shape is a dominant
strength did not change when only the galvanized factor to lower the fatigue strength.
layer was corroded, but it significantly decreased after Cables are the crucial element in cable-supported
the steel corrosion progressed. The fatigue strength bridges. Once cables are corroded, it is not easy to
under wet conditions was lower than that under dry repair or replace them. However, not many studies
conditions due to corrosion fatigue. have been conducted so far to confirm effectiveness
Pit sizes were measured in the wire specimens of various repair methods and no perfect one has been
on corrosion levels-2 and 3. The maximum depth is established.
5.6 mm, and both pit length and width are within Seven cases were studied in evaluating effective-
10 mm. Fatigue tests were conducted for the specimens ness of different repair methods of corroded wires: (1)
with artificial pits of 0.6 mm deep and 3.5 mm long. taking no measure, (2) coating only the surface wires
They have three different pit shapes; round (group-P), with epoxy resin paint, (3) coating only the surface
triangle (group-Q) and triangle with notch (group-R). wires with zinc rich paint, (4) coating and filling the
Each group has three lengths; 3.5, 6.0 and 10 mm. surface and inside wires with oil containing inhibitor,
The group-P specimens with round shape pits did (5) coating and filling the surface and inside wires
not break until the limit cycle of one million. entire wires with epoxy resin, (6) coating the surface
The group-Q specimens with triangle shapes broke layer with thick paste containing zinc powders, and (7)
at the fewer cycles for the shorter pit length. At a pit dehumidifying the inside of a cable.
length of 3.5 mm the wires broke at cycles between In most of the cases mass loss of the inside wires
4,800 and 36,000. At a pit length of 6.0 mm the wires due to corrosion was much less than that of the surface
broke at cycles between 49,000 and 72,000. At a pit wires. As for the surface wires the dehumidification
length of 10.0 mm the wires did not break. The critical method (case 1) was the most effective followed by the
cycles increase with pit length. epoxy resin paint of surface and inside wires (case 5),
The critical cycles did not depend on pit length for the zinc powder paste (case 6), and the zinc and epoxy
the group-R specimens with triangle shapes with a resin paint of surface wires (case 2 and case 3). The oil
notch. All of the wires broke at cycles between 18,000 filling (case 4) was not very effective compared with
and 36,000. other repair methods. The tests show that the dehumid-
From these results of the groups-P, Q and R it can ification method and the epoxy resin paint and filling
be understood that the pits with a triangle shape and of surface and inside wires seem to be very effective
with a notch has the lowest fatigue strength. This is to repair corroded wires.
204
MS6: Measurement systems for bridge weigh-in-motion (B-WIM)
Organizers: B. Bakht, A. Znidaric & D.K. McNeill
Bridge capacity assessment by combined proof-loading and WIM data
J.R. Casas & J.D. Gmez

Civil Engineering Department. Universitat Politcnica de Catalunya. Barcelona, Spain
ABSTRACT of such tests, actual traffic data is mandatory. The paper

shows how the traffic data from 5 European countries
The increasing volume of European transport urgently (The Netherlands, Czcech Republic, Poland, Slovakia
requires an effective road and rail system in Central and Slovenia), obtained via WIM, has been used to
and Eastern countries. To bring this transport infras- propose a set of proof load factors applicable to the
tructure up to modern European standards will require existing bridge in those countries (Gmez & Casas
an immense investment (estimated by the European 2010, ARCHES-D16 2009). Evidently, in order to be
Commission to be about 100 billion), and there- general, they are based on the most heavy traffic condi-
fore difficult to achieve in the medium term. New tions that can be encountered. Therefore, the proposed
motorways will be required with many new bridges. target proof load factors may become too conserva-
Numerous existing bridges will need to be assessed, tive for many bridges located in local or secondary
and a large portion of them improved or replaced. roads that will never experience such level of load-
The overall goal of the European project ARCHES ing. For this reason, additionally, the paper presents
(Assessment and Rehabilitation of Central European a simplified method that with very common traffic
Highway Structures) was to reduce the gap in the data easily recordable by WIM systems may predict
standard of highway infrastructure between Central accurately the most critical traffic actions for any spe-
and Eastern European Countries (CEEC) particu- cific bridge under assessment. The simplified model
larly New Member States (NMS) and the rest of has been checked by comparing its results with those
the EU. This key problem was addressed by a com- derived with a complete simulation process for a long
bined approach. As part of it, takes great importance time period.
the development of more appropriate bridge capacity
assessment tools and procedures to avoid unnecessary
interventions (repairs/replacements) in the existing REFERENCES
road network (Casas 2010, ARCHES-D16 2009). In
the paper, proof loading tests are presented as one ARCHES-D08 2009. Recommendations on the use of results
of monitoring on bridge safety assessment and main-
of these tools. The correct application of the method
tenance. Deliverable D08. ARCHES Project. VI EU
should combine an accurate execution and monitoring Framework Program. Brussels (available on line at
of the loading process and a good estimate of the actual http://arches.fehrl.org)
traffic in the bridge. The last is of crucial importance to ARCHES-D16 2009. Recommendations on the use of soft,
decide on the target value of the load to be introduced diagnostic and proof load testing. Deliverable D16.
by the test. This can be achieved by the most advanced ARCHES Project. VI EU Framework Program. Brussels
WIM techniques available, also developed and (available on line at http://arches.fehrl.org)
applied within the ARCHES project (ARCHES-D08 Casas, J.R. 2010. Assessment and monitoring of existing
2009). bridges to avoid unnecessary strengthening or replace-
ment. Proceedings of 5th International Conference on
The target proof load to be introduced in the bridge
Bridge Maintenance, Safety and Management, IABMAS
during the test can be evaluated on the basis of reliabil- 10. Philadelphia, USA. Rotterdam: Balkema
ity theory as many variables involved are of random Gomez, J.D. & Casas, J.R. 2010. Target proof load factor for
nature. The target proof load can be obtained as the highway bridge assessment in Central and Eastern Euro-
nominal load in the Code multiplied by a so called pean Countries (CEEC). Proceedings of 5th International
proof load factor. In order to get values of the proof Conference on Bridge Maintenance, Safety and Man-
load factor representative for several countries from agement, IABMAS 10. Philadelphia , USA. Rotterdam:
Central and Eastern Europe to facilitate the execution Balkema
207
Using weigh-in-motion data for modeling maximum live load effects

on highway bridges
M. Ghosn
The City College of New York/CUNY, New York, New York, U.S.A.
B. Sivakumar
HNTB, New York, New York, U.S.A.
ABSTRACT that best fits the tail end of the distribution. The mean
of the Normal that best fits the tail end of the distri-
This paper recommends a procedure to estimate the bution is obtained from event = n/m. The standard
maximum expected load effect on a highway bridge. deviation of the best fit Normal is event = 1/m. Given
The procedure describes how site-specific truck weight a total number of events in the return period of inter-
and traffic data collected using Weigh-In-Motion est, N, the most probable value, u, and the dispersion
(WIM) systems can be used to obtain estimates of the coefficient for the Gumbel distribution that models the
maximum live load for the design life of a bridge, spec- maximum value Lmax are:
ified to be 75 years as per the AASHTO LRFD code,
or the two-year return period that may be used for the
load capacity evaluation of existing bridges.
The model requires as input the WIM data col-
lected at a site after being scrubbed and processed
as described by Sivakumar et al (2008). The process
begins by assembling the measured load effects his-
tograms for single lane events and two-lane events. The
cumulative distribution function for each load effect is The mean value of the maximum load effect, Lmax ,
used to obtain the standard deviate of the cumulative and the standard deviation are obtained from:
function. A plot is made of the upper 5% of the val-
ues of the normal deviate versus the load effect X as
illustrated in Fig. 1.
The slope m and intercept n of the best fit regression
line provide the statistics for the normal distribution in which is the Euler number = 0.577216. The
standard deviation is:
The procedure can be used to obtain the maximum

load effect for any return period to design and evaluate
the safety of bridges in different jurisdictions where the
AASHTO LRFD loading may not be representative of
current loading due to variations in truck traffic and
truck weight conditions.
REFERENCES
Sivakumar, B., Ghosn, M. & Moses, F. (2008) Protocols for
Collecting and Using Traffic Data in Bridge Design, Web
Figure 1. Linear fit of tail end of load effect histogram on Report NCHRP Project 1276, Transportation Research
Normal probability paper. Board, National Academy Press, Washington DC.
208
Bridge weigh-in-motion on steel orthotropic decks Millau viaduct

and Autreville bridge
B. Jacob, M. Hannachi & S.S. Ieng

Laboratoire Central des Ponts et Chausses, Paris, France
ABSTRACT Table 1. Statistics of the relative errors of the truck sample

and accuracy classes (Millau, 2009).
Bridge weigh-in-motion (B-WIM) is a technique
which uses an instrumented bridge to weigh heavy Standard
vehicles in motion. Several types of bridges may be Criteria Number Mean Dev. Class
used for that, but they shall be sensitive to traffic
Gross weight 43 3.24% 5.76% C(15)
loads, above all wheel or axle loads. The longitudi- Single axle 86 1.09% 11.18% D + (20)
nal stiffeners of steel orthotropic deck bridges bend Group of axles 39 8.01% 5.32% C(15)
between two cross beams when wheels and axles are Axle of group 115 7.93% 9.46% C(15)
passing on them. The bending strains are measured by
extensometers. The principle of free of axle detector
(FAD) B-WIM, introduced by (nidaric et al. 1999) is
used, with two transverse sections of the bridges instru- The relative errors of the axle loads and gross
mented to get the vehicle speed and axle spacings. A vehicle weights and their statistics were calculated,
global optimization software introduced by (Dempsey and then the European specifications on WIM (Jacob
et al. 1999) calculates axle loads and gross vehicle et al. 2002) were applied to assess the B-WIM system
weights. First experimentations were carried out in the accuracy (Table 1).
European cooperative WAVE project in the 90s, and in The system meets the accuracy class C(15)/D +(20)
2009 a large scale test was done on the Millau viaduct of the European specifications. These results were
in France, the tallest cable stayed bridge in the world, compared to those of the test carried out in 199798
on the motorway A75, using a commercial B-WIM on the Autreville bridge in eastern France, another
system SiWIM. orthotropic steel deck bridge (Dempsey et al. 1999),
The deck is a steel box orthotropic deck, 32 m in and were found very consistent with them. However,
width and 4.20 m in height, which carries 2 lanes and the past results analyzed with a 2-D algorithm instead
an emergency lane in each direction. The instrumented of the 1-D standard algorithm, which is implemented
section is close to the North end of the viaduct in the in the SiWIM, gave a better accuracy. Therefore it
first span. 12 extensometers were stuck on the bot- was concluded that there are perspectives to improve
tom of longitudinal stiffeners at mi-span between two the SiWIM accuracy on orthotropic deck bridges by
cross beams under the slow lane, and partly under the developing a new algorithm.
adjacent emergency and fast lanes, for the weighing
section. Four additional extensometers were stuck in REFERENCES
the previous section 4 m upstream under the slow lane.
The SiWIM was calibrated over 1.5 day with two Dempsey, A., Jacob, B. & Carracilli, J. 1999. Orthotropic
rented trucks, a rigid 3-axle and an articulated 5-axle Bridge WIM for determining Axle and Gross Vehicle
truck, which made repeated runs over the viaduct at Weights, in Weigh-in-motion of Road Vehicles, ed. B.
two speeds (6080 km/h) and three transverse loca- Jacob, Hermes Science Publications, Paris, 227238.
tions. Then 52 trucks from the traffic flow were Jacob, B., OBrien, E.J. & Jehaes, S. (eds.) 2002. Euro-
pean Specifications of WIM, in Weigh-in-Motion of Road
measured by the SiWIM, identified by an observer
Vehicles: Final Report of the COST 323 Action, LCPC
and stopped at the toll gate 5 km downstream, to be Publications, Paris, 293378.
weighed in static on an approved scale (axle load and nidaric,A., DempseyA., Lavric, I.,& Baumgrtner, W. 1999.
gross vehicle weight). After a careful check of the col- B-WIM Systems without Axle Detectors, in Weigh-in-
lected data, 43 of these trucks were kept as correctly motion of Road Vehicles, ed. B. Jacob, Hermes Science
identified on both systems (B-WIM and static). Publications, Paris, 101110.
209
Winnipeg weigh in motion
A. Mufti, H. Vogel & E. Murison

ISIS Canada Research Network, University of Manitoba, Winnipeg, Manitoba, Canada
B. Bakht
JMBT Structures Research, Toronto, Canada
ABSTRACT the performance of the bridges and establish fatigue

life of the girders and the decks. The method chooses
Fatigue analysis is an important consideration for several corresponding strain peaks to calculate vehi-
evaluating the health of the bridges. To develop a cle parameters, such as truck speed, axle spacing,
fatigue analysis, load and resistance are two essen- axle weights, and gross weight. The calculated truck
tial concerns. In fact, load due to traffic vehicle differs parameters are used as input for a program of Semi-
considerably from the designed load. To estimate the continuum method of analysis for bridges (SECAN)
remaining fatigue life more accurately, a refined eval- to check the calculated result and to study the perfor-
uation load procedure is desirable. In this paper, we mance of bridges. This allows utilization of the output
propose a Winnipeg method to incorporate bridge of SECAN for input in the fatigue analysis to estimate
Weigh-in-Motion (WIM) technology into Structure the fatigue life.
Health Monitoring (SHM) data analysis to understand
210
The influence of correlation on the extreme traffic loading of bridges
E.J. OBrien
B. Enright
Dublin Institute of Technology, Dublin, Ireland
T. Dempsey
Roughan ODonovan Innovative Solutions, Dublin, Ireland
ABSTRACT
Accurate traffic loading models based on measured

WIM data are essential for the accurate assessment
of existing bridges. There are well-established meth-
ods for the Monte Carlo simulation of single lanes of
heavy vehicle traffic, and this can easily be extended
to model the loading on bridges with two independent
streams of traffic in opposing directions. However, a
typical highway bridge will have multiple lanes in the
same direction, and various types of correlation are
evident in measured traffic, such as groups of very
heavy vehicles traveling together and heavy vehicles
being overtaken by lighter ones. These traffic patterns
affect the probability and magnitude of multiple pres- Figure 2. Annual maxima smoothed bootstrap (SB) and
ence loading events on bridges, and are significant uncorrelated model (UC).
for the maximum lifetime loading on the bridge. One Note: i + j denotes i trucks in the slow lane and j in the fast
example of these patterns is shown in Figure 1 which lane
suggests that when heavier trucks are in the fast lane,
they tend to be passing another truck in the slow lane. the stream of measured traffic and these are used as
This paper analyses traffic patterns using multi-lane the basis for a multi-dimensional smoothed bootstrap
WIM data collected at two European sites. It describes approach (Efron & Tibshirani 1993) which allows the
an approach to the Monte Carlo simulation of this traf- observed correlation structure to be accurately sim-
fic which seeks to replicate the observed patterns of ulated but also allows for unobserved patterns to be
vehicle weights, same-lane and inter-lane gaps, and simulated. The process has been optimized so as to
vehicle speeds. A method is presented which applies make it possible to simulate traffic loading on bridges
variable bandwidth kernel density estimators (Scott over periods of 1000 years or more, and this removes
1992) to empirical traffic patterns of vehicle weights, much of the variability associated with extrapolating
gaps and speeds. Traffic scenarios are identified in from shorter time periods to calculate the characteristic
maximum lifetime loading. The effects of correlation
on characteristic maximum bridge load effects are
illustrated in Figure 2 for two-lane traffic on a 45 m
bridge which shows that the loading event with one
truck in each lane (1 + 1) governs at the 1000-year
return level, and that the modeling of correlation has
a noticeable effect on all multi-truck loading events.
REFERENCES
Efron, B. & Tibshirani, R. (1993), An Introduction to the
Bootstrap Chapman and Hall/CRC.
Scott, D. W. (1992), Multivariate density estimation: theory,
Figure 1. Inter-lane GVW correlation, the Netherlands. practice, and visualization, New York: Wiley.
211
A filtered measured influence line approach to bridge weigh-in-motion
E.J. OBrien, A. Gonzlez & J. Dowling

ABSTRACT
In Bridge Weigh-in-Motion (B-WIM), an instru-

mented bridge is used as a scales to weigh passing
trucks and their axles. The most common algo-
rithm upon which modern B-WIM systems are based
remains that developed by Moses (1979). The perfor-
mance of this method is well documented; it is very
good at estimating GrossVehicle Weight, but less accu-
rate for individual axles, particularly closely spaced
axles on longer bridges. Many alternatives to Mosess
original algorithm have been tested and some show the
potential to improve accuracy but commercially avail-
able B-WIM systems are still based substantially on
the original approach.
This paper proposes a method of altering the B-
WIM algorithm to improve the accuracy of the pre-
dictions. The measured dynamic signal, to which the Figure 2. Filtered Measured Influence Lines for a range of
algorithm is applied, is first filtered to remove high fre- vehicle speeds.
quency components of the dynamic increment of load,
but filtered at a much lower threshold about 3 Hz the corresponding measured signal into a measured
than is commonly used. The objective is to remove influence line. However, for this work, the measured
those parts of the signal associated with the bridge signal is first filtered to remove much of the high fre-
first natural frequency. quency dynamic components. The exact theoretical
The influence line used by the algorithm is also cal- influence line is illustrated in Figure 1 and the mea-
culated differently. As previously described by OBrien sured filtered influence line in Figure 2. In this way,
et al. (2006) it is determined using a pre-weighed cali- Mosess least squares fitting method is now comparing
bration truck and an algorithm to automatically convert only the low frequency components of the measured
and theoretical responses and produces a much more
accurate fit.
The new approach is tested in numerical models
and it is shown to result in a substantial improvement
in accuracy.
REFERENCES
Moses, F. (1979). Weigh-in-motion system using instru-
mented bridges. ASCE Journal of Transportation Engi-
neering, 105(3), 233249.
OBrien, E.J., Quilligan, M. and Karoumi, R. (2006). Cal-
culating an Influence Line from Direct Measurements.
Proceedings of the Institution of Civil Engineers: Bridge
Figure 1. (Ideal) Influence Line. Engineering, 159(BEI), 3134.
212
Evaluating remaining lifetime of bridges by means of BWIM
M. Petschacher
PEC Consulting, Am Huegel, Feldkirchen/Kaernten, Austria
ABSTRACT
In the evaluation of bridges often considerably more

attention is paid to the modelling of the structure
than to the load this statement mainly applies on
the area of traffic loads. Considering the increasing
amount of bridges with a distinct damage pattern,
state evaluations will become more important, because
also economic considerations depend on them since Figure 1. A fault tree applied on a bridge.
an immediate intervention is not always possible. Due
to ongoing deterioration the requirements of relevant cross-sections are predicted with help of the theory of
codes, which are often mounted on highway traffic, extreme values as well as ongoing deterioration due to
can not be fulfilled. Therefore it is necessary to gain corrosion is considered. With a probabilistic analysis
more information concerning real load and resistance the development of the safety index is calculated and
in order the estimate the failure probability of the decisions for further actions are derived. The model for
structure and to gain decision criteria. the consideration of further findings from following
This paper will deal with the analysis of the inspections is presented.
remaining lifetime of bridges, focusing on a con- The bridge overpasses a highway and is also used
crete example. The model that is displayed employs as entrance and exit ramp of the highway. In 2007 this
a stochastic format for analysis and the dominance of bridge already was evaluated in a similar manner, but
load assumptions in the limit state formulations will mainly data from the literature were used and data from
be pointed out. In consequence of that, the success- a highway was used as a basic to generate a load model,
ful way of using realistic traffic loads will be taken where trailers and semi trailers are dominant vehicle
from BWIM measurements, where BWIM is short groups, see [2]. Now BWIM-measurements showed
for Bridge-Weigh-In-Motion. The data gained at the that at this bridge the local traffic is more impor-
BWIM-measurement are used to form an axle load and tant than the traffic from the highway. The fraction
traffic flow model where the Heavy Goods Vehicles of overloads was quite high.
(HGV) are artificially generated. To gain information Within this paper failure will be described as a fault
on the load impact on the bridge the load step method, tree, consisting of elements shown in figure 1.
see [1], is applied and simulated for n times. Using the
outcomes in a semi probabilistic code manner, char-
acteristic values are evaluated by means of extreme REFERENCES
value statistics. On the other hand the distribution of
extreme values can also be used in a fully probabilistic [1] P. Grundy and G. Boully. Assessment of Bridge
format where of course the deterioration of the bridge Response using Weigh-In-Motion Data. 5th Austro-
has to be considered. roadsBridge Conference, 2004.
In this paper a damaged bridge that can only be kept [2] Petschacher M., 2008. Evaluation of remaining life of a
operational for a limited period is examined. The rated bridge. 4th CCC Symposiom Concrete Engineering in
values for bending moment and shear force in critical Urban Development, Opatija
213
WIM data to assess consequences of new traffic regulations
F. Schmidt & B. Jacob

Laboratoire Central des Ponts et Chausses, France
ABSTRACT Table 1. Auxerre & Libourne bridge lifetime for A9 traffic,

40 or 44 t gross weight limit.
The French traffic regulations allow common articu-
lated trucks and road train to weigh 40 tons on 5 axles. Lifetimes (in years)
However, there are several exceptions to these rules, Auxerre
Fatigue class 40 tons 44 tons
with a weight limit of 44 tons for 40 ft ISO containers in
a combined rail-road transport, trucks picking or deliv- Class 56 10.1 8.5
ering goods in harbors 150 km around them, or some Class 63 14.3 12.0
agricultural good transport. Log trucks are allowed up Class 71 20.1 16.9
to 48 tons on 5 axles and 57 tons on 6 axles. Following
the law Grenelle de lenvironnement on greenhouse Libourne 40 tons 44 tons
gas saving in 2008, the Parliament asked the govern-
ment to study the advantages and disadvantages of Class 50 34.6 28.9
increasing the general weight limit to 44 tons. Class 56 48.9 40.8
To assess and compare the impacts of current traf- Class 63 69.0 58.8
fic loads and future traffic loads with 44 ton trucks
on bridges, WIM data recorded in 2009 on a heavily Rices formula, to assess the extreme stress values for
trafficked motorway in south of France were used by a various return periods from 25 to 5000 years (Flint &
software developed in the LCPC (Eymard & Jacob Jacob, 1996). Rain-flow histograms were also calcu-
1989) to calculate extreme load effects and fatigue lated and used with the Miners law and S-N curves
damages on two composite bridges. These WIM data (CEN EN1993-1-9 2005) to assess the detail lifetime
were then modified by a micro-simulation based on in fatigue (Jacob and Kretz 1996).
simple assumptions on the increase of some truck The modified traffic with 44 ton trucks induced
weights from 40 to 44 tons. maximum stresses 6.5 to 8.5% higher than the recorded
The stress influence lines of details, such as the traffic. The detail lifetime are decreased, up to 20%,
welds between vertical stiffeners and lower flanges with these increased traffic loads (Table 1).
of steel main girders, for a 40 m single span simple
supported bridge (Auxerre) and a 4 continuous span
REFERENCES
bridges (Libourne) were used. They have already been
used for the calibration of the Eurocode 1 (Bruls & al. Bruls, A., Croce, P., Sanpaolesi, L. & Sedlacek, G. (1996).
1996, CEN EN1991-2 2003). The level crossing his- EVN1991-Part 3: Traffic Loads on Bridges; Calibration
tograms of stresses were calculated and used, with the of Load Models for Road Bridges, Proceedings of IABSE
Colloquium, Delft, The Netherlands, IABSE, 439453.
CEN EN1991-2 (2003). Eurocode 1: Actions on structures
Part 2: Traffic loads on bridges.
CEN EN1993-1-9 (2005). Eurocode 3: Design of steel
structures Part 19: Fatigue strength of steel structures.
Eymard, R. & B. Jacob (1989). Un nouveau logiciel : le
programme CASTOR pour le Calcul des Actions et Sol-
licitations du Trafic dans les Ouvrages Routiers. Bulletin
de liaison des laboratoires des Ponts et Chausses (164):
6478.
Flint, A. & Jacob, B., (1996), Extreme Traffic Loads on
Road Bridges and Target Values of their Effects for
Code Calibration, Proceedings of IABSE Colloquium
Basis of Design and Actions on Structures, IABSE, Delft,
469477.
Jacob, B. & Kretz, T. (1996). Calibration of bridge fatigue
Figure 1. Rain-flow histogram of the stresses in Auxerre loads under real traffic conditions. Proceedings of the
bridge at mid-span, lower flange, both traffics. IABSE Colloquium on Eurocodes, Delft, March, 479487.
214
Vehicle loading and effect on the Tsing Ma Bridge using WIM data
Y.L. Xu & Z.W. Chen

The Hong Kong Polytechnic University, Hong Kong, China
K.Y. Wong
Highways Department, Hong Kong, China
ABSTRACT The statistical data on highway traffic composition

in 2006 revealed that a total of 8.5 million vehicles run
Many innovative long span bridges made of new mate- through the Tsing Ma Bridge for the Airport bound
rials and implemented by new technologies have been way and 8.8 million vehicles for the Kowloon bound
built in recent years throughout the world. These long way. The percentage of heavy goods vehicles, includ-
span bridges surrounded by harsher environment than ing rigid heavy goods vehicles and articulated heavy
ever before engender many challenges to profession- goods vehicles, was about 5% to 6% of the total vehi-
als on how to ensure these structures function properly cles. Most of heavy goods vehicles used the slow lane
during their long service lives and how to prevent for both the Airport and Kowloon bound ways. The
them from sudden failure. Recently-developed struc- statistical data on axle load distribution in 2006 man-
tural health monitoring technology provides a better ifested that only 12.5% of the total vehicle axles had
solution for some challenges (ISIS Canada, 2001). an axle load more than 5 tonnes. The statistical data on
In Hong Kong and Mainland China, structural health GVW distribution in 2006 showed that only 7.5% of
monitoring systems have been installed in more than the total vehicles had an GVW more than 16 tonnes.
40 long span bridges. The axle load spectra and GVW spectrum from the
The Tsing Ma Bridge in Hong Kong is a suspension measured WIM data in 2006 and based on BS5400
bridge with a main span of 1,377 m. The bridge deck both were on the safer side compared with the design
carries a dual three-lane highway on the upper level spectrum.
of the deck and two railway tracks and two carriage- The results of fatigue life evaluation demonstrated
ways on the lower level within the bridge deck. The that the fatigue life estimated based on the daily data
Hong Kong Highways Department installed a com- length was almost the same as that based on the
prehensive Wind And Structural Health Monitoring monthly data length. The fatigue life estimated based
System (WASHMS) in the Tsing Ma Bridge in 1997 on the connection class F was much longer than that
(Wong et al. 2001). This paper takes the Tsing Ma based on the connection class F2. The fatigue dam-
Bridge as an example to manifest how the WASHMS age to the steel connections of theTsing Ma Bridge was
installed in the bridge is used for monitoring highway mainly induced by moving trains. The contribution of
traffic condition, vehicle loading, and vehicle load- road vehicles to fatigue damage was rather small. The
ing effects on the bridge. The configuration and the fatigue life of the bridge is very long for some con-
relevant sensors in the WASHMS of the Tsing Ma nections, and all the instrumented components of the
Bridge are briefly introduced. By using the traffic data bridge are in good conditions.
collected by weigh-in-motion (WIM) stations of the
WASHMS, the highway traffic condition and vehicle
loading on the bridge are analysed in terms of vehi- REFERENCES
cle traffic volume, vehicle traffic composition, axle
load spectrum and gross vehicle weight spectrum. The ISIS Canada, 2001. Guidelines for structural health monitor-
measured axle load spectrum and gross vehicle weight ing, Design Manual No. 2. ISIS Canada Corporation.
spectrum are compared with the design spectrum. The Wong, K.Y., Man, K.L. & Chan, W.Y.K. 2001(a). Monitoring
of wind load and response for cable-supported bridges in
strain data collected by 110 strain gauges installed at
Hong Kong, Proceedings of SPIE 6th International Sym-
three sections of the bridge deck are analysed. A prac- posium on NDE for Health Monitoring and Diagnostics,
tical method in conjunction with the method proposed Health Monitoring and Management of Civil Infrastruc-
in BS5400: Part 10 is used to estimate the fatigue life of ture Systems (Chase and Aktan eds.), Newport Beach,
the bridge due to highway loading and railway loading. California, 4337, 292303.
215
Latest practical developments in the bridge WIM technology
A. nidaric, I. Lavric & J. Kalin

Slovenian National Building and Civil Engineering Institute, Ljubljana, Slovenia
ABSTRACT
The European Commission research project WAVE

(2002) that ended almost 10 years ago, introduced
some substantial improvements of the bridge WIM (B-
WIM) technology. Since then developments continued
and today the B-WIM systems are accepted in many
countries around the world as an important player on
the WIM market. They can be equipped with solar or
fuel cell power units, traffic cameras and other sen-
sors and are collecting data for various applications,
from traffic studies, design and reconstruction of pave-
ments, pre-selection of overloaded vehicles and bridge Figure 1. Typical B-WIM installation.
applications.
Bridge WIM systems have several advantages over
the pavement systems. One is their full portability,
which allows rotating of the same equipment around their axles, if they are detected on the bridge at the
several locations. The free-of-axle detector (FAD) same time. In the B-WIM practice this problem, known
technology (nidaric, Lavric, & Kalin, 2005) does not as multiple-presence of vehicles, is being mitigated
require any actions on the pavement during installation by the strips method, which takes into account dis-
and life-time of the system. Results on some types of tribution of traffic loading from one lane to another.
bridges, especially on shorter spans, are very accurate Applying the measured lane factors can be seen as
and, last but not least, they can in parallel measure the an intermediate step towards implementation of a 2-
structural parameters that are essential for optimized dimensional influence surface, which is foreseen as
bridge assessment. the next major improvement of B-WIM algorithms
In the last few years the B-WIM developments but would, contrary to strips, require by far more
focused on one hand on new algorithms that would comprehensive calibration procedure.
further improve accuracy of the results, and on the Recent studies also demonstrated how accuracy of
other hand on new or improved measuring techniques, results is heavily affected by the road roughness and
that would allow them to be used on bridges that were by the type of calibration. Smooth (resurfaced) pave-
traditionally less or not appropriate for B-WIM mea- ment and employing several calibration factors (WIM
surements. Some of these developments are discussed calibration by vehicle type) can considerably increase
in the paper. accuracy of measurements.
Generation of the experimental influence lines that
are based on the measured strain responses of the
REFERENCES
instrumented structural elements has proven crucial
for the high accuracy of B-WIM results, especially the WAVE. (2002). Weigh-in-Motion of Axles and Vehicles for
axle loads. The influence lines used for calculations are Europe, General report. (B. Jacob, Ed.) Paris: LCPC.
obtained from a number of strain responses of random nidaric, A., Lavric, I., & Kalin, J. (2005). Nothing-on-
vehicles passing the bridge. the-Road axle detection with threshold analysis. 4th
On bridges longer than 20 meters the traditional International Conference on Weigh-in-Motion (ICWIM4).
B-WIM algorithm is not able to properly divide the E.J. OBrien, B. Jacob, A. Gonzales, CP. Chou (eds.)
total measured loading between several vehicles and (pp. 95106). Taipei: National Taiwan University.
216
Applications of B-WIM technology to bridge assessment
A. nidaric & I. Lavric

ABSTRACT
Weigh-in-motion (WIM) data has been for many years

used for bridge applications, primarily for calibration
of bridge design and sometimes of bridge assessment
codes. The data used was primarily obtained from the
pavement WIM systems that did provide the necessary
information about realistic traffic loading but could not
give any additional information about the behavior of
the structures. These can be obtained by a bridge WIM
(B-WIM) system if it is installed on the bridge being
assessed (nidaric, Lavric, & Kalin, 2010). Figure 1. Typical relation between the maximum static
Influence lines are the key factor for quality B-WIM strains of loading events and DAF.
measurements. They should be directly derived from
the measured data on the site. deviation with increasing weight. However, as results
Load testing, especially the diagnostic one, effi- at this stage are still limited, the ARCHES project
ciently optimizes structural models used for assess- recommended using, with individual vehicles on the
ment. To avoid high costs of traditional load tests, the bridge, DAF equal to 1.15 to 1.20, depending on the
soft load testing was proposed. With the same objec- smoothness of the pavement.
tive to optimize the structural model, it uses a B-WIM In order to fully verify them, the new DAF mea-
and random traffic rather than some pre-weighed vehi- surement and evaluation procedures are being further
cles, to monitor the real structural behavior of the developed.
bridge under traffic loading. Due to the lower level of The evident goal of the described methods, which
loading the results are more conservative than those include monitoring of the structural behavior (soft
from the traditional diagnostic load test, but can still load testing) and traffic loading, is to optimize bridge
result in large optimizations of results. At the same assessment. If used, these results will prevent from
time the procedure is considerably cheaper than a tradi- spending the always limited financial resources for
tional load test and does not require closing the traffic. unnecessary expensive interventions on bridges, such
To this stage it has been applied to shorter span bridges a strengthening and replacements.
(up to 40 m), but this does include the majority, even
over 90% of the entire bridge stock in most of the
countries. REFERENCES
The B-WIM systems have also been used for mea-
surements of static loading of traffic and dynamic ARCHES. (2009). Report D08: Recommendations on bridge
amplification factors. The results of convolution, traffic load monitoring. Bruselj: Evropska komisija.
to obtain the expected maximum static loading ARCHES. (2009). Report D10: Recommendations on
effects, and the measured Dynamic amplification fac- dynamic amplification allowance. Bruselj: Evropska
komisija.
tors compared very well with the analytical results nidaric, A., Lavric, I., & Kalin, J. (2010). Latest practical
(ARCHES, 2009). For the DAF, the results of thou- developments in the Bridge WIM technology. Proceed-
sands of measured loading events demonstrate an ings of the Fifth International Conference on Bridge
inverse proportionality between vehicle weight and Maintenance, Safety and Management IABMAS2010.
dynamic amplification and also a reduction of standard Philadelphia, USA.
217
MS7: Bridge management tools & research
Organizer: L. Klatter
A methodology to estimate risk related to road links, due to latent processes
B.T. Adey
Infrastructure Management Group, Institute for Construction Engineering and Management,
Swiss Federal Institute of Technology, Zurich, Switzerland
J.D. Birdsall
Parsons Transportation Group, Washington D.C., United States of America
(formerly Infrastructure Management Consultants, Zurich, Switzerland)
R. Hajdin
Infrastructure Management Consultants, Zurich, Switzerland
ABSTRACT
A methodology is presented that has potential to be

used in infrastructure management systems to deter-
mine risk related to road links, due to latent processes.
The methodology was developed taking into consider-
ation the current infrastructure management systems
and the availability of data in Switzerland. It is par-
ticularly relevant for Switzerland, but in general is
applicable for all road infrastructures. The method-
ology was developed with the following conceptual
view of a road link: A road link is composed of objects
that are both clustered based on topographical char-
acteristics and composed of object sections in the
longitudinal direction. Objects are clustered hierarchi-
cally based on the topographical characteristics of their
surroundings since natural hazards, in many cases, are
Figure 1. Graphical depiction of the aggregation of risk
governed by these topographical characteristics. using modification factors.
The risk related to a link is estimated by aggre-
gating the risk of each item (object section, object or
cluster) on the lowest hierarchical level to which they
belong to give the risk of the item on the next high- assessment of the object section using predeter-
est hierarchical level, until the risk related to the link mined guidelines, and
is estimated using appropriate modification factors. A continuous functions be developed from the dis-
graphical depiction of this aggregation for the flood crete data to assess the consequences of failure for
hazard is given in Figure 1. failure modes at each hazard parameter intensity.
The estimation of risk using the methodology The methodology proposed has potential to be used
requires the estimation of the probabilities of occur- in Switzerland for the evaluation of risk related to road
rence of hazard parameter intensities, the determina- links with respect to one hazard that may result in
tion of the failure modes of the objects for each hazard multiple simultaneous failures on one link within one
parameter intensity, and the determination of interven- period. More research, however, is required to verify
tion and other costs incurred for each failure mode. It the assumptions made.
is proposed that:
the probability of occurrence of hazard parameter
REFERENCES
intensities be determined by conducting geographic
coincident analyses between the investigated haz- Birdsall, J., Adey, B.T., Hajdin, R., Erath, A., & Axhausen, K.
ard and the infrastructure objects. in press, A methodology to estimate risk related to links in
the failure mode for each hazard parameter the Swiss road network due to a hazard within one period.
intensity be determined by conducting a failure Swiss National Science Foundation.
221
A framework for comprehensive estimation of user costs for bridge

management: A synopsis of existing practices and discussion of new
considerations
Q. Bai, S. Labi, G.P. Ong, A. Bhargava & K.C. Sinha

Purdue University, West Lafayette, IN, USA
P.D. Thompson
Paul D. Thompson, Castle Rock. CO. USA
ABSTRACT certain bridges, the user may detour for more than one
reason, and not accounting for this situation may lead
Bridge agencies strive to incorporate stakeholder per- to double counting of user costs. This issue of double
spectives in their decision-making processes. One way counting has not been explicitly considered in the liter-
of doing this is to include user costs as a key perature. By using principle of set theory, the paper shows
formance criterion in investment evaluation. While how the estimation of user costs of detouring could
agencies seek to manage their bridges to meet user be carried out in a manner that avoids double count-
expectations of performance, they are often stymied ing. In addition, the paper improves the methodology
by lack of a comprehensive and consistent framework for calculating bridge delay user cost by updating the
for assessing the several different user cost types and speed-volume function.
components. This paper tries to establish a framework The applicability of the framework is then demon-
for comprehensively estimating user costs for bridge strated using a case study that calculated a broad array
management. of bridge user cost categories for a steel bridge.
The paper first identifies the factors of bridge user The results show that when the double counting
cost incurrence and the various categories of user cost issue is not taken into account, the calculated user
(Table 1). This is done on the basis of a synthesis of the cost is higher. Specific results for this case study sug-
existing state of practice and some new considerations. gest that the calculated workzone user cost (whether
For purposes of this presentation, the bridge user the double counting issue is addressed or otherwise)
cost is categorized into detour cost, delay cost and can be extremely high when the bridge is closed for
accident cost. Also, it is recognized that the bridge rehabilitation and that the user cost of delay reduces
life-cycle user cost consists of the cost incurred dur- drastically after the bridge widening intervention.
ing the period of normal operations of the bridge and Summing up, the case study results suggest that it is
those incurred during the work zone period. The lit- prudent to appreciate the problem of double counting
erature review showed that relatively few studies had and to address this issue duly. The results also suggest
addressed the latter category of user cost. The paper that work zone user cost is a critical and dominant com-
provides a refined and detailed procedure to calculate ponent of bridge user costs and needs to be considered
bridge detour cost and delay user cost. In address- in any bridge investment evaluation problem.
ing key considerations, the paper recognizes that for
REFERENCES
Table 1. User cost factors and possible user costs.
Johnston, D.W., Chen, C., and Abed-Al-Rahim, I. 1994.
Detour Delay Accident Developing user costs for bridge management systems.
Traffic Causes Cost Cost cost Transportation Research Circular 423: 139149.
Son, Y. & Sinha, K.C. 1997. Methodology to Estimate User
On Load Capacity Limit Yes Costs in Indiana Bridge Management System. Trans-
bridge Vertical clearance limit over bridge Yes Yes portation Research Record: Journal of the Transportation
Horiz. clearance limit over bridge Yes Yes Research Board 1597: 4351.
Poor alignment Yes Yes Thompson, P.D., Najafi, F.T., Soares, R., and Kerr, R. 2000.
Low capacity Yes Development of Pontis User Cost Models for Florida.
Work zone Yes Yes Yes Transportation Research Circular 498: D-1/1D-1/15.
Under Vert. Clearance limit under bridge Yes Yes Thompson, P. D., Soares, R. H., Choung, J., Najafi , F. T., and
bridge Horiz. clearance limit under bridge Yes Yes Kerr, R. 2000. User Cost Model for Bridge Management
Work zone Yes Yes Yes Systems. Transportation Research Record: Journal of the
Transportation Research Board 1697: 613.
222
Bridge management technique: Implementation of a deterioration model

to different highway networks
P. Crespi
D. Oggionni & G. Pasqualato

Sineco S.p.A., Milan, Italy
ABSTRACT data. Nevertheless, in some particular cases, despite

of a great amount of data available, it is not possible
Sineco performs from several years detailed inspec- to draw up a deterioration model for a medium-long
tions on bridge structures of some motorway networks time forecast.
in northern Italy, in order to perform maintenance To overcome this problem, a theoretical method
programs as an integrated method between the inspec- has been developed to draw up deterioration curves
tion results database created by Sineco since 1995 and using available tests on concrete carbonatation on dif-
the Bms tools provided by the software Pontis. Such ferent elements of the bridges of Savona-Ventimiglia
ask data analysis defines optimal preservation, pol- motorway (a forty years old motorway in which all
icy identification and recommendation; bridge needs the structures are in quite good conditions). In this
and performance measure forecasts and development case both the deterioration models determined by real
of projects are included in agencies economical inspection data and created by the experience or liter-
planning. ature on similar motorways are unable to describe and
Simultaneous use of the two systems consent, by simulate the real condition of analyzed bridges.
one side to preserve the high level of detail of different To calculate the carbonatation thickness also on
data recorded during inspections, that are referred to those elements for which direct tests are not available,
any single defect, and to exploit also that historical have been used diagrams that correlates carbonation
data collected before the introduction of Pontis system, thicknesses s and cubic concrete strength Rck .
and by the other side, to develop cost and deterioration So it has been possible to determinate the whole ser-
forecasting models necessary to achieve life cycle cost vice life of structural elements as the sum between the
analysis and long term programming. corrosion initiation time ti (carbonatation time of con-
Deterioration forecasting models can be developed crete cover) and the propagation time tp (ratio between
from different source of information. Sineco first step limit penetration and corrosion speed). Then the ser-
in defining forecasting models has been compiling vice life of each structural element has been divided in
expert elicitation data. This elicitation process is used the different condition states, and finally the transition
to create deterioration models solely based on expert probabilities to one condition state to the next one is
judgment. This allows reasonable models to be devel- calculated by means of the expression Pj = 1 1/n,
oped before having actual data, which take several where n is the time of staying in one condition state.
years to be collected. The resulting forecasting model is more conserva-
Expert transition probabilities are updated using tive than the previous expert one, but it is useful to
historical inspection data collected by Sineco for sev- obtain significant results and to manage reliable cost-
eral years, accumulating all possible pairs of succes- benefit analysis, and so to draw up a realistic and
sive registered condition state of the elements, subject functional maintenance plan.
to certain constraints (e.g., there should not have been This method, that can be repeated for any other
a spontaneous improvement in the element condition bridges group in which inspection data are not enough
between inspections). or not useful, provide a realistic deterioration model
It is generally possible to create deterioration mod- that allows to simulate future deterioration and obtain
els very customized and reliable for different bridge a priority list of maintenance works in a medium-long
networks using the amount of historical inspection term scenario.
223
Structural monitoring with wireless sensor networks: Lessons learned

from field deployments
G. Feltrin, R. Bischoff, J. Meyer & O. Saukh

Empa, Swiss Federal Laboratories of Materials Testing and Research, Dbendorf, Switzerland
ABSTRACT
In the last years, wireless sensor networks have

emerged as a promising technology that is inducing
a deep innovation in the field of structural monitoring.
The main advantages of wireless sensor networks com-
pared to conventional monitoring technologies are fast
deployment, little interference and self-organization.
However, since wireless sensor network nodes are bat-
tery powered, in long term monitoring applications
the power management influences significantly the
operation of a wireless sensor network. In data inten-
sive applications, e.g. vibration based monitoring, low
power hardware, duty cycling and efficient commu-
nication policies are not sufficient for achieving a
sustainable system life-time. Since data communica-
tion is the most energy consuming task, a significant
data reduction has to be attained in the nodes to achieve
sustainable system life times. This data reduction is
a challenging task, since it has to be performed with
very limited computational and memory resources and
in competition with the basic network functionality.
The objective of the paper is to present our expe- Figure 1. a) Natural frequencies of monitored stay-cables.
rience over the past three years with data intensive b) Monitored natural frequencies smoothed with a moving
structural monitoring using wireless sensor networks. average filter.
Specific aspects of sensing, data quality, stability,
availability, and system life-time are presented and
analyzed. These results stem mainly from a still on- save power, destabilized seriously the wireless sensor
going field test started in November 2006 on a cable network by preventing the basic operations like time
stay bridge. The goal was to monitor the natural fre- synchronization and routing.
quencies of stay-cables by using ambient vibration Nevertheless, the progress made during these years
records (Figure 1). demonstrates that data intensive structural monitor-
These 3 years of field test experience demon- ing with wireless sensor networks is feasible. The
strate that operating a wireless networks reliably problems could be substantially solved with several
over a period of months or years was a non-trivial hard- or software improvements. Despite the severe
task. The problems relied basically on balancing the hard- and software limitations, the generated informa-
requirements of a data intensive application with the tion complies with the quality requirements in civil
requirements of minimizing power consumption for engineering, which usually do not require high preci-
a achieving a sufficiently long battery lifetime. The sion information. The hardware limitations, however,
omnipresent goal to save power generated problems require a tight specialization to the monitoring task.
like sensor signal corruption by duty cycling or switch- This implies a detailed analysis and specification of
on operations. Data reduction, a powerful method to the monitoring goals.
224
An improved model for predicting NBI condition ratings
M.J. Fraher & S. Hwang

Cambridge Systematics, Inc. Cambridge, Massachusetts, USA
J. Fu
Hawaii Department of Transportation, Kapolei, Hawaii, USA
ABSTRACT Table 1. NBI to health index conversion table.
The Federal Highway Administration (FHWA) NBI Ratings

National Bridge Inventory (NBI) Translator pre- NBI Item 4 5 6 7 8
dicts the NBI condition ratings for deck, superstruc-
ture, substructure and culvert given the element-level Deck <25 2580 8089 8998 <98
inspection data. The American Association of State Superstructure <72 7281 8189 8998 <98
Highway and Transportation Officials (AASHTO) Substructure <72 7281 8189 8998 <98
Pontis Bridge Management System (BMS) uses the
NBI Translator as part of the systems program sim-
ulation to predict future NBI condition ratings and
structurally deficient or functionally obsolete status. In With the health index approach, the impact on the
using Pontis, the Hawaii Department of Transportation components health index is minor when small quanti-
(HDOT) determined that the NBI Translator overesti- ties are predicted for the worst condition state, unlike
mates the poor ratings for NBI components and the the large impact predicted by the NBI Translator. This
number of structurally deficient bridges. The Transla- approach generates predicted NBI rating values that
tor logic accentuates the small quantities in the worst are well correlated with the inspector assigned NBI
condition state, and this leads to an overprediction of ratings and consistent with the health index. HDOT
bridges with poor NBI ratings. has implemented this approach for recalculation of
This paper describes an approach developed by the the condition ratings generated by Pontis simulation.
authors for using the health index to translate element- Future versions of Pontis should consider including
level inspection data to NBI condition ratings. The the health index based prediction model.
approach was developed for HDOT to provide more
accurate predictions of future bridge conditions and
to support its bridge programming. The approach con-
sists of devising a conversion table that translates the REFERENCES
health index ranges into NBI ratings and applying the AASHTO. AASHTO Guide for Commonly Recognized
converted NBI ratings to Pontis simulation results. (CoRe) Structural Elements, 1997 (amended 2001).
The review of the HDOT inspection data showed Cambridge Systematics, Inc. 2006. Pontis Release 5 Techni-
that the average health index of each component group cal Manual. Washington, D.C.: AASHTO.
and the NBI ratings assigned by the inspectors were Hearn, G., Cavallin, J. & Frangopol, D.M. 1997. Generation
well correlated. This finding was also true when the of NBI Ratings from Condition Reports for Commonly
same comparison was performed using data from Recognized Elements. Denver, CO: University of Col-
Arkansas, Idaho, Oklahoma, and Wyoming. Based on orado at Boulder, Colorado Department of Transportation,
and the Federal Highway Administration.
this review, a conversion table was developed which Scherschligt, D. & Kulkarni, R.B. 2003. Pontis-Based health
is presented in Table 1. The table assigns values for Indexes for Bridge Priority Evaluation. Transportation
NBI ratings of 48, which are the most prevalent in Research E-Circular E-C049: 9th International Bridge
the HDOT data set. The health index ranges in Table 1 Management Conference,Transportation Research Board.
were used to assign simulation NBI ratings based on Shepard, R. & Johnson, M. 2001. California Bridge Health
the predicted health index. Index. TRNews 215.
225
Bridge maintenance and practical bridge management systems in Japan
H. Furuta
Department of Informatics, Kansai University, Takatsuki, Japan
E. Watanabe
Kyoto University, Kyoto, Japan
1 INTRODUCTION owner of expressway bridges should perform more fre-

quent inspection and repair to maintain the bridges in
In Japan, maintenance of existing bridges is an impor- better conditions. Here two groups of bridge owners
tant and urgent issue, because there are 150,000 are selected; one is the local governments and the other
bridges with more than 15 m span should be main- is expressway corporations.
tained in good conditions. Furthermore, Japan has As local governments Osaka prefectural govern-
severe environments surrounding bridge maintenance ment, Aomori prefectural government and Osaka
as follows: 1) Reduction of investment for infras- municipal government are selected, and three express-
tructures, 2) Increase of renewal investment, 3) way corporations such as NEXCO, Hanshin Express-
Increase of maintenance cost, 4) Increase of stock of way and Honshu-Shikoku Expressway are considered.
infrastructures. Osaka prefectural government, Aomori prefectural
In order to establish a rational maintenance pro- government and Osaka municipal governments are
gram and provide a basis for accountability, several maintaining old bridges based on the concept of
Bridge Management Systems (BMS) have been devel- LCC. NEXCO built BMS in 2003 for the preventive
oped. In this paper, basic ideas and frameworks of bridge maintenance and the system has been applied
bridge maintenance and representative five BMSs are to bridge management in NEXCO since then to keep
introduced, which have been practically applied by the safety and security of the expressway. Honshu-
local governments in Japan, such as Osaka prefectural Shikoku expressway has focused on the preventive
government, Aomori prefectural government, Osaka maintenance so far, because they have to mainly man-
municipal government, Nippon Expressway Company age long spanned bridges. Hanshin expressway has
Limited (NEXCO), Honshu-Shikoku Expressway, and been paying attention to risk evaluation and finan-
Hanshin Expressway. The characteristics are described cial analysis for road maintenance on urban express-
and compared to each other, through which their way based on H-BMS (Hanshin Expressway Bridge
differences are made clear. Management System).
2 BRIDGE MANAGEMENT SYSTEMS IN 3 AOMORIS BMS AND H-BMS

JAPAN
In this paper, as representative BMS for local gov-
In recent years, many Bridge Management Systems ernment and expressway corporation, Aomoris BMS
(BMS) or Asset Management Systems (AMS) have and H-BMS are introduced in detail. Aomori Prefec-
been developed in Japan. Although the basic concept tural Government has developed the Aomori Bridge
and procedure used in the systems do not significantly Management System (ABMS) in 2004 and 2005, and
differ from each other, some details are different by began its bridge management in 2006 by using ABMS.
reflecting the peculiar characteristics of bridge own- H-BMS is a maintenance management system which
ers. Namely, damage states and damage causes are has functions to calculate optimal repair policies by
different so that the focusing or emphasizing points minimizing life cycle costs, to simulate future con-
are not the same. dition state and repair costs and to determine repair
Expressway bridges require higher safety level than priority.
the ordinal road bridges. This means that the bridge
226
Contribution of non-destructive defect detection to bridge management
R. Helmerich & B. Milmann

Federal Institute for Materials Research and Testing, Non-destructive Damage Assessment
and Environmental Methods
ABSTRACT
Regular inspection in structural and bridge engi-

neering may result in doubts about the integrity of
structures or critical structural details. In some cases,
information is not available or not detailed enough
for the planning of further maintenance decisions.
Bridge owners and providers need additional informa-
tion about the integrity of the structure or about the Figure 1. Sensor prototype and image with detected defects.
success of strengthening measures in order to assess
the safe and reliable further service of the structure. thinner than the wave length. The excitation is received
Often, according to the possibilities of the structures by a second sensor. The signal is an information about
owner, only destructive testing (as e.g. coring) can pro- the degree of the bond integrity between any non-
vide the required information of the inner structure magnetisable thin layered materials. The method was
of reinforced concrete structures, about the current applied in laboratory and in field. Bond defects may be
materials condition or about the quality of repair. deteriorating and separating surface layers, insufficient
After the enhancing development of methods and repair quality or debonding of layered strengthening
data analysis within the last decades, Non-Destructive material as CFRP due to degradation or load effects.
Testing is recommended in special inspection to inves- to physical characteristics, material-related properties
tigate the inner structure. Images from evaluated data and sensor reliability.
answer questions of interest and provide information The lack of a reliable technique for the investiga-
on further details. The German Road Bridge Man- tion of delamination between thin non-metallic layers
agement System SIB-Bauwerke proposes to call in resulted first in the confection of an appropriate sensor
object-related damage analysis of a bridge structure with piezo-ceramic point contact transducers working
(special inspection) in any case of doubts that are aris- with low frequencies. The prototype was used during
ing either from regular inspection, heavy deterioration first applications.
or aging, from impact of accidents resp. to identify the Besides laboratory investigation, an extended field
execution quality of repair. study with application of further non-destructive test-
Latest steps in research and development for the ing methods was carried out. Task of the study was i.e.
application of acoustic non-destructive testing to rein- bond control using ultrasonic method.
forced concrete structures resulted in advanced sensors
and scanning tools, allowing reliable data acquisition
for the condition appraisal. The data processing as REFERENCES
well as data fusion of two or more different mea-
surements using different sensors or even different Helmerich, R, Milmann, B.: 2007. Quality assurance for
NSMR, Workmanship control at a the Frvi Viadukt/
measurement techniques increases the accuracy of the
Sweden, unpublished background report, FP7 project
obtained details. Sustainable bridges.
The paper describes the opportunities for detection Helmerich, R., Wiggenhauser, H.: 2008. Non-destructive
of near surface delamination between non-metallic Testing of Concrete Structures, 4th international Con-
materials using acoustic sensors. A dry coupling point ference on Advances in Mechanical Engineering and
contact transducers causes a mechanical wave in layers Mechanics.
227
Bridge inspection: Are we getting it right?
B. Kamya
Kamya Consulting Ltd, United Kingdom
ABSTRACT An evaluation by the author on the current practice of

inspection and bridge management, found that many
For many bridge management authorities, the overar- authorities still rely on methods where an inspector
ching inspection principles are to detect defects likely assesses defects on site, provide them a ranking and
to cause unacceptable serviceability risks, safety to using a bridge management program a grading for each
the public, structure or the environment. Inspections structure is made.
provide information used in the planning and manage- As in previous presentations, the author believes this
ment of bridges. However, there are still noticeable approach will still lead to wasted resources and con-
increases in costs associated with bridge maintenance. tinued backlog to the maintenance programme. For
Catastrophic failures in service and increasing backlog example, an inspector may find a crack in a bridge.
to the management programs still make headline news However, presence of a crack does not necessarily
even with improvements in training, certification and imply a significant risk. Further understanding of the
use of experienced and competent inspection staff. type, characteristics, location, and effect of the crack
In previous paper presentation to IABMAS06 & 08 on the structural and serviceability behaviour of the
I have advocated the issue of training and certification bridge is required.
of bridge inspectors. Unfortunately, current practice This paper, in conclusion, proposes a radical
still relies on prescriptive guidance such as inspec- approach to the training and certification of bridge
tion work being carried out by appropriately qualified, inspector.
trained and experienced personnel. In the UK, for
example, this is prescribed as, a programme of con-
tinuing professional development (CPD) and training REFERENCES
for bridge managers, engineers, inspectors and other
staff should be provided to enable them to understand Inspection Manual For Highway Structures. London TSO
2007.
and implement the process described in the code Management of Highway Structures: A code of Practice 2005
This paper, argues that this prescriptive approach Wisconsin Department of Transportation, An evaluation
falls short of what is required. For example, it lacks of Bridge Inspection Program, 01-17, 20012002 Joint
emphasis on what skills or experience that the inspec- Legislative Audit Committee Members Report, October
tors should be getting training in or experience. 2000.
228
Maximizing return on investment utilizing a bridge depreciation model
H.S. Kleywegt
Keystone Bridge Management Corp., Kingston, ON, Canada
ABSTRACT It is shown that for a normally age distributed bridge

population and assuming straight-line depreciation,
The challenge facing bridge managers is how to the depreciated value of all of the assets should be
maximize investment in their infrastructure given maintained at above 50% of the new value. Similarly
very limited budgets. There are few tools available for a parabolic decay function, the assets should retain
to demonstrate that scarce monetary resources are 67% of their new value under ideal circumstances.
effectively utilized. Capital investment in a bridge either renews a com-
An adaptation of basic accounting principles is ponent or extends the life of a component. When a
employed to rationalize bridge expenditures. The component is renewed, its depreciated value is reset
model is that of a very basic business model that seeks to the new value. If a component is rehabilitated, the
to maximize value by obtaining the most yield from original service life of the component is extended.
strategic investments. The efficacy of investment is measured as the ratio
The fundamental premise is that of maintaining the of improvement of the depreciated value of the bridge
asset value of a bridge population at the highest pos- to the amount of capital investment required.
sible level. Depreciation continually erodes the book There are significant overhead burdens that influ-
value of a jurisdictions structures. Rehabilitation and ence the capital cost of bridge rehabilitation or
renewal of bridge components increases the measured renewal. These overheads are real but do not contribute
asset value of the bridge fleet.The accounting and engi- to improving the asset value of the bridge fleet. Hence
neering challenge is to identify those capital invest- the model gives credence to the notion of Get in, get
ments that yield the highest return as measured by the out, and stay out.
increase in value of the bridge assets. The measure of Calculating the efficacy of investment provides a
yield is termed the efficacy of investment. rational basis for optimizing investment in bridges
The asset value of a bridge is determined in two where all other considerations are relatively equal.
steps; the undepreciated value and the depreciated
value. The undepreciated value is obtained by breaking
a bridge into its components and assigning a value to
each component based on its geometric and material REFERENCES
properties, and assigning a base new value on a unit
rate basis. In this manner the asset value of a single Holmes, O.W. The Deacons Masterpiece or The Wonder-
bridge and a bridge cohort is assessed. ful One-Hoss Shay: A Logical Story. Houghton, Mifflin
The depreciated value of a bridge is calculated by (eds), 1858The Complete PoeticalWorks of OliverWendell
Holmes, Cambridge: The Riverside Press.
applying decay functions to the value of each bridge
Kleywegt, H.S. 2008. Improving Bridge Component Deterio-
component. Each bridge component is given a deemed ration Forecasting Precision. In Hyun-Moo Koh & Dan M.
service life. The decay function depreciates the com- Frangopol (eds), Bridge Maintenance, Safety, Manage-
ponent value by comparing its age against its deemed ment, Health Monitoring and Informatics; Proc. 4th
service life. Both straight-line and parabolic decay intern. conf., Seoul, 1317 July 2008. London: Taylor &
functions are tested. Francis Group.
229
Trends in development of bridge management systems
J.S. Kong & J.H. Kim

School of Civil, Environmental and Architectural Engineering, Korea University
B.T. Adey
Infrastructure Management Consultants GmbH
L. Klatter
Ministry of Transport, Public Works and Water Management, Center for Public Works
ABSTRACT For the characteristics of inspection measure, it can

be said that the current trend in inspection measure is
This paper focuses on the evolution of how latent pro- based on subjective condition even though some safety
cesses are being considered, evolution of the modeling related measures are added to cover the weakness of
of deterioration, evolution of the optimization tech- quantitative but subjective condition based on visual
niques being used to find the optimal intervention inspection.
strategies in different bridge management systems. To construct a bridge management system with
The early generation of most BMSs started as a better performance, an effort needs to be made to
database to handle bridge inventory. The capability of obtain reliable data even though this would be a time
BMS becomes modernized to help manage bridges consuming procedure. Also, the bridge management
over operational criteria satisfying safety and service- system always needs to be used with inspection results
ability restrictions (Cho et al. 2007). As the roll of and suitable engineering judgments to avoid mak-
bridges becomes an essential component for the phys- ing an inappropriate decision because of uncertainties
ical distribution and the transportation, the functional included in the cost, condition, and reliability models.
requirement of modern BMS has been changed also The optimal solution depends on the condition and
(Frangopol et al. 2001). In these days, the concept safety measures in addition to life-cycle cost of man-
of the lifecycle cost (LCC) has become wide spread agement scenarios. However, currently, each BMS
and the long term performance of bridges has been system developed by independent authorities uses dif-
considered to assess the value of bridges as a civil ferent methods resulting in loss of opportunity of
infra- asset. Therefore, advanced BMSs have to have communication for saving funds and man powers.
a decision making function to help bridge owners to Therefore, it might be worth of trying to prepare inter-
obtain the most effective budget distribution and cor- national fundamental standard and guideline of bridge
responding optimum management scenario (Kong and management by an international society.
Frangopol 2003).
The reliability and usability of the bridge manage-
ment system significantly depends on the fundamental REFERENCES
data such as costs of inspection, maintenance, and Cho, H-N., Frangopol, D. M. & Ang, A. H-S. (eds) 2007.
failure; condition profiles based on inspection data; Life-Cycle Cost and Performance of Civil Infrastructure
and reliability profiles considering the deterioration Systems, Taylor and Francis.
model. In this paper, trend of BMSs has been dis- Frangopol, D.M., Kong, J.S., and Gharaibeh, E.S. (2001).
cussed based on some basic features of BMS such Reliability-Based Life-Cycle Management of Highway
as Direction of BMS, Bridge Inventory and Inspec- Bridges, Journal of Computing in Civil Engineering,
tion, Inspection Information, Condition and Safety ASCE, 15(1), pp. 2734.
Measure, Deterioration Prediction Model, Life-Cycle Kong, J. S. and Frangopol, D. M. (2003) Life-cycle reliability-
based maintenance cost optimization of deteriorating
Performance and Cost Evaluation, Optimum mainte- structures with emphasis on bridges, Journal of Structural
nance scenario, Optimization in system level. Except Engineering, ASCE, 129(6), 818828.
these there are more issues to be considered for devel- Liu, M. and Frangopol, D. M. (2004), Optimal bridge main-
oping modern BMS such as network level optimization tenance planning based on probabilistic performance
(Liu and Frangopol, 2004). prediction, Engineering Structures, 26(7), pp. 9911002.
230
The bridge management system of the NYCDOT
B. Kroely
Advitam, Sterling, VA, USA
K. McAnulty
NYCDOT Bridge Department, New York, NY, USA
J. Daza
AECOM-Earthtech, New York, NY, USA
ABSTRACT 3. An offline system designed to work with dis-

connected laptops and PDAs during the field
The New York City Bridge Inspection/Research and inspections.
Development Unit (BIR&D) is tasked with inspec-
Using these 3 components, BDS is able to perform
tions and management of approximately 800 of the
all of the daily tasks typically associated with bridge
more than 2100 bridges in NewYork City. The BIR&D
management.
unit is continually striving to improve inspection
Beta testing of version 1.0 of the application com-
methods in order to meet their own, internal require-
menced in July 2006 with a small group of Key Users
ments as well as those imposed by the NY State
within the Inspection Unit. Full deployment to all of
Bridge Data Management System, and the National
the inspection staff was initiated at the end of July
Bridge Inspection Standards of the Federal Highway
2006. Since its initiation the unit has completed 65
Administration.
biennial inspections, 94 emergency inspections and
In 2003 members of the BIR&D unit attended
490 monitoring inspections. BDS version 1.6, the
a meeting where Advitam was demonstrating the
present version installed at the DOT, addresses 55 of
functionality of its ScanPrint Inspection Software.
the 57 functional requirements identified during the
Shortly thereafter an inquiry was made as to whether
initial stages of the application development, and in
or not the software could be customized to fit the needs
addition addresses 15 more that were identified during
of the Bridge Inspection Unit. The application devel-
the project. A key feature added during the project was
opment process began in October 2004 with two Joint
BDS-Web, an interface that allows access to bridge
Application Design (JAD) sessions to identify short-
information and inspection data without having the
comings in existing systems and desired features of
application installed on your PC.
the new software.
Through the customization of an existing bridge
As a result of the JAD sessions and other interviews
inspection and management tool, the NYCDOT has
with the user community regarding past experiences
been able to develop a comprehensive Bridge Data
with NYCDOT applications, a functional require-
System suited specifically to their needs. The uti-
ments document was created to list the specific tasks
lization of current information and data warehousing
expected of the BDS system.
technologies provides an excellent platform for the
The BDS is an integrated solution consisting of 3
development of future upgrades to address advance-
main components:
ments in Asset Management Techniques and Infor-
1. An extensive inventory which can be accessed by a mation Technology. Most importantly, the BDS will
standard application or via a web based portal, facilitate the inspection processes and provide valuable
2. A Windows based application to manage the inven- decision support information to assist the Depart-
tory, the inspection process and perform software ment in managing its inventory of critical and aging
administration, infrastructure.
231
Integrated bridge inspection and management software for New Jersey

Turnpike Authority
J.H. Laird
New Jersey Turnpike Authority, Woodbridge, NJ
J.A. Paul
LS Engineering Associates Corporation, Montville, NJ
J.K. Shaffer
InspectTech, Pittsburgh, PA
ABSTRACT
The New Jersey Turnpike Authority is one of the

United States largest toll-road systems and operates
both the New Jersey Turnpike and the Garden State
Parkway. In the past these two major roadways oper-
ated as separate entities and had developed different
bridge inspection programs for the nearly 1,000 com-
bined structures (over 20 in length). The Authoritys
bridge inspection program is designed to meet the
needs of a variety of critical stakeholders. The first and
foremost priority is ensuring the safety of the users of
the roadway. The inspection data is also used internally
by the Maintenance Department to plan repairs and by
the Engineering Department for capital improvements Figure 1. The New Jersey Turnpike varies from two 2-lane
roadways to six 3-lane roadways by Newark Airport, with
and overall management. Externally, information is dual-dual roadways over a 32 mile length through the busiest
provided to NJDOT and FHWA. zone.
A number of different consultants are engaged by
the Authority to perform bridge inspections each year.
Organizing all of this information into paper and com- REFERENCES
puter formats for the various stakeholders has proven
to be a difficult challenge. The New Jersey Turnpike Federal Highway Administration., 1995. Recording and Cod-
Authority has recently completed a program to stan- ing Guide for Structure Inventory and Appraisal of the
dardize the format of the reports and information for Nations Bridges, December 1995, FHWA-PD-96-001,
and Errata Sheet, March 2004, Washington, D.C.
all of its structures (bridges, culverts, ancillary). This is
Federal Highway Administration (FHWA), Bridge Inspec-
being accomplished by implementing a computerized tors Reference Manual, December 2006, Washington,
inspection and management program that is capable of D.C.
handling the wide variety of structure types, various New Jersey Turnpike Authority (NJTA), Bridge Inspection
consultants, and output needs. This paper will exam- Reports and Program Summary Reports for New Jersey
ine a brief history of the Turnpike and Parkway and Turnpike and Garden State Parkway Bridges, 2007 and
their respective bridge inventories (over 1000 bridges) 2008, Woodbridge, NJ.
and inspection programs. It will present the challenges
faced as well as the software solution that is being
delivered to solve these issues and provide a greatly
enhanced inspection and management process.
232
Integration of bridge management systems (BMS) and pavement

management systems (PMS)
R.M. Ellis
Stantec Consulting Inc., Edmonton, Alberta, Canada
A. Cheetham & K. Hong

Stantec Consulting Inc., Kitchener, Ontario, Canada
P.D. Thompson
Consultant, Castle Rock, Colorado, USA
ABSTRACT This paper presents the findings of this research and

development project and a description of the prototype
Transportation agencies are interested in finding ways tool which integrates BMS and PMS in a single inter-
to better manage pavement and bridge assets together face. The tool was developed to use Stantecs Pavement
using an intelligent and integrated asset management Management System (HPMA) and Bridge Manage-
approach. More than simply storing asset informa- ment Systems (Thompson et al (2003), Ellis et al
tion and condition in management systems without (2008)), but can be used for any bridge management
providing analytical capabilities, or simply reporting system and any pavement management system.
information from separate management systems, the The Phase 1 prototype includes the ability to dis-
challenge is to find ways to better use inspection and play a variety of inventory and condition data on
testing information, recommended work, prioritiza- several types of Microsoft Vitual Earth maps. In
tion and timing of the integrated work program. Phase 2 (currently underway), an analysis module is
The problem of an integrated analysis is complex being developed based on multi-objective optimiza-
and few solutions exist. One of the earliest inte- tion methodology of the type described in NCHRP
grated MR&R management systems was the Kuwait Report 590 (Patidar et al (2007)). The final product
Infrastructure Maintenance Management System will also be able to display integrated results on the
(Al-Kulaib et al (1997)). Developed for the Kuwait mapping interface as well in tabular and report form.
Ministry of Public Works between 199799, the sys- It is hoped that this research will further advance
tem allows the Ministry to manage their capital concepts of integrated asset management, and bene-
and maintenance planning functions more cost effec- fit agencies who are already implementing integrated
tively for roadway, bridge, storm and sanitary sewer asset management solutions, or who are considering
networks, as well as right-of-way features such as similar projects.
traffic signs, pavement markings, traffic signals and
sidewalks.
Technology has evolved in BMS, PMS, and infor- REFERENCES
mation technology to the extent that many new tools
Al-Kulaib, A., Baumgardner, R., 1997. Kuwait Infrastructure
are available. Computing speeds, network and internet Maintenance Management System (KIMMS). Proceed-
access speeds, and storage availability has made cer- ings, Quatar GIS/GPS Conference 97.
tain processes viable that were previously marginal. Ellis, R.M., Thompson, P.D., Gagnon, R. & Richard, G.,
Yet there has been little progress in the development 2008. Design and Implementation of a New Bridge Man-
of integration tools for these two major transportation agement System for the Quebec Ministry of Transport.
assets and very little progress in integrated priori- Transportation Research Circular E-C128: 7786.
tization. In 2007, Stantec embarked on a research Patidar,V., Labi, S., Sinha, K.,Thompson, P.D., 2007. NCHRP
project to develop a concept design for the integra- Report 590: Multi-Objective Optimization for Bridge
Management Systems. Washington D.C.: Transportation
tion of inventory, condition, and analysis results from
Research Board of the National Academies.
bridge management and pavement management sys- Thompson, P.D., Ellis, R.M., Merlo, T. & Hong, K., 2003.
tems. This included a new integration tool and new Implementation of the Ontario Bridge Management
analysis integration optimization tool. The resulting System. 9th International Bridge Management Confer-
tool is a web-based application with GIS interface ence 2003. Washington, D.C.: TRB, National Research
which can display all information and analysis results Council.
for integrated assets.
233
Development of a bridge network life cycle cost model
T. Reale & A. OConnor

Trinity College Dublin, Dublin, Ireland
L. Duffy & A. Daly

National Roads Authority, Dublin, Ireland
ABSTRACT launched a research project with the primary objective

of developing a bridge network lifecycle cost model
As bridges age and volumes of traffic continue to whereby The model would be developed in conjunc-
increase unabatedly, bridge maintenance has become tion with the NRAs EIRSPAN Bridge Management
a considerable challenge for governments around the System.
world who are managing ageing infrastructure. A Consequently the developed procedures have a
Bridge Management System (BMS) is defined as a Markovian Basis and will achieve the specific objec-
rational and systematic approach to organising and tives of (i) incorporating deterioration parameters from
carrying out all the activities related to managing a net- the full range of materials including, Reinforced Con-
work of bridges. Two of the main objectives of a BMS crete, steel and masonry structures, (ii) considering
are the prediction of the future state of structures, and bridges of all ages and exposure conditions as well
the prioritisation of maintenance and rehabilitation. It as other parameters specific to Irish conditions, (iii)
is used to optimise the inspection, maintenance and identifying budget restrictions going forward as well
rehabilitation of a network of structures with respect as highlighting the future financial impact of invest-
to minimising the total cost, and achieve the best ment restrictions (i.e. % increase in structures of a
performance from the network of structures. particular material, age etc. deteriorating to lower con-
Ireland has over 20,000 bridges on its road net- dition), (iv) highlighting the future financial impact of
work with over 2,000 of these on national roads. various investment options (i.e. % increase in struc-
This stock comprises a range of structural forms, tures of particular material, age etc. deteriorating to
located in different environments and with varying lower condition rating as a consequence of alterna-
ages. In 2001, the EIRSPAN Bridge Management tive investment options for repair) and (v) ultimately
System was introduced to coordinate and integrate providing a consistent basis for improving the way in
activities such as bridge inspection, repairs and reha- which Irelands bridge network is managed.
bilitation work in order to ensure optimal management The principal tasks to be completed in achieving
of this national road structure stock. The system these objectives have been identified as (1) the devel-
includes the essential components of a BMS, consider- opment of Markov transition matrices for different
ing the inter-related activities of inventory gathering, materials at global structural level, (ii) development
principal inspections, routine maintenance and reha- of non-stationary transition matrices to model cor-
bilitation works required for bridges. To date however, relation between components within structures, (iii)
the system has been primarily used as an inventory incorporation of assessment and repair performance
database and prioritisation decisions on maintenance uncertainties, (iv) allocating maintenance costs to the
and repair have been taken on the basis of visual condition ratings assigned within the EIRSPAN sys-
defects highlighted in the Principal Inspection report tem and (v) Bayesian updating of condition ratings
using a simple ranking system which identifies the based upon assessments logged within the EIRSPAN
bridges with the worst condition rating carrying the system.
highest traffic volumes. As such, whole-life cost- The development of this model then is the basis
ing has been available on a bridge specific level of this paper. It is calibrated specifically for the Irish
but not as yet on a network level. To this end, in network in conjunction with the information provided
2009 the National Roads Authority of Ireland (NRA) in the EIRSPAN database.
234
Implementation of a multi-period bridge investment optimization approach

utilizing Pontis results and additional constraints in three agencies
W. Robert & D. Gurenich

Cambridge Systematics, Inc. Cambridge, Massachusetts, USA
R. Thompson
Virginia Department of Transportation Richmond, Virginia, USA
ABSTRACT constraints specified by funding period and work type.

Each alternative has assigned to it a benefit or utility
One use for a bridge management system (BMS) is and set of costs by work type and period. Conceptu-
to optimize allocation of funds and predict future ally, the alternative could be a single project or life
conditions given a set of funding constraints. The cycle profile that includes a set of actions over time.
Virginia Department of Transportation (VDOT), It is assumed that one and only one alternative can be
Oklahoma Department of Transportation (ODOT) and selected per bridge for the analysis period.
Hawaii Department of Transportation (HDOT) all use Solving the optimization entails performing a series
the American Association of State Highway and Trans- of Pontis simulation runs as a one-time or occasional
portation Officials (AASHTO) Pontis BMS for this step, and then performing a secondary optimization
purpose. In recent years staff in each of these agencies with work type constraints. The preliminary Pon-
have invested considerable time refining the models tis simulations are used to specify alternative life
in Pontis to improve their ability to obtain realistic cycle profiles for each bridge in a network, including
results from the system. A critical limitation they have a do nothing for each bridge. Once the simula-
encountered in using Pontis is that the system offers no tions are completed, statistics are compiled for each
functionality for specifying the budget by work type bridge, including the costs, benefits and predicted per-
when performing a program simulation. Instead, when formance for each candidate life cycle profile (also
one runs a program simulation in Pontis, one specifies referred to as alternatives or project alternatives in
an overall budget constraint, and the system attempts this paper). The optimization step selects the set of
to optimize the work recommended in any one year life-cycle profiles that maximizes agency and user
without regard to work type. The lack of functionality benefits, subject to budget constraints for each year
for specifying budget constraints by type of work has and work type. Two systems have been developed to
been perceived by some to be a significant shortcom- support the approach described here: the VDOT Pon-
ing in the system that limits the value of the systems tis Robot for automating Pontis scenario runs, and
results. the VDOT Post-Pontis Optimizer for performing the
This paper describes an approach developed by optimization. Following initial implementation of the
the authors for using results generated by Pontis to modeling approach for VDOT described previously,
optimize bridge investments over a ten-year period subsequent changes detailed in the paper were made
considering budget constraints by year for each of five based on testing performed for ODOT and HDOT.
work types. The approach was originally developed
for VDOT to support VDOTs approach to bridge pro-
gramming, which involves determining the allocation REFERENCES
of funds by district with funding constraints by work
Cambridge Systematics, Inc. 2006, Pontis Release 5 Techni-
type.The resulting allocation is used to determine what
cal Manual. Washington, D.C.: AASHTO.
bridges on which to perform work, given work on a Robert, W., Gurenich, D. & Thompson, R. 2009. Multi-Period
bridge typically is performed no more than once every Bridge Investment Optimization Utilizing Pontis Results
ten years. Formulated as an optimization problem, the and Budget Constraints by Work Type. Presentation at
objective of the problem is to select the set of alterna- the 88th Annual Meeting of the Transportation Research
tives that maximizes benefit subject to a set of budget Board, Washington, D.C.
235
BrIM based bridge operations and management
A.M. Shirole
S & A Shirole, Inc., Minneapolis, Minnesota, USA
S.S. Chen, Q. Gao, & H. Hu

University at Buffalo, State University of New York, Buffalo, New York, USA
ABSTRACT Modeling (acronym BrIM), first introduced in 2005

by co-authors Shirole and Chen, facilitates an inte-
The design, construction, subsequent service life oper- grated approach to managing all phases of bridge
ations and management phases in a bridge life cycle life. This paper describes and illustrates just such
have become increasingly dependent upon informa- an approach to demonstrate how recently developed
tion management. Information to be managed is of software linkages to existing commercial software
various types: engineering analysis and design, meth- utilize a BrIM Data Pool to accomplish a variety
ods/techniques of construction, contracting methods of operational activities, such as load rating, routing
used for project delivery, day-to-day operational prac- and permitting. It also demonstrates the leveraging of
tices such as inspections, load rating, permitting this Data Pool for project and network level bridge
and routing, project and network level planning and inspection, reporting, and planning and programming
programming to meet financial, personnel, equip- of maintenance, repair, rehabilitation and replacement
ment, and contractual needs. Bridge Information activities.
236
Updating bridge deterioration models with irregular inspection intervals
P.D. Thompson
R.M. Ellis
K. Hong
Stantec Consulting Inc., Kitchener, Ontario, Canada
ABSTRACT adds complexity to the analysis. With the longer transi-

tion periods used in the Canadian systems (Thompson
Many transportation agencies are using an element et al. 2003, Ellis et al. 2008), these extra steps have
level condition state inspection methodology to inspect not been necessary, making the model outputs simpler
their bridges. The Provinces of Ontario, Nova Scotia, and easier to understand, and especially suitable for
Prince Edward Island, Manitoba, and Saskatchewan, smaller agencies such as municipalities wanting a less
and many municipalities, have adopted the Ontario elaborate solution.
Structure Inspection Manual (OSIM), while Qubec The analysis module, referred to as the Strategic
has implemented a similar approach. The methodol- Planning Module or MPS includes a module to per-
ogy requires the inspector to quantitatively record the form the regression analysis and Bayesian updating
severity and extent of defects, appropriate to the mate- process to generate new deterioration models.An inno-
rial and the specific component of the bridge. This is vation in the procedure is that it takes advantage of
as opposed to methods in which the inspector assigns a irregular inspection intervals from one to five years,
numeric rating (e.g. 0 to 9) to represent the status of an rather than being tied to a fixed cycle length. The
entire element or component of the structure and thus method takes advantage of the structure of Markovian
masks the quantities and types of defects occurring in transition probability matrices to arrive at a closed-
the element. form solution that is readily implemented in software.
Like many element-based bridge management sys- This improves the utilization rate of inspection records.
tems, Qubecs Strategic Planning Module (MPS) uses The model was demonstrated in a user-friendly
Markovian deterioration models to forecast future ele- Excel spreadsheet, which has been useful for train-
ment condition state distributions. A benefit of these ing purposes. It was also built into the MPS soft-
models is the ability to update them from histor- ware for routine maintenance of the Qubec bridge
ical condition data without a long time series of deterioration models over time.
inspections.
The 5-year period of the Canadian deterioration
model is unusual in bridge management systems but REFERENCES
has some distinct advantages. In a 1-year model, there
is a small but non-zero probability of each transi- Ellis, R.M., Thompson, P.D., Gagnon, R. & Richard, G.,
2008. Design and Implementation of a New Bridge Man-
tion each year. With four condition states, there is a
agement System for the Quebec Ministry of Transport.
small probability of the worst condition state after just Transportation Research Circular E-C128: 7786.
four years, something that would be considered unre- Thompson, P.D., Ellis, R.M., Merlo, T. & Hong, K.,
alistic in practice. Because of the small fraction that 2003. Implementation of the Ontario Bridge Management
quickly arrives in the worst state, bridge management System. 9th International Bridge Management Confer-
systems have to take steps to suppress small needs ence 2003. Washington, D.C.: TRB, National Research
or to postpone the more advanced transitions. This Council.
237
Estimation of enhanced Pontis deterioration models in Florida
P.D. Thompson
J.O. Sobanjo
Florida State University, Tallahassee, Florida, USA
ABSTRACT
The Florida Department of Transportation (FDOT) has

conducted a series of projects to enhance its implemen-
tation of the Pontis Bridge Management System, which
is developed by the American Association of State
Highway and Transportation Officials (AASHTO).
These enhancements have included a Project Level
Analysis Tool to present graphically the scoping and
timing alternatives on a bridge; and a network analy-
sis tool to analyze funding vs performance tradeoffs.
These tools have proved to be a valuable platform
for implementing new improved planning models that
cant be built into Pontis itself, but operate on Pontis
data.
Taking advantage of 14 years (19952009) of
history with 884,678 individual element inspection
records, the agency has amassed sufficient data to Figure 1. Comparison of selected element types.
develop statistically sound deterioration models for its
entire bridge inventory, including specialized elements that can describe the slower deterioration rates typical
for non-bridge structures, such as sign structures and of new elements. This same model form is planned for
retaining walls, and moveable bridge equipment. inclusion in the next major release of Pontis in 2012.
A problem noted in previous research is that the Figure 1 illustrates some of the results of the
Markovian models used in Pontis have fairly rapid ini- research. It was found that decks and expansion joints
tial deterioration. This creates a serious problem for do behave in a manner consistent with the Markovian
multi-year programming models, because it is diffi- model, without a statistically-verifiable age depen-
cult to configure such models to maintain a realistically dency. However, superstructure and substructure ele-
high network condition level. ments evidenced delayed onset of deterioration that
In order to address this problem, Florida in 2009 could be described by a Weibull model.
developed an enhanced form of the Markovian analy- Further research is underway to improve the screen-
sis that features a Weibull survival probability model ing of data points for maintenance activity, and to apply
for the initial transition from the best to second-best the new models in Floridas decision support tools to
condition states. This model has a shaping parameter see how they affect the outputs.
238
Advanced dynamic testing and bridge management system
J. Zwolski, J. Bien & P. Rawa

Wrocaw University of Technology, Wrocaw, Poland
ABSTRACT
Transportation network of each developed country

is one of the most exploited and lively systems.
Bridges are the most critical points in this network
and due to their peculiar structure and extensive use
they are exposed to accelerated deterioration. Results
of dynamic bridge tests offer information important
for the management process covering areas such as
damage detection, data for fatigue analysis and for
life-cycle forecasting as well as for long-term bridge
structure health monitoring.
The purpose of this paper is an overview of
advanced dynamic tests of bridges as well as discus-
sion of their applications in the Bridge Management
Systems. Main attention is paid to various techniques Figure 1. Rotational eccentric mass exciter applied in
of structure excitation: tests of vibration excited by bridge tests (Zwolski 2007).
normal traffic, special vehicles, vibration exciters or
free vibration tests as well as to advanced testing
equipment and efficient data processing procedures. Considered technologies of the bridge dynamic
Data of bridge structures required for use in Bridge tests are illustrated by example of practical applica-
Management Systems (BMS) can be collected during tion for identification of dynamic parameters of the
bridge dynamic tests which are focused on two areas: structure and damage detection. Presented tests were
parameters of the vibrating bridge structure itself performed on a steel railway bridge with damage
(modal tests) and vibration parameters of the struc- caused by dismantling of riveted connections between
ture subjected to a specific load (operational tests). the girder stiffeners. The structure was modally tested
The two philosophies of testing are different in terms by means of the rotational eccentric mass exciter
of applied methods of vibration excitation, data pro- (Bien et al. 2004) presented in Figure 1. For dam-
cessing techniques and field of results applications age detection, based on analysis of dynamic data,
in advanced BMS. General classification of dynamic the tool called UNCOMAC was applied (Zwolski
tests of bridges, taking into account excitation meth- 2007).
ods, analyzed parameters and range of application in
BMS, is presented in the paper.
Dynamic parameters of the bridge structure can REFERENCES
be identified on the basis of a single test as well as
can be collected either by permanently installed mea- Bien, J., Krzyzanowski, J., Rawa, P., Zwolski, J. 2004.
suring systems or by systematically performed tests. Dynamic Load Tests in Bridge Management, Archives of
Conception of the bridge condition monitoring system Civil and Mechanical Engineering, IV(2): 6378.
based on structure modal parameters changes, pro- Zwolski, J. 2007. Identification of Bridge Structures Modal
Parameters Applying Exciters; PhD Thesis. Wrocaw
posed by Zwolski (2007), is presented and discussed
University of Technology, Wrocaw, Poland.
in the paper.
239
MS8: Uncertainty in bridge damageability modelling
Organizer: M. Shinozuka
Failure estimation of highway bridges under combined effects of scouring

and earthquake

Department of Civil and Environmental Engineering, The Henry Samueli School of Engineering,
University of California, Irvine, Irvine, USA
ABSTRACT
Bridge scour is one of the most common causes of

bridge failure in the United States (HEC-18 2001).
In order to prevent the failure of bridges spanning
waterways, it is essential to consider the effects of
scour on the structural performance of bridge and to
design the bridge so that it can resist any structural
degradation due to scour in addition to other natural
hazards. According to AASHTO LRFD (2007), scour
is not considered as a load parameter, but its effects
on changing the structural properties of bridge should
be evaluated to make sure that the structure is not
vulnerable to extreme loads, such as wind and earth-
quake. Towards this goal, the current paper studies the
behavior of reinforced concrete (RC) highway bridges
under the combined effects of scouring and earth- Figure 1. Developed risk curves for the two- and three-span
quake. bridges under different scour depths.
Scour is defined as the water-induced erosion of soil
around the foundation of bridge and can be identified
mainly in three forms: 1) long-term aggradation and
degradation of the river bed due to erosion and depo-
sition of material, 2) general scour which may result
from contraction of the flow, and 3) local scour caused
by an acceleration of water flow at the piers or abut-
ments (HEC-18 2001). In the latter form, the presence
of obstruction to the flow increases the speed of stream
and creates vortices along the horizontal and vertical
axes which result in higher scouring depths.Among the
mentioned three forms of scour, it has been proved that
the local scour is generally the most significant form.
Hence, this paper focuses on the local scour and studies
the uncertainties associated with its modeling around
the bridge pier. Then, a set of two- and three-span RC
bridges are selected to assess the extent of structural Figure 2. Developed fragility curves for the three-span
degradation as a function of scour depth. The seismic bridge under different scour depths (moderate damage).
performance of case study bridges are evaluated at the
next step using fragility parameters and effects of local multi-hazard framework which can be used predict the
scouring on the response of RC bridges are studied probability of occurrence of various seismic damage
during nonlinear time-history analysis. This provides a states for a bridge under scouring effects.
243
Deterministic and probabilistic evaluation of time to corrosion initiation

for bridges located in coastal areas

Department of Civil and Environmental Engineering, The Henry Samueli School of Engineering,
University of California, Irvine, Irvine, California, USA
ABSTRACT
A Chloride-induced corrosion in reinforced concrete

(RC) bridges is a mechanism caused by the intrusion of
chloride ions into concrete. This mode of corrosion is
more probable when RC bridges are located in coastal
regions and exposed to aggressive environmental con-
ditions. Because of the penetration of chloride ions
in structural members, the chloride content of con-
crete gradually increases and when the concentration
of chloride ions in the pore solution on the vicin-
ity of reinforcing bar reaches a threshold value, the
chloride-induced corrosion initiates.
Chloride transport mechanism in concrete is a com-
plex phenomenon that may occur in several forms,
such as ionic diffusion, capillary suction, and per-
meation. The rate of this mechanism depends on the Figure 1. Cumulative distribution function of corrosion
characteristics of concrete, degree of pore saturation, initiation time for three concrete depths.
chloride binding capacity, free chloride content, and
exposure conditions. By increasing the duration of binding capacity, are considered to calculate the chlo-
time through which a bridge is exposed to aggressive ride content in different time intervals. This leads to a
conditions, the deterioration process of reinforcing deterministic approach for the estimation of corrosion
bars can get relatively fast. This results in cracking initiation.
or spalling of RC members and may lead to severe The developed computational framework is then
reduction in serviceability of bridges. improved by taking into account the effects of uncer-
The time between corrosion initiation and service- tainties of influential parameters in the deterioration
ability failure is usually smaller than the required time process. Through a probabilistic approach, the proba-
for corrosion to initiate. Therefore, the realistic estima- bility distribution function of corrosion initiation time
tion of corrosion initiation time has a significant role is determined and the results are compared with the
in the accurate performance prediction of RC bridges estimated initiation time obtained from the determin-
over the time. In the present paper, an integrated istic approach. Finally, the structural performance of
computational framework is proposed to simulate the a set of short-, medium-, and long-span bridges are
penetration of chloride ions into concrete. Towards this evaluated using pushover analysis over a life cycle of
goal, the effects of various parameters, such as water 30 years and the effects of variation in corrosion ini-
to cement ratio, ambient temperature, relative humid- tiation time on the prediction of remained structural
ity, age of concrete, free chloride content, and chloride capacity are studied.
244
Uncertainty in bridge fragility curves and its effect on the seismic

risk evaluation of a highway transportation network
S. Banerjee
Pennsylvania State University, University Park, PA, USA
M. Shinozuka
University of California, Irvine, CA, USA
ABSTRACT involved in the bridge fragility curves is extremely

important for reliable evaluation of seismic risk of
Seismic vulnerability of bridges, commonly expressed highway networks.
in the form of fragility curve, is a key component of
seismic performance evaluation of a highway trans-
portation network. Two-parameter lognormal distribu- REFERENCES
tion functions are generally used to represent fragility Banerjee, S. & Shinozuka, M. 2008. Mechanistic quantifica-
curves. The fragility parameters are estimated using tion of RC bridge damage states under earthquake through
maximum likelihood method. However, these param- fragility analysis. Probabilistic Engineering Mechanics
eters may have some statistical variation, which in 23(1): 1222.
turn introduces uncertainties in the fragility curves. Caltrans. 1994a.The Northridge earthquake. Caltrans PEQIT
Such uncertainties may play a critical role when Report, California Department of Transportation, Div. of
the bridge seismic vulnerability model (in terms of Structures, Sacramento, CA.
fragility curves) is used for the performance evaluation Caltrans. 1994b. The supplementary bridge damage reports.
of highway networks under regional seismic hazards. Caltrans PEQIT Report, California Department of Trans-
portation, Div. of Structures, Sacramento, CA.
The present study evaluates the statistical uncertainty Chang, E.S., Shinozuka, M. & Moore, J. 2000. Probabilistic
associated with such a bridge seismic vulnerability earthquake scenarios: Extending risk analysis methodolo-
model. Monte Carlo simulations are performed to gies to spatially distributed systems. Earthquake Spectra
develop 90% confidence intervals (within 5% and 95% 16(3): 557572.
confidence levels) of bridge fragility curves. These Shinozuka, M., Feng, M.Q., Kim, H.-K., Uzawa, T. & Ueda,
confidence intervals are taken as a measure of the T. 2000. Statistical analysis of fragility curves. Journal of
uncertainty involved in the bridge vulnerability model. Engineering Mechanics, ASCE 126(12): 12241231.
This vulnerability model is used then in the seismic Shinozuka, M., Feng, M. Q., Kim, H., Uzawa, T. & Ueda,
performance evaluation of a regional highway trans- T. 2003a. Statistical analysis of fragility curves. Report
MCEER-03-0002, Multidisciplinary Center for Earth-
portation network. The network is analyzed under a quake Engineering Research (MCEER), The State Uni-
set of USGS hazard consistent scenario earthquakes. versity of New York at Buffalo, NY.
Monte Carlo simulations are performed to predict Shinozuka, M., Murachi, Y., Dong, X., Zhou, Y. &
the degraded performance of the network when all Orlikowski, M.J. 2003a. Effect of seismic retrofit of
constituent bridges are damaged under the scenario bridges on transportation networks. Earthquake Engineer-
earthquakes considered herein. The expected loss due ing and Engineering Vibration 2(2): 169180.
to network degradation is evaluated in terms of total Zhou, Y. & Shinozuka, M. 2006. Development of proba-
social cost. Seismic risk curves are developed that rep- bilistic scenario earthquakes for seismic risk analysis of
resent the annual exceedance probabilities of various spatially distributed systems. Proceedings, EERIs Eights
U.S. National Conference on Earthquake Engineering
levels of expected loss in the network under future (8NCEE), San Francisco, California, April 1822.
seismic events. Results indicate that the developed risk Zhou, Y., Banerjee, S. & Shinozuka, M. 2010. Socio-
curves have variation due to the use of bridge fragility economic effect of seismic retrofit of bridges for highway
curves with different confidence levels (e.g. 5%, 50% transportation networks: A pilot study. Structure and
and 95%). Hence, the quantification of uncertainty Infrastructure Engineering 6(12): 145157.
245
Sequential bridge fragility updating using long-term monitoring data
M. Shinozuka & M. Torbol

University of California, Irvine, USA
S. Banerjee
The Pennsylvania State University, USA
ABSTRACT
The present study evaluates the performance of a real-

life bridge under a set of earthquake ground motions.
A three-span continuous cast-in-place pre-stressed
post-tensioned box-girder bridge (known as Jam-
boree Road Overcrossing (JRO)), located in Orange
County, CA, is considered for this purpose. The bridge
is instrumented by California Department of Trans-
portation (Caltrans) with 13 servo-type accelerometer.
These sensors are recording bridge vibration response
under ambient and traffic loading conditions over past
seven years. Additionally, the bridge response under
several earthquakes was also recorded.
In this study, the natural frequencies of the Jamboree
Bridge are calculated from the database containing
recorded bridge response. Following the flow-chart in
figure 1 a finite element (FE) model of the bridge is Figure 2. Fragility curves using model updating.
developed using the computer code OpenSees and
including non-linear characteristic like plastic hinge, History Analysis is performed after each model updat-
gap element, bearing, piles and soil modeling. The ing and the damageability of the bridge, in term of
model is initially built using the design data and suc- fragility curves is updated.The comparison of the dam-
cessively updated using the long-term monitoring data ageability of the bridge over the time is studied, in
from the installed sensor. A dataset of Inelastic Time figure 2 the comparison of fragility curves obtained
from the initial model built from the design data and
the updated model using field data is made. The out-
come of this research provides useful insights for
analyzing RC bridges under seismic ground motions.
REFERENCES
Hossein Tahghighi, Kazuo Konagai, Numerical analysis of
nonlinear soil-pile group interaction under lateral loads,
Soil Dynamics and Earthquake Engineering 27 (2007).
Shinozuka, M., Banerjee, S., Kim, S.-H. (2007 Statistical
and mechanistic fragility analysis of concrete bridges,
MCEER, State University of New York at Buffalo.
Soyoz, S., Feng, M.Q. (2008) Long-Term Monitoring
and Identification of Bridge Structural Parameters,
Computer-Aided Civil and Infrastructure Engineering,
Figure 1. Flow chart of the study. 24(2009), 8289.
246
MS9: Performance-based asset & risk management of the
highway infrastructure system
Organizer: A.E. Aktan
Mitigating infrastructure performance failures through risk-based

asset management
A.E. Aktan & F.L. Moon

Drexel Intelligent Infrastructure Institute, Drexel University, Philadelphia, PA, USA
ABSTRACT is the need to include both psychological and cul-

tural weighing factors within the risk formulation.
There is a pressing need for a more conceptual and This allows both the public perception related to the
generic definition of risk in the context of infrastruc- objectional nature of the specific risk and the cul-
ture performance. The objective of this paper is to pro- tural expectations for infrastructures to be explicitly
pose such an expanded definition of risk appropriate defined and considered. The paper concludes with a
for underpinning infrastructure decisions. This paper description of an eight step procedure for construct-
presents a brief history of risk and discusses the curing an integrated asset management system, based on
rent limitations of convention risk definitions related to the proposed risk definition, to effectively underpin
infrastructure decision-making. Of particular interest infrastructure decision-making.
249
One Team bridge maintenance strategy MACs in the UK
A.G. Ardrey
Halcrow Inc., New York, USA
B. Stanforth
A-One Plus, England, UK
ABSTRACT construction, network operations, design and imple-

mentation of processes.
The UKs strategic road network is managed through The current proposed US Transportation Bill is
a series of innovative contracts called MAC (Man- expected to create a greater focus on performance
aged Agent Contractor) contracts whereby a fully and to include funding mechanism changes. The cur-
integrated, team consisting of a maintenance contrac- rent US State budget shortages an increasing need
tor and consultant, effectively take on the management to improve the management of existing infrastruc-
of large sections of highway including bridge structure are strong drivers for change that may provide
tures. The team plans, manages and executes routine, an opportunity to re-think bridge maintenance con-
preventative and rehabilitation works (to a defined tracting. An option is to bundle existing contracts and
value limit). Driven by government policy and inde- create longer term performance based contracts con-
pendent reports recommending an industry re-think, necting bridge management policy with maintenance
the MAC contracts have been developed over a long activity to gain the efficiencies and innovations neces-
period of time and have resulted in interesting devel- sary for the future. The MAC contract offers a proven
opments in the bridge maintenance strategy adopted model that can be adapted to the US market.
for the network. A private sectors view of the bridge maintenance
With overarching strategic goals (safe, reliable jour- strategies adopted, the reasons for success and the
neys, informed travelers) and aligned incentives, the issues arising is also provided. The paper illustrates
out-come-based, performance contracts and specifi- how bridge maintenance activities are driven by pol-
cations have allowed the private sector to innovate icy and the effects of strategic goal alignment and
and work with the UKs Highways Agency to develop risk-based approaches. It also draws parallels with
a bridge management and maintenance process that current bridge maintenance outsourcing activities and
is considered cost-effective and sustainable. The pro- maintenance strategies in the US.
cess includes a risk-based approach that guides the
maintenance activities conducted and pro-actively REFERENCES
encourages a culture of continuous improvement.
All aspects of service delivery are based on the National Audit Office. 2009. Report by the Comptroller and
Highway Agencys strategic goals safe, reliable jour- Auditor General. HC 959. HA: Contracting for Highways
neys with in-formed travelers. The success of the MAC Maintenance. Session 20082009.
has required an intelligent and pro-active client. Suc- UK Highways Agency. BD79/06 Design Manual for Roads
cess has required partnership between the client and and Bridges The Management of sub-standard Highway
Structures.
the contractors who take ownership and pride in their
UK Highways Agency. November 2009. Delivering best
Network Areas with an optimal allocation of risk and value solutions and services.
focus on value for money. UK Highways Agency. Network Management Manual and
Future contracts will need to demonstrate and Routine and Winter Maintenance code.
actively manage sustainability issues. Continued effi- www.amotia.org Association for the Management and
ciencies are required to eliminate waste particularly in Operations of Transportation Infrastructure Assets.
250
TEAMS: Floridas Turnpike Enterprises answer to asset management
D.L. Blake
Floridas Turnpike Enterprise PBS&J, Ocoee, FL, USA
ABSTRACT customizable, user friendly and compatible with other

legacy systems currently in place. The asset man-
Floridas Turnpike Enterprise has an application called agement system gives Floridas Turnpike Enterprise
the Turnpike Enterprise Asset Management System, a highly efficient means of identifying, tracking, and
or TEAMS as it is commonly referred. The mission maintaining the performance history of right-of-way
of TEAMS is to provide an improved approach to assets, generates lists of detailed periodic maintenance
management and preservation of infrastructure assets, needs, and develops renewal and replacement budget
to protect bondholder investment, and sustain system forecasts at the users desktop.
performance. This paper focuses on how asset management prin-
TEAMS was developed as a web-based comput- ciples have been implemented at Floridas Turnpike
erized means to catalogue Floridas Turnpike Enter- Enterprise in the form of the TEAMS application. Dis-
prises assets, evaluate current conditions, predict time cussion will start with the history of Floridas Turnpike
to renewal and replacement, and develop appropriate Enterprise, lead to the early stages of asset manage-
periodic maintenance budgets. The system resides in ment at the agency, and then to the development of the
centralized servers and users access it via the Floridas asset management system.Afterwards the current state
Turnpike Enterprise Intranet using a graphical web of TEAMS and an analysis of its future at Floridas
browser interface. The system is comprehensive, Turnpike Enterprise will be discussed.
251
Optimal resource allocation for seismic retrofitting of bridges

in transportation networks
F. Carturan, C. Pellegrino, C. Modena, R. Rossi & M. Gastaldi

Dipartimento di Costruzioni e Trasporti, Universit degli Studi di Padova
ABSTRACT link damage state and how a link damage state can
influence its functionality, has been created a set of
In this study bridges fragility is evaluated considering damaged networks. In this set of damaged networks
the effect caused to the transportation network and to has been assigned the transportation demand and the
the economy of the region affected by the earthquake. fluxes have been calculated. The total network time
This is a multi-disciplinary approach since the pro- has been calculated in order to evaluate the network
cedure is composed by several modules: geological, delay that the damage causes. At the end of the pro-
structural, transportation, economic. The procedure is cess the network risk curve (probability of the seismic
based on a stock of bridges included in a database action vs. total delay) is derived. Some possible appli-
and starts from a set of possible seismic scenarios. cations of the procedure are the evaluation of the order
By using fragility curves of bridges the damage state of bridge retrofitting and the analysis of the emergency
of the network links, in which bridges are included, response. The procedure developed in general terms
can be obtained. By making a series of hypothesis has been applied to a test network near Treviso in the
on how a bridge damage state is correlated with a north-eastern part of Italy.
252
Inspection process of Bulgarian Bridges
A. Chaperon & V. Morisseau

Advitam, France
B. Kroely
Advitam Inc, Sterling, VA
The Road Administration of Bulgaria and Advitam set Software and trained teams to periodically inspect
up jointly, in 2003, a national scale solution to all these the bridges condition. The correct following of the
questions. Rehabilitation is not a matter of deciding inspection calendar defined in the Bulgarian norms
how much money to deploy, but shall/can fit to avail- is now facilitated.
able budgets. The project mainly stands on three steps: Software to help the decision about actions of main-
tenance. The systematic link between defects and
A global audit including regulations in effect as
respective actions to undertake can now be done
well as the infrastructure network. Practical meth-
automatically.
ods and techniques allowing it within limited time
Preliminary repair projects for 15 critical bridges.
and controlled budget from the beginning
These projects will serve as basis to launch repair
A global infrastructure network evaluation, using
works and can be used as reference in terms
objective indicators through synthesis tools
of actions and costs for future similar projects,
The setting and application of a progressive director
including:
scheme, from partial rehabilitation projects towards
Defect total quantities from detailed inspections,
a global preventive maintenance management.
which allowed to state directly on the necessary
To support this project, Advitam was in charge light repair budget for the 15 bridges
of developing a Bridge Management System (BMS) Recommendations for heavy repairs (replace-
based on its software suite: ScanPrint. From mid of ment of equipments as bearings or expansion
October to mid of December 2003, inspectors trav- joints, structural reinforcement, etc. . .)
eled around the country and gathered information and Recommendations for further investigations
pictures about the 1312 bridges. Advitam provided (durability tests, monitoring, etc. . .)
technical assistance in the various regions to finalize A global maintenance master plan on the 1312
the training. Bridges, stating when to re-inspect the bridges,
By the end of the project, the REA has been according to their technical and final condition rat-
provided with the following information system: ing, and to the criteria selection. A global decision
tree has been worked out with the REA in order to
Standard procedures for bridge inventory and
step towards a global preventive maintenance man-
inspection
agement plan of all bridges over 20 m in 10 years.
Database with information about all bridges. This
information may be accessed by many means and
be easily transmitted to local administrations:
ScanPrint interface
Excel tables resuming information
Web interface containing inspection results
253
Current state of highway transportation asset management in Germany
P. Haardt & J. Krieger

Federal Highway research Institute, Germany
ABSTRACT Cornerstones are a federal database, which covers all

the required information about the road network and
In Germany the federal road network carries the main management systems for pavements and structures.
load of the transit traffic by reason of its central posi- Whereas operative tasks are performed by the state
tion in Europe. Currently it contains about 53,000 km administrations, the federal Ministry of Transport has
of motorways and trunk roads including about 38,300 the significant mission to define specifications and to
bridges. supervise and control the implementation and achieve-
The main part of the network was built in the 60th ment of objectives. Aims of the Asset Management
up to the 80th of the last century. Over the years a very are:
high quality of construction methods was achieved.
a sustainable and efficient securing of substance and
Nevertheless maintenance needs today are in the range
function of the network
of about 1.7 Bio per year. Maintenance costs have
a reliable assessment of the development and exten-
to be spent in a way to obtain the greatest possible
sion needs and
benefit. This task is supported by an comprehensive
the efficient operation of the federal network
Asset Management.
Asset Management can be considered to be a sys- However, the complexity of these system continues
tematic process of maintaining, upgrading and oper- to increase and an holistic approach is necessary to
ating assets. It provides tools to facilitate a more understand the effect of technological, environmen-
organized approach to making the decisions neces- tal, economical, social and political interactions on
sary to achieve the publics expectations. In Germany life cycle aspects. In order to accomplish this, meth-
the Asset Management procedures include all the ods have to be developed to systematically analyze
processes, tools, data and policies necessary for the the whole lifecycle, and models have to be formulated
effective management of the federal road network. for evaluating and comparing the risks and benefits
In Germany the system is essentially influenced by associated with various alternatives on a probabilistic
the federal constitution. The Federal Government is the basis.
owner of the federal road network however the states
administer these roads under their own responsibility.
In the frame of the current German Asset Man-
REFERENCES
agement firstly the Federal Ministry of Transport is
supported with an overview of the current situation BMVBS 2001. Aufstellung koordinierter Erhaltungspro-
of the road network and with statements concerning gramme fr Straenbefestigungen, Bauwerke und son-
financial requirements as well as possible strategies stige Anlagenteile von Bundesfernstraen. Allgemeines
for realizing long-term objectives and fulfilling basic Rundschreiben Straenbau Nr. 26/2001. Bonn: unpubl.
conditions for maintenance routines. Secondly, the Haardt, P. & Stadler, A.. 2004. The German approach to
states and authorities are supplied with recommen- bridge management. Planning at network level. Proceed-
dations for performing improvements in compliance ings of 2nd Int. Conf. on Bridge Maintenance, Safety and
Management, Kyoto.
with strategies, long-term objectives, basic conditions
OECD 2000. Asset Management for the Roads Sector.
and budgetary restrictions. Organisation for Economic Co-operation and Develop-
The required consulting service includes for exam- ment, Directorate for Science, Technology and Indus-
ple network-related analysis and evaluation. For this try, Expert Group on Asset Management Systems,
purpose, a methodological basis was developed. DSTI/DOT/RTR/IM1.
254
Network level deterioration modeling: A case study on masonry

arch bridges
M. Imran Rafiq, S. Sathananthan & M.K. Chryssanthopoulos

University of Surrey, Guildford, Surrey, UK
ABSTRACT
Engineering systems comprise a series of intercon-

nected elements, which behave in a complex but
integrated manner to perform their intended function.
Being exposed to harsh environmental conditions, and
owing to their critical position within the transport
infrastructure, effective management of bridges is vital
for the reliability of overall transport system. Deterio- Figure 1. BBN model of network level deterioration for
ration of bridge elements and sub-elements may lead masonry arch bridges.
to a reduced level of service derived from the trans-
port network. Traditionally, element level management
approaches have been extrapolated in managing sys-
tem performance, especially in dealing with possible from the sample bridges and is shown in Fig. 1. The
changes due to wear and deterioration. overall condition of the network is predicted using the
Predicting deterioration is a vital component of any BBN. The figure shows that a majority of the elements
modern bridge management system. It is generally are in Fair condition, leading to a higher probability
established for individual bridge elements, through for the overall group being in the Fair condition.
inspection records, thus leading to an approximate A desirable feature of the BBN models is their abil-
prediction of bridge performance over time. Several ity to analyze what-if scenarios. This is particularly
methods are available in literature to account for such helpful for the prioritization of assessment and main-
elements vs system interactions. BBNs are capable of tenance intervention activities. This feature has been
handling complex relationships between the elements demonstrated though a sensitivity analysis to obtain
and the system by means of conditional probabili- the relative importance of each minor element on the
ties, and can also address the variation of performance overall bridge group condition.
with time and the updating of performance with new It is evident from the results that initial element
information, both important considerations in a bridge conditions are crucial in determining not only the
management context. network condition but also in understanding the main-
The development of a network level deterioration tenance/intervention priorities. Hence, for practical
model using Bayesian Belief Networks is presented in applications these should be obtained as accurately
this paper. Conditional probabilities between bridge as possible. This, in turn, leads to the question of
elements and initial element conditions are assumed on what inspection techniques should be used, how often
the basis the inspection methodology and inspection and with what priority. In answering those questions,
results from a sample group of UK masonry arch rail- the usefulness of the BBN model lies in its ability to
way bridges. A BBN model representing the network deal with a wide range of scenarios both readily and
level deterioration of masonry arch bridges is shown in transparently.
Fig. 1. Three condition states are considered for each The BBN model can be used to compare vari-
element in the BBN. Poor, Fair and Good states ous inspection techniques and to estimate appropriate
are represented by SCMI score range of 045, 4580 inspection intervals with the objective of maintaining
and 80100 respectively (see full paper for details). a consistent risk/reliability target within any particular
The probability of an element in these states is derived bridge network.
255
Road infrastructure asset management a holistic approach to road

infrastructure supply
J.M. Jansen
Danish Road Directorate, Operation & Maintenance Division
ABSTRACT sub-asset decisions to total asset decision-making,

which means that old habits has to be reconsid-
This paper gives a view on the concept of road infras- ered, and the traditional budget competition between
tructure asset management in context with the concept road network engineering disciplines will be replaced
of modern road administrations as corporate busi- by objective measurement in stakeholder surveys on
nesses, which mission is delivering of a contemporary perception of delivered asset management outcome.
transport network and traffic management to support Different stakeholder groups have their own percep-
societal prosperity and growth. tion of outcome value, and this is the true driver of
Management of the road infrastructure asset effective asset management decisions.
encompasses a holistic management of all road infras- This paper presents the complexity of asset man-
tructure elements in the corridor, both capital works agement for modern road administrations and give
on pavements, structures and installations and routine examples of how this process has been approached;
maintenance works on areas and equipment. in view of nobody has a complete asset management
Based on work in PIARC committee on Asset in operation.
Management and national research on asset manage- Major learning from asset management is to adjust
ment approach, this presentation will paint a pic- the choice of enablers for intervention to secure
ture of the multidiscipline complexity that has to be stakeholders expectations for perceived outcome.
accomplished to execute effective asset management.
Asset management is not only to provide best
practice maintenance, but also to deliver perceived REFERENCES
outcomes to stakeholders value. To do this, much
more knowledge than just engineering has to be incor- Jan M. Jansen, 2006. Managing Asset Allowance for Main-
tenance Investments. XXVI International Baltic Road
porated in the process of road infrastructure asset
Conference, Estonia
management. Jan M. Jansen, Sren Brnchenburg and Bjrn Lassen, 2008.
With introduction of road infrastructure asset man- Asset Management: Technical input to decision making,
agement also internal processes in the road adminis- Denmark National Report. World Road Congress, Paris
trations has to be reengineered, as the holistic approach Jan M. Jansen et al, 2008. Asset Management practice: TC
will change the basis of decision-making from 4.1.1 report. World Road Congress, Paris
256
A holistic approach to asset management in The Netherlands
L. Klatter, J. van der Velde & J. Bakker

Ministry of Transport, Public Works and Water Management, The Netherlands
ABSTRACT at the end of 2011. In this program four sub-programs

are defined:
The Directorate General for Public Works and Water
1. Reliable and accurate asset data
Management is responsible for the management of
2. Stable long term maintenance programs
the national roads and water infrastructures in the
3. Clear objectives and transparent requirements
Netherlands. This agency, further to be referred as
4. Transparent procurement procedures.
Rijkswaterstaat, is an executive agency within the
Dutch Ministry of Transport. Asset data management calls for a network oriented
Approximately 9.000 fte are working on 3.300 km approach. In this approach a decomposition of all
(2,000 miles) strategic highways, 1.700 km (1.000 object classes can be handled effectively.
miles) strategic waterways and 65.250 km2 (25.000 Development of stable long term maintenance pro-
square miles) water systems (such as dunes and surface grams (including prioritization) is another important
water). The annual budget for maintenance, opera- development at Rijkswaterstaat. The existing object
tions, renewal and expansions of those networks is class oriented programs contain elements for strategic
approximately 5 billion. and operational planning, but they were not integral
A comprehensive approach to infrastructure man- and not based on strategic goals. (Klatter & van der
agement needs integration in the decision making Vusse,2004).
process and therefore reliable asset data, a long term Clear objectives and transparent requirements are
performance based maintenance program, clear steer- important for justifying the maintenance programs.
ing objectives and transparent procurement strategies. Each hierarchal level in the decomposition needs
The introduction of asset management is used as the a specific description. Also requirements should be
integrating principle. The goal is: to deliver best ser- connected throughout the hierarchical levels. System
vice to the public at lowest life cycle cost, given public Engineering principles are used to connect all these
acceptable risk. requirements and to assure compliance within the
Within the asset management process three roles different levels.
can be distinguished: the asset owner, the asset man- The policy in the Netherlands is to outsource activi-
ager and the service provider (table 1). ties to private parties to an extend as large as possible.
Rijkswaterstaat decided to implement asset man- All maintenance and construction works are contracted
agement in 2008. The program should be completed out to private companies, contractors and engineering
firms. This means that 100% of the final output and
outcome is realized through these contracts. The asset
Table 1. The asset management roles. management program will be completed within two
years. The full implementation and management of the
Asset owner Asset manager Service provider system thereafter should not be underestimated. Today
de results of the program can already be seen in the
Responsibilities and tasks
organization, in a grown insight in the asset quality,
maintenance needs and effects on performance and
- Overall network - Investment - Project delivery
policy strategies - Maintenance risks.
- Targets for - Maintenance execution and
performance and concepts services
condition on a - Technological - asset data REFERENCES
network level standards management
- Target for - Risk management - project Klatter, H.E. & van der Vusse, J.J.,2004, Total Asset Manage-
acceptable risk - Network management ment in the Netherlands, 2nd International Conference
profile management on Bridge Maintenance, Safety and Management, Kyoto,
2004.
257
Towards sustainable life cycle design of highway bridges
Z. Lounis & L. Daigle

National Research Council, Ottawa, Canada
ABSTRACT The long-term efficiency of high performance con-

crete (HPC) containing supplementary cementing
Highway bridges provide a core public service and are materials (SCM) is evaluated and compared to that
needed for a very long time to support the sustain- of normal performance concrete (NPC) in terms of
able development of neighboring communities. They service life, life cycle cost, environmental impacts and
are critical links in Canadas transportation network social impacts. A limited number of performance indi-
that should be kept safe and functional during their cators are chosen to illustrate the proposed approach
service lives to enable personal mobility and trans- to sustainable design, which include:
port of goods to support the economy and ensure
Economy: life cycle agency costs and life cycle user
high quality of life. The average service lives of these
costs.
assets vary from 50 years up to 100 years, which
Environment: CO2 emissions and volume of con-
are continually extended by using different manage-
struction waste materials.
ment strategies that include different combinations
Social: Accident cost and traffic congestion.
of preservation, rehabilitation, strengthening, replace-
ment, mitigation and adaptation actions. The growing The use of HPC materials results in bridge decks
concerns for environmental protection and the shift with extended service lives, reduced life cycle costs,
towards achieving sustainable transportation infras- and better environmental and social profiles when
tructure are now requiring the use of approaches that compared to conventional normal concrete bridge
can achieve an adequate balance between social, eco- decks.
nomic and environmental performance over the entire In this particular example, all the performance mea-
life cycle of the bridge. sures used indicated that the HPC deck alternative
Relevant and practical performance measures are would be a better choice. There is no need to trade off
needed to support the decision-making at different lev- or reach an equitable balance between environment,
els of hierarchy and to ensure the social, economic and social and economical. Only on initial construction
environmental sustainability of bridges, as well as the cost would the NPC bridge deck option perform better,
sustainability of transportation and the communities but initial construction cost is not a suitable indicator
they serve. for sustainability.
This paper discusses some performance indicators, In life cycle performance analysis, the reliable esti-
such as safety, serviceability, costs, traffic disruption, mation of service life and associated schedule and
greenhouse gas emissions, which can be used for extent of MR&R activities is critical. Beyond their
life cycle design of highway bridges. An example, significant influence on economic performance, they
taken from the NorthAmerican context, illustrates how also drive the results of the social and environmental
different design and rehabilitation approaches can con- impacts analysis. For bridges with high level of traffic,
tribute to achieve the desirable balance between social, the social and environmental costs of frequent and/or
economic and environmental sustainability criteria. extended interventions should be taken into consider-
The example illustrates the benefits of implementation to move towards a sustainable approach for the
ing a life cycle-based design approach through the design and management of highway bridges.
construction of high performance concrete highway
bridge decks that could yield reduced environmental
footprints and lower social pressure.
258
Green stormwater infrastructure: A prototypical multi-domain, complex

infrastructure system
F.A. Montalto & A. Waldman

Drexel University, Philadelphia, PA, USA
ABSTRACT cars and trucks. Streets and highways can be habitat

corridors and infiltration parks, or be used to daylight
Runoff generated from impervious surfaces in urban- historic streams, or create new ones. We need to view
ized watersheds is an extensive liability for water buildings as more than simply places for us to live and
utilities which must meet increasingly stringent federal work. Buildings can also cool our cities, and reduce
regulations governing its discharge. Traditional grey our consumption of energy and natural resources.
infrastructure solutions to this problem are costly
because they involve expansion of existing collec-
tions systems and/or creation of in-line storage. A new
suite of green infrastructure (GI) approaches would
reduce the quantity of runoff by addressing its source,
while also addressing many urban sustainability goals.
However, to be effective, such approaches (1) would
need to be implemented widely in urban watersheds,
(2) would require coordination/collaboration between
multiple stakeholders, (3) would need to be imple-
mented relatively rapidly and efficiently, and (4) would
require the development of new decentralized mainte-
nance and quality control measures.
Through description of these four challenges, this
paper presents green infrastructure as a case study
in multi-domain, complex infrastructure planning.
For green infrastructure to become part of urban
stormwater management plans, many challenges need
to be overcome. These include overcoming spatial
constraints, and engendering meaningful levels of par-
ticipation, in reasonable timeframes, with adequate
quality assurance. To be effective, green infrastructure
projects also need to be implemented in both the public
right of way, and on private lots. While the challenges
are formidable, they must be considered as the only
alternative to investing billions of dollars in concrete
tanks and tunnels that, while they may control urban
runoff, provide few other benefits to urban residents.
Indeed, the future of the management of water
infrastructure assets will be informed both by partic-
ipatory processes and lessons learned from the fields
of urban, transportation, and watershed planning. It
will also require new vision. We need to view high-
ways as more than simply conduits for the passage of
259
Managing catastrophic risk to Istanbuls housing infrastructure
R.W. Muir, N. Sunger, R. Wittke, & P.L. Gurian

ABSTRACT alternatives. Separate evaluations were conducted for

both building types. The RC units were analyzed at
This study conducted a comparative analysis of miti- three levels of damage, while two probabilities were
gation alternatives for a destructive scale earthquake considered for masonry structures; destructive scale
(magnitude 7.68) in the Zeytinburnu District in the damage and no destructive scale damage. The total
City of Istanbul, Turkey. Zeytinburnu is particularly loss calculated for each outcome was the sum of prop-
vulnerable to earthquakes due to the districts close erty loss, mortality loss (in high damage scenario)
proximity to the North Anatolian Fault, alluvial soil, and cost of the selected action. The expected prop-
and seismically weak residential building stock. The erty loss was a function of the damage ratio and value
mitigation strategies investigated within the scope of the housing unit. The expected costs for the RC
included seismic retrofit to reduce the probability of scenario were $9,372 for the retrofit option, $9,760
building collapse during an earthquake, relocation of for relocation, and $17,612 if no action were taken.
the family to a seismically-stable dwelling, or to take The expected costs for the masonry building scenario
no action at all. was $18,137 for the retrofit option, $18,301 for reloca-
The study considers economic loss for the most tion, and $23,157 if no action were taken. While there
vulnerable residential building groups including poor does not appear to be substantial benefit of adopting
quality, mid-rise reinforced concrete (RC) frame build- retrofitting measures over relocating to a seismically-
ings and weak masonry structures. The focus on these safe home, mitigation clearly is preferred over taking
two building types is driven by the fact that 86% of no action at all. A sensitivity analysis indicated that
the residents of Zeytinburnu live in mid-rise, clus- the mitigation alternatives were preferred over the no
tered, apartment houses, 69% of which are considered action alternative over the full range of parameters val-
unsafe. Masonry dwelling units in Zeytinburnu are ues. While this is a preliminary study, that required
generally of poor quality and would not likely pro- many assumptions and is subject to substantial uncer-
vide the lateral resistance to withstand a major seismic tainty, the expense of the mitigation efforts appears
event. Viable retrofitting measures appropriate to these justified. While future efforts to improve the risk
types of buildings were identified in the study. assessment are justified, efforts to improve risk com-
Decision trees were developed in order to compare munication and develop public policies to reduce this
expected values of the three earthquake mitigation risk.
260
System-identification of multi-domain (human, natural, and engineered)

infrastructure systems
A.R. Pradhan, P.L. Gurian, F.A. Montalto, F. Moon & A.E. Aktan
ABSTRACT and culminate in the development of software sys-

tem for integrated asset management. The five steps of
Population growth and urbanization trends pose cha- system development are as follows: 1) Identification
llenges for existing civil infrastructure systems. Unfor- of Information Flows, Institutional Knowledge Repo-
tunately planning for such infrastructure systems has sitories, and Management Metrics 2) Assessment of
usually not considered system-level impact, which has Forecasting Tools and Needs 3) Development of Asset
led to a wide range of unintended impacts. Construc- Management System Architecture, 4) Database Devel-
tion of highways resulted in low density sprawling opment, and 5) Database Interface Development and
development and increased the extent of impervious Refinement.
surfaces, creating storm water runoff which adversely The process should be based on what software
affected aquatic ecosystems. developers call an iterative and incremental devel-
While the importance of the interactions between opment model, rather than the traditional waterfall
infrastructure systems is becoming increasingly evi- management model of software development by suc-
dent, models that can be used to simulate, visualize, cessive execution of pre-defined tasks. The iterative
and predict these interactions are still rare. Even within and incremental approach allows the software devel-
infrastructure systems, existing asset management oper to implement only a small subset of software
approaches are fragmented, generally internalizing the requirements and iteratively enhance a sequence of
opportunities and constraints of only a single asset versions with additional functional requirements until
class within a given system, such as bridges or pave- the full system is implemented. The need for such
ment. During an era of such critical planning for the an approach is derived from the uncertain nature of
future, there is a need to devise new integrated, multi- large, layered infrastructure systems with dynamic
domain decision-making models that directly incor- boundaries used and managed by multiple stakehold-
porate the important interactions that exist between ers. A key management objective in such systems
layered infrastructure systems, and the human and nat- is to be able to accommodate changing performance
ural communities that are impacted by them. In this goals in response to the needs of multiple stakehold-
paper we propose a process for systematically iden- ers with influence over multiple time horizons and
tifying the human, natural, and engineered systems scales.
comprising infrastructure systems and developing an As future research, the authors plan to develop a
asset management system that manages these assets prototype asset management system based on the pro-
in an integrated fashion. The proposed approach con- posed five step approach and validate the prototype
sists of five steps that begin with system identification system with real case studies.
261
A systems approach to assessing the sustainability of the Grand

Canal of China
N. Tsung, R. Corotis, P. Chinowsky & B. Amadei

University of Colorado at Boulder, USA
ABSTRACT influence them. From this preliminary list, a subset

of indices are selected that take into account limi-
The prevailing approach to infrastructure sustainabi- tations on the availability of data. These indices are
lity studies today is to implement changes in existing then normalized so that they can be combined into
facilities, monitor these changes, and compare the each sustainability measure, using weighting func-
results to the initial conditions (Bell and Morse, 2003, tions. The three sustainability measures (economic,
2008). One of the most obvious drawbacks of this environmental and social) are then combined into
forward approach is that people tend to draw conclu- the Infrastructure Sustainability Indicator (ISI), again
sions too quickly. The uniqueness of this study is that using weighting functions.
of looking back by studying the historical materials This brief paper is dedicated to the background of
of the Grand Canal of China, a centuries-old critical the approach and the selection of appropriate indices.
infrastructure system. This process examines the issues that were important
The Grand Canal of China started in Beijing in the over the 100-year study period, and utilizes the advan-
north and ended in Hangzhou in the south, with a tage of hindsight to relate these factors to available
total length of about 1,250 miles. For more than six data.
centuries (12891905), the grain supplied to the per-
sonnel of the imperial court, as well as to the troops
stationed in Beijing and along the Great Wall, was
transported via the canal (Li, 2008; Wu, 1999; Yao, REFERENCES
1998). The activities of the canal were recorded in
detail throughout its life cycle. Many of these doc- Bell, S. and Morse, S. 2003. Measuring Sustainability:
Learning from Doing. Earthscan: London.
uments have survived and are in libraries in the US,
Bell, S. and Morse, S. 2008. Sustainability Indicators:
China, and Taiwan (Liao, 2006). Measuring the Immeasurable? Earthscan: London.
The temporal focus of this study is from 1770 to Clayton, A. M. H. and Radcliffe, N. J. 1996. Sustainability:
1870, which is the period during which China as well a Systems Approach. Earthscan: London.
as the canal declined rapidly. Although his early years Li, D. 2008. Hydraulic Works for Chinas Grand Canal and
saw the continuing of an era of prosperity in China, the Yellow River and Their Influences to the Society and
historians generally agree that later in his reign China Environment, 13681911. A dissertation submitted to the
began declining. Spatially, this study focuses on the Faculty of the Graduate School of Fudan University at
canal at Huaian, where it crossed the Yellow River Shanghai in partial fulfillment of the requirements for
the final defense of Doctoral Degree, Institute of Chinese
until 1855
Historic Geography.
Sustainability indicators such as economic, insti- Liao, S. 2006. Customs Stations and Regional Economy in
tutional, and environmental as well as indices rel- Qing Dynasty. A dissertation submitted to the Faculty of
evant to these indicators can be examined in the the Graduate School of the Normal University of Shang-
historical-geographical context of the Grand Canal of hai in partial fulfillment of the requirements for the final
China. In this systems approach (Clayton and Rad- defense of Doctoral Degree, College of Humanities.
cliffe, 1996), an Infrastructure Sustainability Indicator Wu, Q. 1999. Grain Shipping and Chinese Society. Huazhong
(ISI) is developed, based on the three sustainability Normal University Press: Huhan.
measures: economic, environmental and social. Yao, H. 1998. History of Chinas Grand Canal. China
Hydraulic and Hydropower Publisher: Beijing.
Each of these sustainability measures is then con-
sidered and examined in terms of factors that might
262
Strategies of investment in the management of urban bridges
G. Vacheyroux
cole des Ingnieurs de la Ville de Paris, France
R.B. Corotis
University of Colorado, Boulder, Colorado, USA
ABSTRACT too theoretical or vague. The chosen example is Paris,

France. Crossed by 37 bridges on a 13 km-long path,
Decisions on maintenance and rehabilitation strate- the Seine River has exposed the French capital to
gies for a portfolio of bridges should be based on life floods for millennia. The data are mainly provided by
cycle cost and benefit analysis (LCC). With a limited the Office of the Mayor of the City of Paris, and are
source of funding, however, further strategy is needed complemented by assumptions when required. This
to determine the priority and timing of expenditure. application is especially interesting since Paris has a
The communication and presentation of programs to very diverse portfolio of bridges, with a wide variation
the public at large are extremely important. in dates of construction, material and their dimensions.
This paper discusses and broadens the perspectives
of bridge management, providing a life cycle cost sav-
ings and investment tool to guide decisions regarding REFERENCES
bridge maintenance and rehabilitation. A spreadsheet
computer program is developed to incorporate user Life-Cycle Costing Manual for the Federal Energy Manage-
supplied costs and benefits, and to rank different ment Program, NIST Handbook 135, S. K. Fuller & S. R.
projects by priority, using the savings to investment Peterse, 1995
ExpertFit users guide, Law, A., and Vincent, S., 1994
ratio (SIR), a measure of cost-effectiveness.
User Benefit Analysis for Highways, American Association of
The program is based on a probabilistic approach, State Highway and Transportation Officials, Washington,
with the SIR involving the initial investment; annual DC, 2003
operation, maintenance and repair/rehabilitation costs; A Multi-attribute Framework for RiskAnalysis of Natural Dis-
and failure costs. These future costs are linked to prob- asters, Evan Martin Hammel, B.S., University of Colorado
able events, such as natural hazards, vehicle and boat at Boulder, 2005
crashes, and user time delay. Life cycle cost analysis Alternative valuation of highway user delay costs, A. J.
can be sensitive to the selection of an appropriate time Trzcinski and R. B. Corotis, University of Colorado at
period, and this paper introduces the concept of infi- Boulder, 2006
Highway Engineering 6th edition, Wright, P.H., J. Wiley &
nite lifetime for bridge maintenance. Consideration is
Sons, New York, 1996
also given to aesthetic considerations, which for a city Road User and Mitigation Costs in Highway Pavement
such as Paris has economic consequences. Project, National Cooperative Highway Research Pro-
An example of use of the program is shown, so gram Synthesis of Highway Practice 269, D. L. Lewis,
that the developed method is not viewed as either 1999
263
Eco-audit of seven green infrastructure practices
U. Wegst, C. Barr & F. Montalto

ABSTRACT BMP scenario 5: sidewalk reservoirs with street

trees, 0.1 ha (Table A7)
This paper describes an Eco Audit of seven different BMP scenario 6: sidewalk biofiltration, 0.1 ha
urban green infrastructure practices. It analyses the (Table A8)
amount of energy embodied in the materials used in the
The picture that emerges from the analysis of the
construction of the BMPs and the CO2 emissions foot-
two way street is similar to that of the one way street.
print associated with their production. Different BMP
BMP scenario 4, the sidewalk with permeable paving,
scenarios are applied to both a typical one-way and a
leads to a significant reduction (719%) in embodied
typical two-way New York City Street and compared.
energy; BMP scenario 6, the sidewalk with biofiltra-
For the 152 m long, one-way street, the following
tion, has an embodied energy and CO2 footprint that
scenarios were compared:
is very similar to the do nothing case. In contrast,
No BMP BMP scenario 5, the sidewalk reservoirs with street
BMP scenario 1: sidewalk permeable paving, 0.1 ha trees, leads to a significant increase (7383%) in the
BMP scenario 2: sidewalk reservoirs with street environmental impact.
trees, 0.1 ha The design of environmentally benign BMP sce-
BMP scenario 3: parking lane permeable paving, narios has many aspects; one is eco aware materials
0.1 ha selection, the choosing of construction materials with
low embodied energies and CO2 footprints. The Eco
Both BMP scenario 1, the sidewalk with permeable
Audits reported above are quick, approximate assess-
paving, and BMP scenario 3, the parking lane with
ments of energy consumption and CO2 emissions that
permeable paving that receives runoff from sidewalk
are associated with the construction materials used in
and street, lead to a significant reduction (923% and
alternative BMP scenarios applied to a typical one-
29%, respectively) in embodied energy and CO2
way and a typical two-way New York City street.
footprint. In contrast, BMP scenario 2, the sidewalk
They reveal which components of the design domi-
reservoirs with street trees, leads to a significant
nate the environmental impact and provide guidance
increase (5875%) in the environmental impact.
for the decision-making process and the exploration
For the 152 m long, two-way street, the following
of alternative scenarios.
scenarios were compared:
No BMP (Table A5)
BMP scenario 4: sidewalk permeable paving, 0.1 ha
(Table A6)
264
Determination of a performance baseline for lifecycle consideration

of bridges
H. Wenzel & R. Veit-Egerer

VCE, Vienna, Austria
ABSTRACT
There is global agreement that bridge rating based

on subjective visual inspection needs to be improved.
Shrinking budgets of our transportation infrastructure
are contrary to the growing demand identified for our
aging structures. It is a fact that not all deficiencies
identified can be mitigated. Therefore risk manage-
ment is a must in order to determine priorities and a
ranking of necessary investments.
The European approach has been focused on ambi-
ent vibration monitoring since 15 years. Considerable
progress has been made and prototypes of bridge man-
agement systems providing decision support based on
case based reasoning routines have been built. In the Figure 1. Expected (analytical) lifeline of new structures.
large European Research Project IRIS (CP-IP213968-
2) this procedure is further developed into a tool,
The operation and maintenance aspect covering
which, in combination with a reduced visual inspec-
mainly the optimization of the financial demand
tion, has demonstrated that our current rating practice
for preventive maintenance and budget planning
produces in average too low ratings.
In the European bridge stock (about 1 million Conventional lifecycle models are based on the
bridges operated by some national authorities) 12% are information provided by the respective databases.
rated deficient. It is anticipated that only 2% deserve In order to introduce objective values for assess-
special attention. The proposed procedure therefore ment monitoring results are utilized. In the following
knows 2 levels of approach: description this process is elaborated.
The presented approach has been developed for
A quick and cheap basic assessment campaign to Central European conditions and the related database
produce a ranking of the structures and information structure. Nevertheless the principle
An elaborate risk monitoring and assessment rou- can be applied globally when the respective interfaces
tine which goes deeper into detail and produces are fitted for the specific purpose.
recommendations for rehabilitation
The methodology has been applied to a large bridge
REFERENCES
stock in Europe and is able to show first results. For
structurally deficient bridges the ranking is improved Bondini, F. and Frangopol, D.M., eds. 2008. Life-Cycle
in average by over 25%. Civil Engineering. CRC Press Taylor and Frances Group,
Lifecycle Modeling has 2 aspects: London. ISBN: 978-0-415-46857-2 (hbk)
Cruz, P.J.S, Frangopol, D.M., Neves, L.C., eds. 2006.
The safety and reliability aspect covering the bridge Bridge Maintainence, Safety, Management, Life-Cycle
owners responsibility to provide a save and reliable Performance and Cost. Taylor & Fracis, London. ISBN:
infrastructure 0415403154
265
Study on evacuation simulation system for urban expressways using

multi-agent system
M. Yamawaki
Graduate school of Engineering, University of Kagawa, Japan
W. Shiraki & H. Inomo

Department of Reliability-based Information Systems Engineering, University of Kagawa, Japan
Y. Kawakami
Hanshin Expressway Company, Osaka, Japan
K. Yasuda
NEWJEC Inc., Osaka, Japan
ABSTRACT change the simulation condition in real time. There-

fore, using the system, different persons concerned
Recently, the occurrence of the large-scale earthquake with disasters such as inhabitants, the persons con-
will be expected with the high probability in Japan cerned with fire fighting, the expressway manager, the
within 30 years in the future. Therefore, performance administrations, the experts and so on can exchange
securing of infrastructure facilities and the users life opinions about the evacuation and reflect and plan a
securing are strongly demanded when disaster occurs, different of view.
the disaster prevention and disaster mitigation mea- Moreover, by using this developed system and
sures are matters of great urgency. Since Great Hanshin reproducing the disaster situation (Impassable by
Earthquake, introduction of new earthquake-proof destruction of structure, Collision of car to structure,
and seismic isolation technology that accompanies Collision of car to structure, Fires, etc.) on the express-
review of earthquake resistant design code of struc- way that can happen when the large-scale earthquake
ture and earthquake strengthening of structure made occurs, the method that evacuates car and person on
from design standard before Great Hanshin Earth- the expressway from the situation promptly and safely
quake are executed. As a result, the protection in the is examined. In this paper, Influence that transmission
hardware side is steadily promoted. However, to do of disaster information and roadway securing instruc-
prompt restoration and restart of business after suf- tion for emergency car from expressway manager gives
fering, it is not only measures of the hardware side, to evacuation time of car and Influence that number
and it is important to examine disaster prevention and of evacuation gates gives to evacuation time of evac-
the disaster mitigation measures on a soft side con- uee when the disaster occurs are examined. And, the
cerning a prompt evacuation conduct from collapsed effectiveness of the evacuation guide is examined.
structure. Especially, early restoration is demanded as
for the expressway. Because it is an important infras-
tructure for transportation of the rescue supply and REFERENCES
prompt disaster recovery.
In this study, simulation system for Hanshin Higashi Oouchi, Masato Yamamoto, Hidenori Kawa-
mura, Basis and application of Multi-Agent System-
Expressway that is urban expressway is developed to
Computing paradigm of complex systems engineering-,
examine disaster prevention and disaster mitigation Corona company, 2002.
measures from soft measures. And, car and persons Katuhiro, Nishinari, Masataka Tamura, Traffic jam study
evacuation guides in the expressway are examined. that understands well, Natume company, 2009.
Specifically, the evacuation simulation system is devel- Haruki Aritomo, Wataru Shiraki, Hitoshi Inomo, Deve-
oped that constructs behavior model of evacuation lopment of participatory evacuation simulation system
car and evacuee when disaster occurs using MAS for Live Design, Journal of Safety Problems, Vol. 1,
(Multi-Agent System). This system made the simu- November, 2006.
lation process visible by using JAVA applet, and it can
266
MS10: Bridging the data gaps for effective management
Organizer: H. Ghasemi
Defining and implementing data collection goals for the Long Term
Bridge Performance program
M.C. Brown
Virginia Transportation Research Council, Charlottesville, VA, USA
ABSTRACT 3. Determine what data the states currently collect and

use for their decision-making processes and what
The Federal Highway Administration has initiated the gaps they see in their currently available data.
Long Term Bridge Performance (LTBP) program as 4. Identify the aspects of bridge performance on
authorized in the Safe, Accountable, Flexible, and which the states would like the program to focus.
Efficient Transportation Equity Act: A Legacy for 5. Develop and support a community of practice for
Users (SAFETEA-LU), signed into law in August of the Long Term Bridge Performance program.
2005. The LTBP Program is a 20-year, multi-faceted
Based on the comments and recommendations pro-
research effort that is strategic in nature and has both
vided by state highway agencies, the LTBP team
specific short-term and long-range goals. It is antic-
identified a list of research topics related to various
ipated that the LTBP program will provide a better
aspects of bridge performance. The list was compiled,
understanding of bridge deterioration, focusing on its
and each topic was supplemented with supporting
numerous causes including corrosion, fatigue, envi-
background information and literature review, defi-
ronment and loads. The program will collect informa-
nition of general scope. The compiled list was then
tion regarding the effectiveness of current maintenance
shared with a select group of experts from government,
and improvement strategies, and information on the
academia and industry who provided input regarding
operational performance of bridges, focusing on con-
the perceived urgency and importance of each of the
gestion, delay and accidents. It is envisioned that the
proposed study topics.
quantitative database developed under the LTBP pro-
Each of the research topics have been developed by
gram will be used to solve a variety of bridge condition
outlining known background information, identifying
assessment and management problems, to develop new
knowledge gaps or limitations in practice, and then
tools, and to advance the state-of-the-knowledge of
posing a series of research questions, with associated
bridge design, maintenance and preservation.
hypotheses and data requirements, to arrive at spe-
An interview process was adopted to assist the team
cific types of data needed to address the topics. These
in achieving the following objectives:
data requirements will be further refined into proto-
1. Develop an understanding of how representative cols for the gathering of specific data and methods for
states manage and track bridge performance. conducting associated tests or field evaluations.
2. Identify the most common concerns and the most Using the methodology outlined herein, the LTBP
costly activities of the representative states in team has established priorities for the program, deter-
maintaining, repairing, rehabilitating and replacing mined specific study objectives and identified data
bridges. required to fulfill those objectives.
269
Bridging the data gaps for effective management, session 3:

performance LTBP data infrastructure and data integration
M. Dejori, R. Ciprian & C. Neubauer

Siemens Corporate Research, Princeton, NJ, USA
ABSTRACT Three requirements and their solutions will be

described in more detail. Namely, a software architec-
The Long Term Bridge Performance (LTBP) program ture to ensure extensibility, a data integration frame-
aims at an in-depth understanding of the underly- work to allow for the seamless integration of additional
ing causes for bridge structures performance issues, data sources, and a data security concept to enforce
such as deterioration processes for example. A bet- data safety.
ter understanding will ultimately lead to a better A hybrid data integration framework is proposed
way of managing the bridge infrastructure, mini- taking advantage of the two main data integration
mizing life-cycle costs and improving the current approaches, the fused data and the federated data
asset management strategies. In order to achieve this approach. The introduced framework allows the inte-
goal, during the LTBP program massive amounts gration of multiple heterogeneous data through a
of data will be collected from different sources, centralized data warehouse as well as through a service
such as retrospective inventory data, environmental oriented architecture (SOA) that links data from decen-
information, visual inspection reports or instrumen- tralized locations. This framework is highly scalable
tation data, to mention a few. The main hypothesis and extensible, providing the necessary foundation for
is that putting together all these pieces of informa- the Long Term Bridge Performance Program.
tion will help in a better understanding performance
issues and generate more accurate predictive main-
tenance strategies, optimized life-cycle cost models REFERENCES
and overall improve maintenance policies. The central
piece of this undertaking is a sound data manage- Ziegler P. and Dittrich K.R. Three Decades of Data
Integration All Problems Solved? In the 18th IFIP
ment framework that facilitates the integration of
World Computer Congress (WCC), Volume 12, 2004.
information and the mining of knowledge nuggets Data Integration Primer. Federal Highway Administration.
commonly referred to as data mining. This publica- 1999.
tion provides an overview of the data integration and Review Of Data Integration Practices And Their Applications
management framework build as part of the LTBP To Transportation Asset Management. Federal Highway
program. Administration. 2003.
270
LTBP pilot study methodology and preliminary results
A.J. Foden & V.R. Sibley

Parsons Brinckerhoff, Princeton, NJ, USA
ABSTRACT next generation of inspection technology, allowing

researchers to evaluate corrosion in a bridge deck
The Federal Highway Administrations (FHWA) Long throughout its lifecycle, from detecting corrosion
Term Bridge Performance (LTBP) Program is a potential at its earliest stages to characterizing ongo-
20-year, flagship research program aiming to obtain ing delamination. The LTBP Program is using half-cell
scientific quality data which can be used to, among corrosion potential surveys, ground penetrating radar,
other things, better understand bridge performance and impact echo, and ultrasonic testing at all pilot bridges.
drive data-based bridge management methods. Utilizing this full-range of NDE tools creates a com-
The Pilot Study Phase of the LTBP Program has prehensive understanding of the current health of the
begun, seeking to test data collection methodologies bridge and allows an early diagnosis of underlying
and obtain useful data early in the life of the Program. problems, enabling researchers to trace the progres-
Seven bridges scattered throughout the United States sion of deterioration before it can be detected at the
are involved in the Pilot Study, each representative surface.
of typical environment conditions and structure types Various concrete testing techniques are also being
common in the Nations broader bridge population. used in the Program to obtain more detailed informa-
These bridges provide a platform for researchers to tion on material properties, such as chloride content,
validate, streamline, and field test the inspection, test- and to serve as ground truth verification of NDE
ing, and instrumentation procedures prescribed earlier testing.
in the development phase of the Program, and ensure Live load and dynamic testing are being used at
the necessary components are in place to achieve the each pilot bridge to analyze the structural behavior
Programs objectives once it is expanded to a larger of the bridge system. The baseline structural condi-
sample of the population. tions currently being recorded will allow researchers to
To achieve the objectives of the LTBP Program, track changes in structural behavior and critical detail
researchers are using a complementary package of behavior over time, and compare it to the material
both traditional and state-of-the-art data collection deterioration data obtained in inspection, NDE, and
techniques. At the time of this paper, researchers have concrete testing.
conducted initial testing at four of the pilot bridges to Close coordination of all field activities is vital to
establish a baseline condition for comparison during the success of the pilot program. To minimize disrup-
long-term monitoring. tion to traffic flow, multiple tasks are consolidated and
A detailed visual inspection, the industry stan- completed during a single lane closure whenever pos-
dard for monitoring deterioration, has been completed. sible. This requires numerous LTBP teams, including
More detailed than a typical biennial inspection, LTBP the visual inspectors, NDE/NDT operators, and deck
Program inspections feature the use of ScanPrint monitoring crew, to share space on the bridges with
computer software, allowing inspectors to accurately limited periods of time to conduct their work.To ensure
record the size and location of a defect electronically activities are carried out efficiently, safely, and to the
onto a CADD bridge plan while in the field. satisfaction of the bridge owners, tremendous amounts
In conjunction with the visual inspection, LTBP of coordination and communication are required.
Program Pilot Study bridges are also evaluated The preliminary results of this array of tools are
using a suite of non-destructive testing and evalu- now being analyzed and correlated to begin to answer
ation (NDT/NDE) techniques. These tools are the critical questions regarding bridge performance.
271
Enhancing bridge performance defining/measuring/improving

bridge performance
H. Ghasemi
Federal Highway Administration, McLean, Virginia, USA
J.M. Hooks
Highway R&D Services, Cape Coral, FL, USA
ABSTRACT Table 1. High priority bridge performance issues.
The transportation user community in the United Issue

States expects and deserves a highway system that
routinely provides high quality service in terms of Performance of untreated concrete bridge decks
Performance of bridge deck treatments (membranes,
safety, efficiency and economy while having little or
overlays, coatings, sealers)
no negative impact on the local or global environ- Performance of precast reinforced concrete deck systems
ment. Optimal (i.e., safe, efficient and economical) Performance of alternative reinforcing steels
operation of public highway systems is dependent on Influence of cracking on the serviceability of
many factors. Bridges are critical nodes in the highway high-performance concrete decks
infrastructure and poor performance of any highway Performance and repair of bridge deck joints
bridge can reduce the operating condition of the high- Performance of jointless structures
way system of which it is part. Four aspects of bridge Performance of coatings for steel bridge elements
performance are critical to owners and highway users Performance of weathering steels
Performance of bare or coated/sealed concrete
alike. These are: the structural condition and durability
superstructures and substructures (splash zone, soils, or
of the bridge, the functional capacities of the bridge, exposed to deicer runoff)
the integrity of the bridge in relation to potential fail- Performance of embedded or ducted prestressing wires
ures and the overall costs of maintaining a bridge in and post-tensioning tendons
service over its lifespan. Performance of prestressed concrete girders
In the United States, there exists a current knowl- Performance of bridge bearings
edge base of inventory information and condition data Direct, reliable, timely methods to measure scour
on bridges that is by far the best in the world. Cur- Performance of scour countermeasures
rent programs and methods of bridge inspection and Performance of structure foundation types
Performance of innovative materials and designs
the tools used for managing bridge programs are also
Risk-based management approach
unmatched anywhere.Yet the level of understanding of Operational performance of functionally obsolete bridges
how bridges perform and how to satisfactorily measure
their performance falls well short of desirable. This is
because the performance of any single bridge is depen-
dent on complex interactions of multiple and often
interrelated factors which include: the original design and ultimately of replacement actions. All of these fac-
parameters and specifications (bridge type, materi- tors combine to impact the condition and operational
als, geometries, load capacities); the initial quality capacities of the bridge and its various structural ele-
of materials and of the as-built construction; varying ments at any given point in the life of the bridge. This
environmental conditions of climate, air quality, etc; session will provide a clear and complete perspec-
corrosion and other deterioration processes; traffic vol- tive on the concept of bridge performance and present
umes and types, weights and volumes of truck traffic; the path to a better understanding of what bridge per-
and the type, timing and effectiveness of preventive formance means and how to reliably measure it and
maintenance, of minor and major rehabilitation actions improve it.
272
NDE condition assessment of reinforced concrete bridge decks

within the LTBP program
N. Gucunski, F. Romero, H. Parvardeh & F.A. Fetrat

Rutgers University, Center for Advanced Infrastructure and Transportation, Piscataway, New Jersey, USA
S. Kruschwitz & R. Feldmann

Federal Institute for Materials Research and Testing (BAM), Berlin, Germany
ABSTRACT technologies is illustrated on material characterization

and defect/deterioration detection of RC bridge decks.
Reinforced concrete (RC) bridge decks are exposed The NDE evaluation was complemented by ground-
to several types of deterioration processes: corrosion, truth measurement on the cores extracted from nine
alkali-silica reaction (ASR), carbonation, shrinkage, bridge decks.
freeze-thaw actions, etc. The most commonly found The condition assessment using the six NDE tech-
problem is corrosion induced concrete deterioration, nologies has clearly shown their advantages and limi-
which is initiated by rebar corrosion and afterwards tations. For example, the GPR survey provides assess-
extends into concrete cracking, delamination and ulti- ment of concrete deterioration at relatively high speeds
mately concrete spalling. Previous studies have shown of data collection. On the other hand, the IE and UE
that surveys of bridge relying on a single nonde- provide high accuracy in detection and characteri-
structive evaluation (NDE) technology provide only zation of delaminations in the deck, but at a lower
limited information about condition of reinforced con- testing speed. Half-cell potential test provides accu-
crete elements. To overcome limitations of individual rate assessment of likelihood of corrosion, while the
technologies, a complementary approach using several USW test provided accurate assessment of the effects
NDE technologies should be used in RC evaluation. of deterioration processes and defects on mechanical
The presented approach utilizes a suite of NDE properties, primarily the elastic modulus degradation.
technologies, namely: impact echo (IE), ultra- Most importantly, the surveys have shown advantages
sonic echo (UE), ultrasonic surface waves (USW), of the use of multi-modal NDE surveys in the com-
ground penetrating radar (GPR), half-cell potential prehensiveness of condition assessment of concrete
and electrical resistivity. The application of listed bridge decks and RC elements in general.
273
Performance indicators for highway bridges and their integration

in bridge management

ABSTRACT
Performance indicators are key and a crucial decision

tool. The uncertainties in the available information
for computing the performance indicators and their
changes over time impose a great challenge in the
decision making process.
The assessment of reliability of highway bridge
structures based on component reliability leads to
either over-conservatism in evaluation of redundant
structural systems or to overestimation of the actual
load carrying capacity of non-redundant structural
systems (Hendawi and Frangopol 1994). However, the
performance of a highway bridge structure can be eval- Figure 1. Interaction among programs and the analyst for
uated by assessment of its structural system reliability computing the life-cycle performance.
(Estes and Frangopol 1999).
To compute the system reliability, Estes and
Frangopol (1999) and Akgl and Frangopol (2004a,b) (3) forecast the average daily traffic, (4) perform statis-
used AASHTO guidelines to formulate limit state tics of extremes for finding the live load multiplier,
functions for the components of various bridge system (5) perform the response surface analysis, (6) perform
against different failure modes. The system relia- the Latin hypercube sampling for resistance compu-
bility was then computed by establishing a series- tations, (7) compute the point-in-time reliability, (8)
parallel model for an assumed failure scenario for the compute the cumulative-time failure probability, and
structural systems. The shortcoming in this approach (9) compute the life-cycle performance.
is clearly the need to assume a model for system In this paper, various life-cycle performance indi-
failure. cators are reviewed, and their integration in bridge
Incremental nonlinear finite element analyses (INL- management is discussed.
FEA) has recently been used in the computation of
system reliability of bridge structures. Using INL-FEA
REFERENCES
accurately allows for the prediction of the resistance of
the entire system. In fact using this approach, service- Akgl, F. & Frangopol, D.M. 2004a. Computational platform
ability as well as ultimate limit states can be considered for predicting lifetime system reliability profiles for dif-
accordingly. ferent structure types in a network. Journal of Computing
An interaction among several computer softwares is in Civil Engineering, ASCE 18(2): 92104.
an integral part of the success of the process. Figure 1 Akgl, F. & Frangopol, D.M. 2004b. Lifetime perfor-
shows a diagram that illustrates this interaction. The mance analysis of existing steel girder bridge superstruc-
connectors in the diagram show the directions of inter- tures. Journal of Structural Engineering, ASCE 130(12):
18751888.
action among the software programs (ABAQUS, MAT- Estes, A.C. & Frangopol, D.M. 1999. Repair optimization of
LAB, VisualDOC, RELSYS, and RELTSYS) and the highway bridges using system reliability approach. Jour-
numbers show the steps that each interaction is estab- nal of Structural Engineering, ASCE 125(7): 766775.
lished to achieve. Some steps are in fact performed on Hendawi, S. & Frangopol, D.M. 1994. System reliability and
a single program. These steps are: (1) build the finite redundancy in structural design and evaluation. Structural
element model, (2) generate the response envelope, Safety, Elsevier 16(1+2): 4771.
274
Parameter identification of a reinforced concrete T-beam bridge
Y. Zhou, J. Weidner, J. Prader, N. Dubbs, F. Moon & E. Aktan

Drexel Intelligent Infrastructure Institute, Drexel University, Philadelphia, PA, USA
ABSTRACT infrastructures. The methodology proposed within

aims to use the Application Programming Interface
Parameter identification based on model updating is (API) function in the Strand7 FE software package
one of the most important links in Structural Iden- to automatically update selected parameters using
tification. The benefits of combining finite element Matlab. The proposed methodology was applied to a
(FE) analysis with on-site measurement through model three span, simply supported T-beam bridge. Both the
updating are significant and growing as more reli- elastic modulus for the global structure and the crack
able experimental, modeling and model correlation height of the primary girders were utilized as update-
approaches become available. The authors have suc- able parameters in two separate cases. The results show
cessfully applied Structural Identification to numerous that the average crack height parameter reproduced the
bridges over the past decade. However, manual model static measurements with a high degree of accuracy
updating of structural parameters has significantly and is a reasonable parameter choice for this class of
limited the application of model updating on large structure.
275
MS12: Computational prediction & in field validation
of bridge performance
Organizers: F. Biondini, F. Bontempi & P.G. Malerba
Probabilistic performance prediction modeling for bridges considering

maintenance effects within a combined computation, visualization
and programming environment
F. Akgl
Department of Engineering Sciences, Middle East Technical University, Ankara, Turkey
ABSTRACT performed as part of this application specific solution

using the toolboxes. For the simulation of the ran-
Various models have been developed in recent past dom variables, Latin Hypercube Sampling technique
for lifetime performance prediction of bridges. Studies is used which is available in the computational envi-
for developing new theoretical models and improve- ronment. Random variables may be assigned normal
ment of the existing ones are part of ongoing research distributions. A function is also developed to generate
in bridge deterioration and management. However, samplings from other distributions such as triangu-
improvement of such theoretical models so that they lar distribution. Instead of using superposition of the
can be practically implemented in bridge manage- effects of maintenance actions on no-maintenance pro-
ment systems is a challenging task. In order to achieve files, the newly developed algorithm consists of a
such task, newly developed models must be tested and simulation loop which contains a time loop embed-
further improved to achieve practicality. ded in it. Such an algorithm enables the generation
Lifetime performance prediction for bridges can be of the performance profile over the whole lifetime at
performed using either a safety or condition criteria. once for a single simulation of all random variables
In a well designed bridge management system, it is involved. This approach may prove to be a faster algo-
well established that both of these criteria should be rithm than the one previously developed. Mean value
implemented and monitored at the same time. The and standard deviation profiles for the condition and
Markovian decision process-based models, linear or safety indices as well as the probability density dis-
polynomial curve-based techniques and profile-based tributions of these indices over the bridge lifetime are
probabilistic methods are the methods to conduct con- graphically generated using the integrated modeling
dition or safety-based performance prediction and and simulation environment.
maintenance optimization. Linear or curve-based tech- An overview of the condition prediction or dete-
niques use linear or polynomial functions to predict rioration models, ranging from the simplest linear
lifetime deterioration based on the deterioration rate model to more complex probabilistic piecewise lin-
and time as the parameters. Profile-based robabilistic ear model based on simulation, is also presented.
methods are based on formulas defining the profiles Using the developed environment, effect of various
and effects of maintenance actions which contain ran- maintenance-repair-replacement scenarios on condi-
dom variables having probability distributions.In this tion and safety profiles of bridges is investigated. The
study, an investigation and improvement of an exist- computational platform may further be developed
ing piecewise linear lifetime condition and safety into a deterioration model analysis tool for infras-
prediction model is performed. tructure systems such that the desired model type,
A bridge performance prediction model developed whether linear, polynomial, Markovian or piecewise-
by researchers in this field has been implemented in linear probabilistic, can also be selected from a list of
Matlab environment for further improvement. Com- deterioration model options.
putation, visualization and programming tasks are
279
Structural maintenance and rehabilitation of a 1900s iron bridge
P. Bernardi, E. Michelini & I. Iori

Department of Civil and Environmental Engineering and Architecture, University of Parma, Italy
ABSTRACT ventilation and dewatering. Material losses are mainly

localized in the horizontal flange of iron angles, where
The Tre Martiri iron bridge, across the Oglio river moisture stagnation is more significant. Other signs
near the city of Cremona (in the north of Italy), was of deterioration can be observed in the diagonal ele-
built by the National Society Savigliano and opened ments of the lateral bracing system, which have been
in 1900 as a road bridge. The original project, drawn realized by the coupling of two angle profiles through
up in 1899, was based on the three-moment equation the riveting of their flanges. Water seepage and stag-
proposed some years before by Clapeyron. nation between the angles have caused the formation
The bridge structure is constituted by a continu- of a variable thickness layer of rust, with consequent
ous beam supported by six metallic piers between two swelling of the iron elements.
masonry abutments. The two end spans are 8.10 m After the survey of the main factors leading to
long, while the five central spans are 18.0 m long, for bridge deterioration, the bridge structure has been crit-
a total length of 106.20 m. The bridge is 5 m wide ically analyzed in its actual state and a numerical model
from curb to curb. Its horizontal bearing structure is has been performed so to reproduce the theoretical
constituted by two main continuous iron trusses, with state of stress and verify the ULS conditions of the
parallel chords, braced with St. Andrews cross bracing bridge bearing elements.
system. The floor system consists of iron floor-beams Some hypotheses for a restoration project have then
and stringers, with St. Andrews cross bracing system. been proposed and discussed. Even if the most dete-
The bridge deck has been realized by using iron Zors riorated elements should be removed and substituted
profiles, with a concrete topping characterized by a by new ones, repairs should be done with sensiti-
variable thickness (around 50 mm at lateral sides and vity, so that the historic character of the bridge and
100 mm in the central part). its aesthetic aspect are not jeopardized in the pro-
Even if the existing structure is basically the same cess. Consequently, the selection of proper materials
originally built, the bridge has undergone considerable and structural solutions in the bridge rehabilitation
repairs in 19461947, since it was seriously damaged is crucial in order to minimize the visual impact of
during the 2nd World War. These restoration works, restoration works and provide at the same time an
carried out by the Ansaldo Company of Genoa, inter- adequate safety level.
ested different parts of the bridge and particularly its
deck, iron trusses and foundations. Nowadays, after
about one century from its construction, the bridge REFERENCES
shows signs of major deterioration, mainly due to
weather and environmental factors as well as to an Malerba, P.G. 2007. Inspection, maintenance and rehabi-
inadequate routine maintenance. litation of bridges with river-bed piers (in Italian). Il
The principal factor leading to deterioration of Po 8.
structural members and their joints is surface cor- Croci, G. 2001. Preservation and structural restoration of
architectonic heritage (in Italian). Turin: UTET.
rosion, which has caused a uniform destruction of
Bertolini, L. 2006. Deterioration, prevention, diagnostics,
relatively large surface of iron, so determining the restoration (in Italian). Milan: Citt Studi Eds.
reduction of cross-sections in the structural elements. Cigada, A. 1993. Structural elements and properties of
Damages due to corrosion are mainly located in metallic materials (in Italian). Milan: Citt Studi Eds.
the bottom chord of the iron truss, near the abut- Radomski, W. 2002. Bridge Rehabilitation. London: Imperial
ments, where the vegetation growth causes insufficient College Press.
280
Credibility indicator for bridge service life prediction
F. Biondini & E. Garavaglia

Department of Structural Engineering, Politecnico di Milano, Milano, Italy
D.M. Frangopol
Department of Civil and Environmental Engineering, ATLSS Center, Lehigh University, Bethlehem, PA, USA
ABSTRACT
The assessment of the life-cycle performance of dete-

riorating bridge structures can be formulated as a
reliability problem where a significant loss of perfor-
mance results in failure. Therefore, when a failure is
reached, the system passes from the current state into
another state characterized by a lower level of perfor-
mance. On the other hand, structural performance can
also be improved by maintenance and/or rehabilitation
interventions. In this case the system may move from
the current state into another state characterized by a
higher level of performance. In both cases the process
may be defined as a transition process through dif-
ferent states. Since the problem is affected by several Figure 1. Comparison between the credibility of two distri-
sources of uncertainty, the assessment of the life-cycle butions in competition in the modeling of the failure times.
performance must be based on a suitable modeling The radius of each circle is proportional to the value of the
of damage and maintenance process (Biondini et al. credibility index.
2008).
The probabilistic modeling of the deterioration pro-
cess is affected by uncertainty. In this paper the REFERENCES
influence of the epistemic uncertainty on the service
life prediction of deteriorating structures is investi- Biondini F., Frangopol D.M., and Garavaglia E., 2008. Life-
gated by means of a relative criterion of validation Cycle Reliability Analysis and Selective Maintenance of
introduced by Grandori et al. (1998, 2003). This cri- Deteriorating Structures. First International Symposium
terion is based on a credibility index able to compare on Life-Cycle Civil Engineering (IALCCE08), Varenna,
Italy, June 1014, 2008, F. Biondini & D.M. Frangopol
two models and decide which one is the most reliable (Eds.), CRC Press, Taylor & Francis Group, London, UK,
(Grandori et al. 1998, 2003, Guagenti et al. 2003). 483488.
The effectiveness of this approach is shown with Grandori, G., Guagenti, E. & Tagliani, A., 1998. A pro-
reference to an existing steel bridge. Two probabilis- posal for comparing the reliabilities of alternative seismic
tic distributions, Gamma and Exponential, are put in hazard models, Journal of Seismology, 2, 2735, 1998.
competition in the modeling of the failure times, and Grandori, G., Guagenti, E. & Tagliani A., 2003. Magni-
their credibility evaluated (Fig. 1). The failure pro- tude distribution versus local seismic hazard, BSSA, Bull.
cesses are simulated using Monte Carlo techniques of Seismological Society of America, June 2003, Vol.93,
and the credibility index is built on the basis of a large No.3, pp 10911098.
Guagenti E., Garavaglia, E. & Petrini L., 2003. Probabilis-
set of samples, with different population size. tic Hazard Models: is it Possible a Statistical Validation?
The obtained results show that lifetime predictions in: System based Vision for Strategic and Creative
can significantly depend on the probabilistic modeling, Design 2nd Int. Structural Engineering and Construc-
and prove that the credibility index is able to recognize, tion Conf (ISEC02), Rome, Italy, September 2326, 2003,
in a qualitative and quantitative way, the most reliable F. Bontempi Ed., A.A. Balkema Publishers, Lisse, The
modeling. Netherlands, Vol. II, pp. 12111216.
281
Structural lifetime and elapsed time between first failure

and collapse: Application to an arch bridge
F. Biondini
D.M. Frangopol
ABSTRACT The prediction of structural lifetime and elapsed

time between local failures and collapse of concrete
The structural performance of concrete bridges bridges is investigated in this paper by means of a gen-
exposed to the effects of environmental aggressive- eral approach to probabilistic assessment of concrete
ness is time-variant. In a life-cycle oriented approach structures under diffusive attacks from aggressive
to bridge design it is therefore necessary to evaluate the agents (Biondini et al. 2004, 2006). The proposed
structural performance not only at the initial time of approach is applied to an arch bridge considering dif-
construction, when the bridge is intact, but also during ferent levels of uncertainty (Biondini and Frangopol
the expected lifetime, by taking into account the effects 2008, 2010). The results show that the uncertainty on
of damage under uncertainty (Frangopol et al. 1997, structural behavior and damage process may signifi-
Biondini et al. 2004). This allows to assess the resid- cantly affect the prediction of structural lifetimes and
ual lifetime under prescribed reliability levels for each of the elapsed time between failures. Moreover, the
failure mode of interest, or to plan proper maintenance expected elapsed time between failures, and then the
and repair interventions to achieve the prescribed value required rapidity of a system to be repaired, decreases
of structural lifetime. with increasing the level of uncertainty.
Structural failures can develop at local level, for
example with the formation of a plastic hinge, or at
global level, as a consequence of the activation of a REFERENCES
set of plastic hinges leading to structural collapse. For
bridge systems the identification of the local failure Biondini, F., Bontempi, F., Frangopol, D.M., Malerba, P.G.,
2004. Cellular Automata Approach to Durability Anal-
modes and of their occurrence in time can represent a
ysis of Concrete Structures in Aggressive Environ-
crucial information to maintain a suitable level of per- ments, Journal of Structural Engineering,ASCE, 130(11),
formance and to avoid collapse over the structural life- 17241737.
time (Mori and Ellingwood 1994). In fact, repairable Biondini, F., Bontempi, F., Frangopol, D.M., Malerba, P.G.,
local failures can be considered as a warning of pos- 2006. Probabilistic Service Life Assessment and Main-
sible occurrence of more severe and not repairable tenance Planning of Concrete Structures, Journal of
failures. Therefore, the time interval between the first Structural Engineering, ASCE, 132(5), 810825.
local failure and structural collapse, or the elapsed Biondini, F., Frangopol, D.M., 2008. Probabilistic Limit
time between these failures, can be considered as a Analysis and Lifetime Prediction of Concrete Structures,
Taylor & Francis, Structure and Infrastructure Engineer-
measure of the required rapidity of the system to be
ing, 4(5), 399412.
repaired after first failure and represents an important Biondini, F., Frangopol, D.M., 2010. Structural robustness
performance indicator in a life-cycle design approach. and redundancy of deteriorating concrete bridges, Fifth
It is worth noting that the tolerance to local fail- Int. Conference on Bridge Maintenance, Safety, and Man-
ures can also be measured in terms of structural agement (IABMAS 2010), Philadelphia, PA, USA, July
redundancy (Frangopol and Curley 1987). However, 1014.
even though they are related concepts, elapsed time Frangopol, D.M., Curley, J.P., 1987. Effects of Damage
between failures and structural redundancy refer to dif- and Redundancy on Structural Reliability, Journal of
ferent system resources. In fact, structural redundancy Structural Engineering, ASCE, 113(7), 15331549.
Frangopol, D.M., Lin, K-Y., and Estes, A.C. 1997. Reli-
denotes the ability of the system to redistribute the load
ability of reinforced concrete girders under corrosion
after the occurrence of a local failure and does not pro- attack, Journal of Structural Engineering, ASCE, 123(3),
vide a measure of the failure rate, which depends on 286297.
the damage scenario and damage propagation mecha- Mori, Y., and Ellingwood, B.R., 1994. Maintaining reliability
nism. However, under this perspective the elapsed time of concrete structures. I: Role of inspection/repair. Journal
between failures can also be considered as a measure of of Structural Engineering, ASCE, 120(3), 824845.
system redundancy in terms of rapidity of evacuation.
282
Role of uncertainties on time dependent behaviour of prestressed

and cable stayed concrete bridges
G. Camossi, P.G. Malerba & L. Sgambi

ABSTRACT
The behavior of pre-stressed and of cable stayed con-

crete bridges may be strongly affected by the time
dependent response due to creep and shrinkage of
concrete. Same experiences showed us that, even
after accurate design assessments, some unsatisfac-
tory service behavior can arise, due to some level of
uncertainty in the reference data. In this paper the role
of the uncertainties has been investigated.
As sources of uncertainty, the pretensioning forces
in the cables, the relative humidity and the concrete
strength have been considered. For a given set of data,
the creep effects has been handled by means of the Figure 2. Prestressing time t0 = 3.5 days. Vertical displace-
A.A.E.M. Method. The effects of the uncertainties ments at the tip of the cantilever over time. (1) dead loads
have been simulated through a Monte Carlo approach. only; (2) prestressing only; (3) = (1) + (2).
Two topical cases have been studied: one less sensitive
to uncertain data, the other one strongly affected by
them. deviation with respect to the mean tension in the cables
In the case of a cable stayed bridge, made of a con- is relatively small and, most of all, it progressively
crete deck, subjected to creep effects and suspended to reduces. The system seems to be self stabilizing over
a set of pretensioned cables, the role of uncertainties time. On the contrary, in the case of a prestressed
in the pretensioning forces, in the relative humidity cantilever beam, the effects of uncertainties in the pre-
and in the concrete strength do not influence the time tensioning forces and in the concrete strength cause
dependent behavior sensibly (Figure 1). The standard a relevant variance of the tip deflections (Figure 2).
Both deflections and their variance increase with time
and they may strongly modify the vertical attitude of
the structure. These effects are emphasized when the
prestressing is applied few days after curing.
The achieved results and, in particular, those con-
cerning the prestressed cantilever beams, outline the
limits of the traditional deterministic analyses.
REFERENCES
Camossi G. 2009 (in italian). Analisi di strutture in regime
elastoviscoso in presenza di incertezze. Graduated thesis.
Politecnico di Milano.
Ghali, A., Favre, R. & Eldbadry, M. 2002. Cocrete Structures,
Figure 1. Effects of uncertainties in the initial pretension- Stresses and Deformation, London and NewYork: E & FN
ing of the stays. Tension variation in a stay over time. Spon.
(1) self-weight only; (2) pretensioning only; (3) = (1) + (2): Martinez y Cabrera, F., Malerba, P.G., Bontempi, F. & Bion-
mean value (thick line) and standard deviation from the dini, F. 1997. Creep Effects on Prestressed Cable Stayed
mean (thin lines). Bridges. Proceedings of the ICCCBE-VII. Seoul, Korea.
283
Variability of measured modal frequencies of a suspension bridge

under actual environmental effects
Y.L. Ding & A.Q. Li

Key Laboratory of Concrete and Prestressed Concrete Structures of Ministry of Education,
Southeast University, Nanjing, Jiangsu Province, China
F.F. Geng
Chengxian College, Southeast University, Nanjing, Jiangsu Province, China
ABSTRACT decrease with the increase of acceleration RMS.

it is found using the linear regression models of
This paper addresses the quantification of normal acceleration RMS-frequency that the maximum
modal variability of a suspension bridge making use and averaged relative variations of frequency
of long-term monitoring data. The output-only modal induced by traffic are 0.262% and 0.104%. On
frequency identification of the bridge is effectively the whole, the correlation between modal fre-
carried out using the Iterative Windowed Curve- quency and traffic loadings is very week and of
fitting Method (IWCM) in the frequency-domain. The no significance.
seasonal temperature-frequency correlation is formu- (4) The influence of wind speed on modal frequencies
lated using multi-sample averaging technique for each is also not significant. The modal frequencies are
vibration mode to eliminate the random variations ris- observed to slightly increase with the increase of
ing from the identification algorithm. Then the effect wind speed. It is found using the linear regres-
of temperature on the measured modal frequencies is sion models of wind speed-frequency that the
eliminated and the traffic-induced and wind-induced maximum and averaged relative variations of fre-
modal variability are quantitatively evaluated, respec- quency induced by wind speed are 0.215% and
tively. The analysis results reveal that: 0.163%. On the whole, the correlation between
modal frequency and traffic loadings is very week
(1) From the identified frequency sequences in a
and of no significance.
typical day, it can be observed that the mea-
(5) The modal variability caused by wind speed and
sured frequencies present instantaneous changes
traffic loadings is notably less than by temperature
because of the nonstationary properties of the
and identification algorithm. And the influence
ambient loadings. In order to eliminate the random
of identification algorithm on modal frequencies
variations rising from the identification algo-
is as significant as the temperature effect. Thus,
rithm, the daily averaged frequencies using multi-
for reliable performance of vibration-based dam-
sample averaging technique were obtained, i.e.,
age detection methods, the improvement of the
the seasonal correlation analysis of frequency-
modal frequency identification algorithm prop-
temperature. The maximum and averaged relative
erly considering the nonstationary properties of
variations of modal frequencies reduced from
the ambient loadings is very important.
3.013% and 2.633% measured at 10-min inter-
vals to 2.168% and 1.268% using multi-sample
averaging technique. REFERENCES
(2) Temperature is the critical source causing modal
variability, and there is an overall decrease in Ding, Y.L., Li, A.Q., Sun, J., et al., 2008. Experimental and
modal frequency with temperature for all the iden- analytical studies on static and dynamic characteristics
tified modes. A 6-order polynomial regression of steel box girder for Runyang Cable-stayed Bridge.
model is further applied for the modeling of daily Advances in Structural Engineering, 11(4): 425438.
averaged frequency and temperature. It is found Ni,Y.Q., Hua, X.G., Fan, K.Q., et al., 2005. Correlating modal
that the maximum and averaged relative variations properties with temperature using long-term monitoring
data and support vector machine technique. Engineering
of frequency induced by temperature are 1.975%
Structures, 27(12): 17621773.
and 1.175% for the variation of temperature Zhang, Q.W., Fan, L.C. and Yuan, W.C., 2002. Traffic-
covering a full cycle of in-service conditions. induced variability in dynamic properties of cable-stayed
(3) The influence of traffic loadings on modal fre- bridge. Earthquake Engineering and Structural Dynamics,
quencies is not significant as the temperature. 31(11): 20152021.
The modal frequencies are observed to slightly
284
Fuzzy reliability assessment of bridge piers in presence of scouring
S. Dordoni, P.G. Malerba & L. Sgambi

S. Manenti
Department of Hydraulics, University of Rome La Sapienza, Rome, Italy
ABSTRACT
The design of bridge piers placed in the riverbed has to

deal with variable hydraulic forces acting on structures
with foundations which may be prone to scour effects
(Da Deppo 1994, Ballio 1999). The conventional
design of drag forces, scour depths and foundation
bearing capacity is carried out by means of equa-
tions which involve a certain number of uncertainties
regarding flow, sediments, structural and geotechnical
parameters. A sound reliability assessment of bridges
piers against scouring risks must take into account the
uncertainties inherent in these parameters.
Due to the phenomenological nature of these uncer-
tainties, a fuzzy approach seems to be the most
adequate to deal with this kind of problems (Dordoni
1999). In the first part of this study the role of the
uncertain parameters is assessed by considering sepa-
rately their effects on the drag forces and on the general
and local scour evaluation. In the second part, the pre-
viously obtained responses regarding the drag forces
and the scouring depths are used as input data for a
fuzzy analysis of the pier foundation reliability.
Different design contexts are studied: embedded Figure 1. Schematic representation of the restriction and
foundations and foundations with exposed piles; piers positioning: front view (upper panel) and section of the
effects of clear water and live bed scour conditions; dif- bridge.
ferent ratios between vertical superimposed and lateral
hydraulic loads.
A sensitivity analysis allows to estimate the influ- the deterministic results. The results of the cases stud-
ence of the various uncertain variables on the partial ied (pier shaft exposed and foundation piles exposed)
and on the global responses. are summarized in Table 1.
The defuzzified output, expressed in terms of centre
of mass of output membership functions, assumes val-
ues that are noticeably different and more severe than REFERENCES
Ballio, F., Bianchi, A., Pranzetti, S., De Falco, F. & Mancini
Table 1. Percentage difference between the deterministic M. 1999 (in italian). Vulnerabilit idraulica dei ponti flu-
analysis and the result of the fuzzy assessment. viali. Atti del XXVI Convegno di Idraulica e Costruzioni
Idrauliche, Catania, 99(3): 6979.
Pier shaft Foundation piles Da Deppo, L. & Datei, C. 1994 (in italian). Le operei
exposed exposed drauliche nelle costruzioni stradali, Cosenza: Editoriale
Bios.
Hydraulic force +19% +46% Dordoni, S., 2009 (in italian). Ruolo delle incertezze nella
Scour depth +9% +13% progettazione delle pile da ponte in alveo. Graduated
Thesis. Politecnico di Milano.
285
Non linear finite element analysis of a 50 years old reinforced concrete

bridge: Comparison with experimental data and sensitivity analysis
S. Epaillard & B. Richard

Paris-Est University, LCPC, Paris, France
C. Cremona
Direction for research and innovation, MEEDDM, La Dfense, France
L. Elfgren
Lulea Technical university, Lulea, Sweden
ABSTRACT
In this paper, a non linear finite element analysis

of a full scale reinforced concrete bridge tested in
Ovik (Sweden) is presented. In order to quantify the
load carrying capacity of the bridge, a collapse load
test combining bending and shearing load effects was
performed within European Union founded project
Figure 2. Damage pattern obtained numerically for a
Sustainable Bridges Assessment for Future Traffic 100 mm mid-span displacement.
Demands and Longer Lives.
The first objective of the numerical simulations is A sensitivity analysis was also performed to quan-
to evaluate the robustness and the relevancy of the dif- tify the influence of local degradations on the global
ferent constitutive laws (concrete and steel/concrete mechanical response of the bridge. To achieve this
interface) to model large-scale structures subjected objective, a new steel/concrete interface constitutive
to complex mechanical loadings coupling bending model has been used to represent corrosion.
and shear up to failure. By using accurate damage Two aspects were investigated. The first one con-
mechanics based models, satisfying qualitative and cerns the effects of steel cross sections reductions
quantitative results have been obtained. Figure 1 pro- and the second one is related to the effect of bond
vides a quantitative comparison between the numerical strength variations. Due to the absence of corrosion in
results and the experimental ones. It leads to prove that the bridge reinforcement, results can only be corrobo-
the proposed finite element model of the reinforced rated with literature information: it comes out that it is
concrete bridge is relevant. nevertheless in good agreement with available results.
A satisfying damage pattern was also obtained
(figure 2). In particular, the cracks appear in shear REFERENCES
areas, in concordance with experimental results.
Guideline for Load and Resistance Assessment of Railway
Bridges. Prepared by Sustainable Bridges a project
within EU FP6, Available from: www.sustainablebridges.
net, 2007.
Ouglova A. 2004. Mechanical behaviour study of corroded
reinforced concrete structures (in French). PhD thesis,
ENS Cachan, Cachan, France.
Richard B., Ragueneau F., Cremona C., Adelaide L.H. &
Tailhan J.L. 2009. Isotropic damage model coupled to
frictional sliding to model the ultimate behaviour of rein-
forced concrete elements, ComplasX 2009, Barcelona,
Spain.
SB D7.3, 2007. Field test of a concrete bridge in
rnskldsvik, Sweden. Sustainable Bridges a EU FP6
project, Background Document D7.3 Concrete Bridge
Sweden, www.sustainablebridges.net.
Vu N.A. 2008. Study of the mechanical behaviour of rein-
Figure 1. Comparison between the experimental load/mid forcement concrete beams degraded by corrosion (in
span displacement and numerical results (proposed analysis French). Ph.D thesis. National Institute of Applied Sci-
and Sustainable Bridges numerical results). ences, Toulouse, France.
286
Field test and analysis of the dynamic factors for bridge in urban railway
transportation system
X.H. He & P. Li
School of Civil Engineering and Architecture, Central South University, Changsha, China
A. Scanlon
Department of Civil and Environmental Engineering, Pennsylvania State University, USA
ABSTRACT
Linear Induction Motor (LIM) system is a new urban

transportation system between maglev and traditional
rail transportation. In China, many cities have adopted
the Linear Induction Motor (LIM) rail transit sys-
tem, but specialized codes for design or assessment
of this type of vehicle loading in the urban rail-
way transportation system have not yet been estab-
lished. In this paper the dynamic factor formulations
for railway bridges adopted in various countries are
reviewed. Then, a typical bridge on the Guangzhou
metro line 4 is evaluated both experimentally and Figure 1. LIM vehicle-bridge vertical coupling vibration
theoretically to determine vehicle-bridge coupling model.
vibration characteristics. The vehicle is represented
as a whole vehicle model of secondary suspen- Table 1. Comparison of theoretical and experimental
sion with 6 degrees of freedom, and the bridge is dynamic factors.
modeled by beam elements. The coupled equation
is formulated using the principle of total potential Theoretical values
energy with stationary value in an elastic system and Velocity Experimental
solved by using Newmark- method. Field dynamical (km/h) Without EF With EF values
tests were also performed on the bridge. The cal-
culated and experimental displacement time-histories 40 1.046 1.052
50 1.058 1.068
for LIM trains crossing the bridge were obtained and
60 1.072 1.087
dynamic factors were developed. Through the theo- 70 1.068 1.075 1.0451.10
retical and experimental analysis for vehicle-bridge 80 1.066 1.073
coupling vibration, a formula for the dynamic factor
that can be used in bridge design and assessment is
proposed.
On the basis of analysis of electromagnetic force REFERENCES
(EF), the equation of LIM vehicle-bridge interaction
Akoussah, K.E., Fafard, M., Talbot, M., and Beaulieu, D.
is derived in terms of the principle of total potential 1997. Parametric analysis of the dynamic amplifica-
energy with a stationary value in elastic system dynam- tion factor of loads for simple-span reinforced-concrete
ics. A special dynamic analysis program based on the bridges. Canadian Journal of Civil Engineering, 24 (2),
MATLAB language of LIM vehicle-bridge interaction 313322.
was developed. Broquet, C., Bailey, S.F., Farad, M., Bruhwiler, E. 2004.
Because the LIM train was less axis weight, lower Dynamic behaviour of deck slabs of concrete road
moving speed etc, the DFs for design formula refer- bridges. Journal of Bridge Engineering, 9 (2), 137146.
enced railway codes or highway codes are shown to Cao, X. Q., and Zhu, J. L. 2003. Field test and analysis of the
be conservative. Adopting the proposed formula of impact factor of the vehicle loading on bridges. Urban
Mass Transit, (5), 3035. (In Chinese).
dynamic factor may be more reasonable for urban rail Chen, J. W., Xia, H., and Guo, W. W. 2006. An Analysis
transit bridge design. The proposed DF formula can on Dynamic Interaction of Linear Induction Motor(LIM)
be used in the development of bridge design criteria Train and Elevated Bridge System. Urban Rapid Rail
for urban rail transit. Transit, 19(1), 4448. (In Chinese).
287
A parametric study on the rocking behavior of bridge columns with spread

footing foundation
H.H. Hung & K.Y. Liu

National Center for Research on Earthquake Engineering, Taiwan
K.C. Chang
Department of Civil Engineering, National Taiwan University, Taiwan
ABSTRACT
Some spread footings of real engineering practices

of bridges in Taiwan were found to be uneconomi-
cally large due to the restriction of foundation uplift
regulated in design code. Although the rocking mode
of spread footings induced from foundation uplift is
not acceptable in general, some researches have indi-
cated that rocking itself can act as a form of isolation
mechanism. Uplift of foundation can limit the earth-
quake forces transmitted to the column base, hence to
decrease the plastic deformation occurring at the plas-
tic zone. Besides, unless very massive footing, some
uplift on the tension edge of spread footing during a Figure 1. Analytical model.
major earthquake can not be avoided. Thus, the anal-
ysis based on the assumption that the foundation and
soil are firmly bonded would not only lead to unrea- tension force and thus can take the uplift of footing into
sonable large internal forces in columns, sometimes it account automatically.
would also underestimate the disadvantage that may Using this analytical model, an extensive para-
be brought by rocking, such as large lateral displace- metric study was performed and the influence of
ments at deck and permanent settlement in soils. In this foundation size, soil stiffness and soil ultimate bear-
regards, realistically consider the effects of the interac- ing capacity on rocking behavior was calculated. From
tion between the foundation, column and soil becomes this analytical results, the isolation effect of rocking
an important issue. In order to gain better understand- spread foundation was identified. It also shows that
ing of rocking mechanism and consequently to have as the footing size and the underlying soil stiffness
more confidence in taking its benefit into consider- decrease, a better performance at the columns base
ation, an analytical model which can consider the can be obtained. However, it also results in an increase
nonlinear interaction between column, foundation and in displacement response at the decks level. On the
the underlying soil was established. other hand, when the yielding of the underlying soil
The analytical model established using Sap2000N is allowed to occur, the decrease in soil ultimate bear-
is schematically shown in figure 1. In which, column ing capacity can lead to minor plastic deformation at
at the plastic hinge zone was idealized by a 3D fiber column base. However, with the yielding of the soils,
beam-column element, while the other parts of the the level of permanent settlement in underlying soil
column were modeled by elastic beam elements. A increases, too.
lumped mass and point load were also added at the
top of the column to simulate the tributary mass and
the weight of superstructure. The foundation was mod- REFERENCES
eled by rigid bars with a lumped mass at the center of
FHWA. 2006. Seismic Retrofitting Manual for Highway
gravity. Figure 1 also shows that the vertical bearing Structures: Part 1-bridges, Reports No. FHWA-HRT-
resistance of foundation was modeled by an elastic- 05-032.
perfectly plastic spring with its yield points equal to Hung, H.H., Chang, K.C., Liu, K.Y. & Ho, T.H. 2008. A study
the ultimate bearing capacity of the underlying soils. on the rocking response of spread foundation. Report No:
Also, this zero-length nonlinear spring can not take the NCREE-08-040 (in Chinese).
288
Tests to collapse of masonry arch bridges simulated by means of FEM
T. Kaminski
Wrocaw University of Technology, Wrocaw, Poland
ABSTRACT
Modelling and analysis of masonry arch bridges is

still a current and complicated topic. No effective or
reliable method was developed so far which could
reasonably and accurately simulate behaviour of real
structures under the collapse loading. The reason for
this deficiency is an exceptional complexity of that
kind of structures. An alternative approach is pro-
posed in the paper which locally simplifies structure
of the bridge but provides satisfying representation
of its global behaviour. Characteristic feature of the Figure 2. Prestwood bridge at the end of testing (Page
method is application of the mezo-modelling tech- 1987).
nique, being between micro- and macro-modelling, in
discretization of the arches barrel. The arch is divided
into even segments which represent parts of masonry
consisting of several bricks or stones separated by indi-
vidual mortar joints. Details of the model are shown
in Figure 1.
The proposed model was applied to simulations
of tests to collapse of several full scale single-span
masonry arch bridges: 12 laboratory models erected
for the purpose of the testing and 5 real bridges
excluded from exploitation. One of the tested bridges
(Prestwood) during the test is presented in Figure 2.
The limit state of all the analyzed structures was
reached very easily during the calculations. An exam-
ple of the results in form of deformed mesh of a
selected structure at the limit state with map of plastic Figure 3. Deformed mesh of 3D model for Dundee bridge.
strains is shown in Figure 3. In the picture plas-
tic hinges (marked with lighter colour) indicate the
four-hinge failure mechanism possible to be simulated
by the model.
In all cases the calculated limit load value, mode of
failure as well as the location of the plastic hinges were
in a good agreement with the corresponding results
obtained from the experiments.
Apart from the satisfying results several other
advantages of the approach can be found, like: lim-
itation of the numerical problem size, simplification
of the FE mesh or effectiveness and ease in reaching
the limit state including development of the localized
cracking of material which is a common and serious
Figure 1. 3D FE model of a masonry arch bridge. difficulty in case of macro-models.
289
Dynamic evaluation of bridge mounted sign structures
D. Kim & A. Shroff

Jacobs, Inc, New York, NY, USA
ABSTRACT this instance, need to be variable depending on the

location and speed of the moving vehicles. In order to
The design requirement for ground mounted sign solve the equation of interaction, Bridge-vehicle inter-
structures are fairly well defined in the AASHTO action finite elements were developed using Modified
Standard Specifications for Structural Supports for Newmarks beta method. Using this element, full scale
Highway Signs, Luminaries, and Traffic Signals and 3D model of typical span of Gowanus Expressway with
consists of applying an equivalent pseudo-dynamic moving vehicle was then developed.
loading to account for the dynamic effects of wind The amplification factor of static loading and
loads and ignores the dynamic effect due to moving fatigue analyses for two dynamic loadings generated
vehicle loads. This design approach, however, should from moving vehicle and wind loads were studied
not be applied to the design of bridge mounted sign using this model. Dynamic wind load were gener-
structures because ignoring the dynamic effects of the ated using AASHTO Specifications noted earlier. This
moving vehicle loads may produce non-conservative procedure allowed meaningful computation of Fatigue
results, since the stiffness of the bridge structure can lives of each component of CCTV support structure.
greatly influence the behavior. Not enough informa- It was concluded that the bridge-vehicle interaction
tion is available in the literatures which provide guide finite element developed can provide a more accurate
lines to include the influence of moving vehicles in representation of the behavior of bridge mounted sign
the design of the bridge mounted sign structures. This structures.
paper describes a theoretical methodology that can The result of these analysis enabled development
be utilized to account for the dynamic effect of mov- of simple and effective retrofitting scheme for the
ing vehicles. A case study is also included where this existing CCTV support system
methodology was successfully applied.
In 2007, a Closed Circuit Television Camera
(CCTV) support system mounted on the superstruc- REFERENCES
ture of GOWANUS Express way in New York City AASHTO LRFD. 2004. Bridge Design Specifications, Amer-
failed after 9 years of service. The superstructure of ican Association of State, Highway and Transportation
GOWANUS Expressway consisted of steel pier bent Officials, Inc., Washington D. C.
with cantilevered cap-beams, steel stringers and 8 inch AASHTO. 2004. Standard Specification for Structural Sup-
concrete slab. The CCTV support attached to the rail- ports for Highway Signs, Luminaries and Traffic Signals.
ing consists of aluminum pole with cantilever arm Washington, D.C.
for the camera support at the top. This support sys- Anil K. Chopra. 1995. Dynamics of Structures. Upper Saddle
tem was subjected to the dynamic effects due to wind River, N.J.: Prentice Hall.
Arthur Huckelbridge. Dynamic Load Environment of
loads as well as moving vehicle loads on the bridge. Bridge Mounted Sign Support Structures. Paper pub-
In order to evaluate this behavior, a set of differen- lished by Case Western Reserve University.
tial equations of motion for the interaction between John W. Fisher. 1984. Fatigue and Fracture in Steel Bridges.
the bridge structures, moving vehicles and camera N.Y.: John Wiley & Son.
support system were derived. Unlike conventional Y.B. Yang. 2004. Vehicle-Bridge Interaction Dynamics. NY
equation of motions, the coefficients of equation, in World Scientific.
290
A special kind of analysis: CFD modeling for design and assessment

of bridge passive control devices
S. Manenti & F. Bontempi

Department of Structural and Geotechnical Engineering, University of Rome La Sapienza, Rome, Italy
P.G. Malerba
Department of Structural Engineering, Polytechnic of Milan, Milan, Italy
ABSTRACT
During its life-cycle a structural system is generally

subject to both static and dynamic loads: for the second
type of actions inertial effects become more impor-
tant resulting in dynamic amplification and cyclic
response.
The effects of such loads can be controlled for exam-
ple by adopting structural protective systems such
as seismic isolation, semi-active and active controls
and passive energy dissipation: the latter type can be
effective against seismic and wind induced motion
(Harris & Crede, 1976) requiring no kind of external Figure 1. Sketch of a viscous fluid damper with annular
orifice.
supply.
Among the different types of passive energy dis-
sipation systems, viscous fluid damper with annular of Non-Newtonian behavior on the velocity distribu-
orifice is deeply investigated in this work with the aim tion across the orifice is investigated by means of both
to set up a reliable numerical model supporting the ANSYS and ADINA codes.
design phase. Comparison of numerical and analytical results in
The operating principle (Fig. 1) consists of a viscous term of velocity and viscosity profiles across the orifice
fluid, filling both chambers inside a cylinder whose is carried out. Velocity vector plots at the orifice-
volume is defined by a one- or two-rod piston, that chamber interface obtained with both finite element
is forced, as the piston oscillates, to move back and models are compared and discussed.
forth through the gap between the piston head and the
cylinder pot: this causes high energy dissipation den-
sity due to the frictional stress produced by the viscous REFERENCES
fluid (Soong & Dargush, 1997).
Harris C.M., Crede C.E. (1976) Shock and Vibration Hand-
In the first part of the work a brief description book. 2nd Ed. McGraw-Hill, New York.
of the fluid-dynamics is provided from an analyt- Hou C.Y. Fluid Dynamics and Behavior of Nonlinear Viscous
ical point of view: the basic equations are solved Fluid Dampers. (2008) J. Struct. Eng. January p.5663.
considering an highly idealized situation in order to Jiuhong J., Jianye D., Yu W., Hongxing H. Design method
provide a reference solution; both Newtonian and for fluid viscous dampers. Arch. Appl. Mech. (2008) 78:
Non-Newtonian incompressible viscous fluid are con- 737746.
sidered (Wilkinson, 1960). Makris N., Constantinou M.C. (1991) Viscous dampers: test-
In the second part finite element modeling of the ing, modelling and application in vibration and seismic
problem is carried out: initially the idealized analyti- iso-lation. Technical Report NCEER-90-0028, National
Center for Earthquake Engineering Research, Buffalo NY.
cal model has been reproduced by means of ANSYS Soong T.T., Dargush G.F. (1997) Passive energy dissipation
code considering only the central portion of the annu- systems in structural engineering. J. Wiley & S.
lar conduit; subsequently an improved model is built Wilkinson W.L. (1960) Non-Newtonian Fluids. Pergamon
which includes a portion of the lateral chamber near Press.
the orifice outlet: influence of boundary effects and
291
Residual service life of existing railway bridges
A. Pipinato, C. Pellegrino & C. Modena

Department of Structural and Transportation Engineering, University of Padova, Italy
ABSTRACT on historical metal bridges, a deep analysis consisting

in a numerical study, based on previous experimental
The residual service life of existing railway bridges investigations (Pipinato et al. 2009a, 2009b), has been
is analyzed and assessed in this paper. Fatigue and carried out at the University of Padova, in collabo-
damage assessment are considered to be the most cri- ration with the railway national authority, RFI-Rete
tical topic issues in order to estimate the residual life Ferroviaria Italiana: non-linear Finite Element analy-
of these types of structures. In fact, with the aging ses have been carried out, giving some key issues on
of existing steel bridges and the accumulated stress the fatigue assessment of railway bridges. In particular,
cycles under traffic loads, assessment of remaining according to the experimental findings of Pipinato et al
fatigue life for continuing service has become more (2009a, 2009b) related to flexural and shear behaviour
important than ever, especially for decisions on struc- of some critical fatigue riveted details, a Finite Ele-
ture replacement, or retrofit interventions. Experience ment (FE) analysis, taking into account material non
from engineering practice indicates that fatigue analy- linearities due to plasticization of materials and con-
sis based on code provisions often underestimates the tact phenomena between the rivets and the plate and the
remaining fatigue life of existing bridges due to over- plates themselves, is developed to obtain some indica-
estimating live load stress ranges, and, at the same tions on the failure mode of the critical fatigue details
time, underestimating the stress history. In order to and accurate stress variations for the estimation of the
bridge the gap between common engineering practice residual fatigue life.
assessment of existing bridges and code provisions,
this analysis has been performed along with traffic
prevision.
Railway bridges represent a strategic part of an REFERENCES
ancient network and, in several cases, they have
already reached their traffic capacity limit. In this Pipinato A., Pellegrino C., Bursi O., Modena C (2009a). High
cycle fatigue behavior of riveted connections for railway
context, bridge condition state assessment and conse-
metal bridges. Journal of Constructional Steel Research
quently maintenance/replacement operations become DOI:10.1016/j.jcsr.2009.06.019
more and more necessary. The average age of sixty Pipinato A., Molinari M., Pellegrino C., Bursi O., Modena C.
percent of Italian railway steel bridges is about one (2009b). Fatigue tests on riveted steel elements taken from
hundred years as they were built between 1900 and a railway bridge. Structure and Infrastructure Engineering,
1920. In order to bridge the gap of a minor attention DOI: 10.1080/15732470903099776
292
Curved-tendon and fiber beam-column element for analysis of cable-stayed

PSC bridges
D.J. Seong & H.M. Shin

Department of Civil and Environmental Engineering, Sungkyunkwan University, Korea
J.R. Cho, K.T. Park & B.S. Kim

Structural Engineering & Bridge Research Division, Korea Institute of Construction Technology, Korea
ABSTRACT
The objective of this study is to develop time-

dependent fiber beam-column element and tendon
element, which are capable of predicting the long-
term behavior of cable-stayed PSC(prestressed con-
crete) bridges. For this purposes, analysis models
for finite element method were developed based on
the flexibility based fiber beam-column model origi-
nally proposed by Spacone et al. (1996). For the PSC
girders with pre-stressing tendon, additional curved
time-dependent tendon element cooperates with fiber
element. The performance of proposed analysis mod-
els is evaluated by comparing with the experimental
results in the literature. Finally, program simulated the Figure 3. Deflection of deck at various construction phase.
construction of the Pasco-Kennewick intercity bridge
in the state of Washington. Khalil et al. (1983) simulate this bridge with time-
dependent analysis but material was considered in
elastic range.
Most of bridge dimensions and material proper-
ties are borrows from that simulation. An attempt was
made to adhere to same concrete dimensions, cable
properties, prestressing information, and construction
phase with actual bridge, but some assumption regard-
ing time schedule, concrete age, and initial tension in
cables were made.
Figure 1. Half elevation of Pasco-Kennewick inter-city The largest difference between time-dependent
bridge. analysis and mechanical analysis was about 8,300 kN
m, 4,100 kN m, and 4,760 kN m at the approach span,
side span, and main span, respectively.
The simulation result of cable stayed bridge may be
helpful in determining the required camber or initial
tension in the cables during each construction phase
in order to achieve a desirable profile at final phase.
REFERENCES
Khalil, M.S., Dilger, W.H., and Ghali, A. 1983. Time-
dependent analysis of PC cable-stayed bridges. Journal
of structural Eng. 109(8): 19801996.
Spacone, E., Filippou, F.C., andTaucer, F.F. 1996. Fiber beam-
column model for non-linear analysis of R/C frames : Part
1. Formulation. Earthquake Engineering and Structural
Figure 2. Moment of deck at final construction phase. Dynamics, 25: 771725.
293
Remaining structural capacity of an early post-tensioned bridge verified

after removal
T. Vogel
Swiss Federal Institute of Technology (ETH) Zurich, Zurich, Switzerland
ABSTRACT Applying the lower bound theorem of plasticity

requires identifying an admissible stress state. For
In southern Switzerland a post-tensioned concrete haunched girders these calculations become quite
bridge (Fig. 1) built in 1952 showed signs of improper complicated.
grouting of the ducts and subsequent corrosion of the Applying the upper bound theorem of plasticity
prestressing steels. That is why a monitoring system means identifying the relevant failure mechanisms.
was installed to detect wire breaks based on acous- Changing the values of loads i.e. external work and
tic emission analysis. The monitoring revealed a wire resistances i.e. dissipation is an easy task and the same
break about every 35 days and the owner decided to applies for extrapolations into the future.
prop the bridge by additional piers and to replace it The calculations have shown that the partial safety
by a new one. The demolition of the bridge was used concept is not adequate for this kind of questions,
to establish a detailed condition survey of both, the because design values have no physical meaning. For
concrete and the reinforcement, especially the ten- material properties and resistance models, characteri-
dons. Like this, excellent data is available on the actual stic values may be applied, but for permanent and
condition of the bridge prior to removal. This infor- traffic loads expected values for the considered sce-
mation was used to calculate the structural capacity nario should be used. To account for the probabilistic
of the bridge (Fig. 2) and possible failure scena- aspect of structural safety, it is more feasible to con-
rios depending on applied traffic loads and ongoing sider the frequency of an expected load scenario rather
deterioration. than the probabilistic distribution of the action force.
Monitoring the change of condition parameters like
detecting wire breaks by acoustic emission analysis is
a good tool for early warnings. Since normally mon-
itoring is installed in an advanced aging stage of the
structure, the initial condition is not known and inte-
grating measured condition changes does not lead to
a reliable result.
Figure 1. Longitudinal section of the bridge. REFERENCES (EXTRACT)

Fricker S. & Vogel T. 2007a. Site installation and testing with
the acoustic monitoring system SoundPrint. Final report
No. 609; research project AGB 2002/009. Zurich: Swiss
Association of Road and Transportation Experts (VSS).
Fricker S. & Vogel T. 2007b. Site installation and testing
of a continuous acoustic monitoring. Construction and
Building Materials 21: 501505.
Fricker S., Vogel T., Ungricht U., Hunkeler F. Acoustic mon-
itoring of a prestressed concrete bridge and its condition
survey during demolition. Final report; research project
AGB 2005/014. Zurich: Swiss Association of Road and
Transportation Experts (VSS), in preparation.
Figure 2. Bending capacity (left span only).
294
An analysis of long time deflection of long span prestressed

concrete bridges
L. Vrablik & V. Kristek

CTU in Prague, Faculty of Civil Engineering, Department of Concrete and Masonry Structures
ABSTRACT
The results of the presented solutions and the deve-

loped analytical and design methods will help creating
the sufficient theoretical tools for reliable and eco-
nomic structural design bridges without deflection
impairments. The paper reviews the causes of under-
estimation of the long-time future deflections. Special
attention is directed to prestressing a procedure to
find the optimal arrangement of tendons layout is pre-
sented, allowing avoiding the tendons contributing to
deflection increases. Developed computer program is
freely available and a method of repair procedure and
a repair example, as well as recommendations for the
design practice are also presented.
The design of structures is more and more directed
towards the entire lifetime design with multiple con-
current objectives. Apart from durability, the most
important factor in the whole life design of rein-
forced and, in particular, prestressed concrete bridges,
is the Service Limit State. From this point of view,
prestressed concrete bridges are very sensitive to long-
term increase of deflections. This phenomenon has
Figure 1. Bridge with excessive deflections after reparation
paramount importance for serviceability, durability process.
and long-time reliability of such bridges.
Due to excessive deflections, several bridges had deceived engineers monitoring the deflections. A sur-
to be either closed or repaired well before the end of vey of many bridges monitored in various countries
their initially projected lifespan. The cost of reduced showed that all of them have experienced similar
service life of structures is tremendous for society, the deflection histories. It has frequently been experi-
owners and users. In fact, it greatly exceeds, in strictly enced that the box girders of many prestressed con-
economic terms, the cost of catastrophic failure due to crete bridges deflected far more than predicted in
mispredicted safety margin. design. The deflection evolution has often been coun-
Reliable prediction of bridge deflections during terintuitive, with slowly growing deflections in the
their construction as well as during their service life early years, followed later by a rapid and excessive
is of crucial importance for achieving good durabi- deflection growth.
lity and long-term serviceability. Obviously, difficulty
of predicting deflections is closely related to the
properties of concrete (strength, elastic modulus, non- REFERENCES
linearity, creep, shrinkage, etc.), both initially and with Kristek, V., Vrablik, L, Bazant, Z.P., Li, G., Yu, Q., 2008.
time. Misprediction of Long-Time Deflections of Prestressed
The long-term deflection behaviour of long-span Box Girders: Causes, Remedies and Tendon Layout Effect,
prestressed concrete box girder bridges has often Proc. of Concreep8.
295
Reliability-based assessment of RC bridges in a marine environment

considering spatial and temporal variability of deterioration processes
J.S. Zhu
Tianjin University, Tianjin, P.R. China
D.M. Frangopol
ABSTRACT bridges. Monte Carlo simulation is used to compute

the probability of violating various limit states.
Durability analysis of deteriorating structures under For the sake of illustrating the proposed model and
uncertainty has to be based on structural reliability method, an external girder of a typical RC bridge is
concepts and methods. In this paper, a stochastic considered herein. The results from the space- and
finite element computational framework for reliability time-dependent reliability analysis of the girder are
assessment of deteriorating reinforced concrete (RC) presented in terms of serviceability and strength, for
bridges is proposed. Spatial and temporal variability of different concrete covers and overload factors. This
the deterioration due to corrosion under uncertainty is information can be used to predict the reliability and
considered. The analysis procedure considers the dete- residual life of deteriorating RC bridges.
rioration processes of concrete, steel and bond between
concrete and reinforcing bars.
Nonlinear models for steel, concrete and bond are
used to describe the behavior of deteriorating RC REFERENCES
bridges. The concrete is modeled by means of four- Akiyama, M., Frangopol D.M. & Suzuki M. 2010. Integra-
node plane-stress elements, while the steel bars are tion of the effects of airborne chlorides into reliability-
represented by two-node truss elements. The bond- based durability design of reinforced concrete structures
link element exhibits a relative slip between the two in a marine environment. Structure and Infrastructure
materials. Engineering., doi:10.1080/15732470903363313.
A computer code written in MATLAB version Enright M.P. & Frangopol D.M. 1998a. Probabilistic anal-
R2006a is used to generate the random variable inputs ysis of resistance degradation of reinforced concrete
and to conduct the reliability analysis. The com- bridges beams under corrosion. Engineering Structures
20(11):960971.
mercially available software ANSYS performs the Enright M.P. & Frangopol D.M. 1998b. Service-life pre-
nonlinear finite element analysis of deteriorating RC diction of deteriorating concrete bridges. Journal of
Structural Engineering 124(3):309317.
Frangopol D.M. & Liu M. 2007. Maintenance and manage-
ment of civil infrastructure based on condition, safety,
optimization, and life-cycle cost. Structure and Infrastruc-
ture Engineering 3(1):2941.
Frangopol D.M., Lin K-Y. & Estes A.C. 1997. Life-cycle cost
design of deteriorating structures. Journal of Structural
Engineering 123(10):13901401.
Marsh P. S. & Frangopol D. M. 2008. Reinforced con-
crete bridge deck reliability model incorporating tempo-
ral and spatial variations of probabilistic corrosion rate
sensor data. Reliability Engineering and System Safety
93(3):394409.
Val D. V. & Stewart M. G. 2009. Reliability assessment of age-
ing reinforced concrete structures-current situation and
future challenges. Structural Engineering International
19(2):211219.
Zhu J.S., Xiao R.C. & He L.Z. 2007. Reliability assess-
ment of large-span cable-stayed bridges based on artifi-
Figure 1. Probability density functions for live load deflec- cial intelligence. China Civil Engng. J. 40(5):4147. (in
tion at mid-span at different bridge ages. Chinese).
296
MS13: Management & maintenance of long span bridges
Organizers: A. Chen & D.M. Frangopol
Anticorrosive coating and maintenance of suspension bridges in China
X. Gongyi & H. Yanqing

China Railway Major Bridge Reconnaissance & Design Institute Co., Ltd., P.R. China
Z. Junhui
Beijing Institute of Aeronautical Materials, P.R. China
ABSTRACT
The main cables and hangers of suspension bridge, cor-

rodes easily due to exposure in the sun, the rain, and
repeatable stress combination. Its anticorrosive coat-
ing is of great importance among maintenance work.
Based on research and study of anticorrosive coating
development of suspension bridges home and abroad,
this paper summarizes the anticorrosive coating
schemes and maintenance, repair and service practice
of suspension bridges in China, and proposes coating
system suitable for bridge in China. It can be referred
to in main cable protection and maintenance scheme.
The anticorrosive coatings applied in China include:
non-drying anticorrosive paste+ wrapping steel
wire+ paint, non-drying anticorrosive paste+ wrap-
ping steel wire+ rustproof paint+ polysulfide sealant+
Figure 2. The Shantou Bay Bridge, China.
finish, and the compressive dehumidification sys-

tem introduced from Japan. The most widely appli-
cation is phosphating rustproof paint+ non-drying
polyisobutylene sealing paste+ wrapping steel wire+
phosphating rustproof paint+ polysulfide rubber
sealant+ polyurethane (epoxy)finish.
REFERENCES
Tatsuya et al. 1999. Development of Corrosion protection
Methods Using S-shaped Wire Wrapping System. Nippon
Steel Technical Report. No.82 November 1999. 814.
John Sun P.E. et al. 2004. Suspension Cable Design of
the New San Francisco- Oakland Bay Bridge. Journal
of Bridge Engineering. (ASCE/JANUARY/FEBRUARY).
Figure 1. The Shantou Bay Bridge. 101106.
299
Integrity management of concrete bridges using Spatial Information

Systems
J. Qin & M.H. Faber

ETH Zurich, Zurich, Switzerland
ABSTRACT The general system representation in the risk

assessment is adapted from Straub et al. (2009) and
Risk assessment forms a cornerstone in the life-cycle Malioka (2009) utilizing a hierarchical Bayesian tem-
integrity management of concrete structures subject to poral and spatial modeling approach. In regard to the
corrosion. To formulate the required risk assessment visual representation and management of spatially dis-
models, it is necessary to account for all available expe- tributed data very little literature is available as of
rience as well as data of relevance for the modeling present. Visual condition mapping has been reported in
of the deterioration processes. In the context of life- Wawrusiewicz (2007). However, for what concerns the
cycle integrity management concrete bridges represent integration of all model components as well as data and
large-scale systems and this necessitates that risk mod- information required to support risk informed decision
els account for the system effects and that such models making in regard to optimal life-cycle integrity man-
together with relevant input data and results may be agement of concrete structures, little or no research
managed efficiently. The present paper presents a Spa- has been reported as of yet. In the present work the
tial Information System (SIS) which facilitates on the approach to the spatial risk modeling is thus adapted
one hand a consistent representation of system depen- from Bayraktarli et al. (2006).
dencies in the modeling of corrosion deterioration of Finally, to illustrate the capabilities of the developed
concrete structures and on the other hand the efficient framework and SIS tool an example is provided where
and intuitive management of the required input data deterioration risk assessment and maintenance strat-
and the analysis results. egy optimization for the Far bridge are considered.
The main aims with the proposed framework and
tool are the following:
To develop a probabilistic model of the deterioration
for an entire large concrete structure taking into ac- REFERENCES
count the most relevant uncertainties and dependen-
Bayraktarli Y.Y., Yazgan U., Dazio A., & Faber M.H.
cies in the exposures (environmental conditions), 2006. Capabilities of the Bayesian probabilistic networks
the material characteristics and the model uncer- approach for earthquake risk management. Proceedings
tainties associated with the applied deterioration ECEES First European Conference on Earthquake Engi-
models. neering and Seismology, 38 September 2006. Geneva,
To account consistently for the uncertainty associ- paper No. 1458.
ated with the performed inspections as well as the Malioka, V. 2009. Condition Indicators for the Assessment of
spatial and temporal allocation of inspections. Local and Spatial Deterioration of Concrete Structures.
To facilitate a practical applicable framework for the Zurich: ETH Zurich.
Straub, D., Malioka, V., & Faber, M. H. 2009. A framework for
visualization and management of input data to the
the asset integrity management of large deteriorating con-
developed models, inspection data, repair activities crete structures. Structure and Infrastructure Engineering
and analysis results. 5(3): 199213.
To facilitate that risk assessments and maintenance Wawrusiewicz, A. 2007. Multi Mapping in Forensic Engi-
management strategies may be easily updated on the neering. Structural Engineering International 17(2):
basis of inspection results. 172177.
300
Research on structural performance evolution of a long span concrete bridge
H. Tian & A. Chen

Department of Bridge Engineering, Tongji University, Shanghai, China
ABSTRACT
Structural performance evolution analysis is one of

critical problems in safety assessment of concrete
bridges in life-cycle. For the purpose of quantitative
analysis on structural performance evolution of long
span concrete bridges, numerical simulation meth-
ods of several routine mechanical problems and some
degradation mechanical problems are firstly solved.
Then an integrated computational methodology for
simulation of carbonation- or chloride-induced degra-
dation process of concrete bridges is proposed and an
analysis program named Concrete Bridge Durability
Analysis System (CBDAS) is compiled by FORTRAN
95. Finally, simulation results are presented for a
long span prestressed concrete continuous rigid frame Figure 1. Integration of CBDAS.
bridge exposed to concrete carbonation environment.
The mainly research works are as follows: of a 83 + 150 + 150 + 83 m prestressed concrete
(1) Simulation of complex mechanical problems continuous rigid frame bridge exposed to con-
According to mechanical features of routine struc- crete carbonation environment. The degradation
tural analysis, simulation of prestressing effect, process of various mechanical property indices
concrete creep-shrinkage effect and structural sys- on key sections are given respectively in service-
tem transformation in construction process are ability limit state and ultimate limit state. The
briefly introduced. According to mechanical fea- results show that: for prestressed concrete bridges,
tures of durability structural analysis, simulation degradation of mechanical property indices in
of deterioration of materials mechanical prop- serviceability limit state should be paid more
erties, loss of various materials section and the attention to.
evolution of structural mechanical performance
are mainly solved.
(2) Concrete Bridge Durability Analysis System REFERENCES
A computational methodology for simulation
Richardson, MG. 2002. Fundamentals of durable reinforced
of carbonation- or chloride-induced degrada- concrete. London: Spon Press.
tion process of concrete bridges is proposed Tian, H. 2009. Research on Performance Evolution of Con-
through integrating the above theoretical analysis crete Bridges in Given Service Life[doctor dissertation].
on specific simulation methods of several rou- Shanghai: Tongji University.
tine mechanical problems and some degradation Li G.P. 2009. Design Principles of Prestressed Concrete
mechanical problems. According to the proposed Structures. Beijing: China Communications Press.
finite element-based method an analysis program JTG D62-2004. 2004. Code for Design Reinforced Concrete
named Concrete Bridge Durability Analysis Sys- and Prestressed Concrete Bridges and Culverts. Beijing:
China Communications Press.
tem (CBDAS) is compiled by FORTRAN 95.
JTG D60-2004. 2004. General Code for Design of Highway
CBDAS totally includes eight subsystems that are Bridges and Culverts. Beijing: China Communications
shown in Figure 1. Press.
(3) Numerical example CECS 220:2007. 2007. Standard for Durability Assessment
The CBDAS is implemented to simulate struc- of Concrete Structures. Beijing: China Architecture &
tural performance evolution in given service life Building Press.
301
Example of bridge maintenance program applied by toll bridges agencies

in Europe: The Rion-Antirion Bridge
V. Stathopoulos
Gefyra, Greece
G. Hovhanessian
Advitam, France
B. Kroely
Advitam Inc, Sterling, VA, USA
ABSTRACT water detection system in the piers, 60 m under the

sea level.
Located 250 km west of Athens in Greece, the The acquisition system on the bridge is composed of
Charilaos Trikoupis bridge, also called Rion-Antirion four independent acquisition units, which are power-
Bridge is a 4-pylon cable-stayed structure. Its 2,252 ful real-time data processing computers, one for each
meter continuous suspended deck is the longest cable span. It digitalises the analogue signal at 500 Hz and
stayed deck in the world and its reference span of 560 provides the necessary data to structural engineers.
meters ranks in the world top 10 longest cable stayed Various types of files are created; dynamic ones with
spans.The Concessionaire, Gefyra S.A. is responsible 100 Hz sampling in case of an alert, automatic ones
for the design, construction, financing, maintenance at a specific time interval or even after request, and
and operation of the bridge during the 35-year con- finally historical ones with lower sampling frequency.
cession period. Therefore, Gefyra decided to work The software is parametric and modifications on the
with Advitam to build methodologies and system to sampling, duration and so on are possible if required
optimize structural inspection & maintenance process. for more specialised post-processing and analysis. The
The structural health monitoring system (SHMS) of SHMS has been providing valuable information and
the bridge, which was initially developed to follow the data to structural maintenance engineers and design-
long term structural deteriorations and improve main- ers since 2004, concerning the long term evolution of
tenance, is now used for real-time earthquake manage- the structure and the impact of special events.
ment as well. The SHMS of the Rion-Antirion Bridge The recent development of the system is based on
was initially developed through a risk analysis process the automatic analysis of earthquake data. A deci-
performed during the design phase. The inspection and sion tree for earthquakes was developed, based on the
maintenance manual was issued in accordance with the ground acceleration intensity and on other structural
results from the risk analysis. This provides the guide- parameters, which specifies all the immediate actions,
lines for all inspection and maintenance activities on for example the level of required inspection, and the
the bridge and in order to complete the surveillance, need to interrupt the traffic. This principle has been
the SHMS was set up to provide engineers with real- implemented in the SHMS in order to provide the
time data, as well as with other information which is operator with an automatic decision-making system.
unobtainable from visual inspection. Based on the knowledge brought by the SHMS
The structural health monitoring system was since 2004 on the dynamic Bridge behaviour, Gefyra
designed, supplied and installed by Advitam and and Advitam will continue to develop and implement
it consists of hundreds of sensors. These include new functionalities, including a real-time wind man-
accelerometers on the free-field ground and pier agement system, and a general event detection and
bases to evaluate earthquakes intensity and bridge management system called Smart Monitoring.
response, and accelerometers on stay cables, towers The objective of the latter is to minimise the vol-
and deck to address the dynamic effects. Load cells ume of redundant data and store data with a specific
on the stay cables monitor their tension, while there sampling duration for special post-processing and
are deformation transducers on the expansion joints, analysis.
meteorological stations on the deck and automatic
302
Construction stages simulation for cable-stayed bridge on geometrical

nonlinear analysis
J. Wu
Tongke College, Tongji University, Shanghai, China
Q.-L. Zhang, X.-Q. Luo & L. Kang

Department of Building Engineering, Tongji University, Shanghai, China
ABSTRACT
There are three main sources of geometrically non-

linear behaviour of cable-stayed bridges: the cable
sag effect, the beam-column effect and the large dis-
placements effects. In this paper the cable sag effect
is considered by the multiple-straight truss elements
and the stress stiffness matrix is presented, the beam-
column and the large displacements effects are con-
sidered using geometrical nonlinear analysis based on
the Updated Lagrange method.
Stress stiffness matrix of cable elements contains
unstress length of cables. Repeatedly stretching of
cables can be used at the construction of the cable-
stayed bridge for the purpose of improving structural
mechanical behavior. The unstess length of cables may
change at each stage. Supposing the tension force Figure 1. Three activated method of new elements.
applied on cable is P1 , The unstess length can be
achieved by Equation below:
stages analysis is performed with parallel model firstly,
the incremental displacement, U , is obtained, the old
node position is gained by subtract U from original
nodal coordinate and the newly activated node coordi-
nate is the initial model coordinate which is shown in
where, l1 is the distance of the two nodal coordinates Figure 1c. The nonlinear iteration is performed until
of the cable element, E and A are theYoung modulus of the convergence tolerance is achieved.
the material and the cable cross section, respectively. The program for construction stages simulation
The geometric configuration of the old elements and for cable-staged bridge, Binas software, is compiled,
nodes is changed during the construction analysis of the computational accuracy and the effectiveness are
cable-stayed bridge. The node positions of the new ele- proven by the examples of the cantilever beam with
ments can be gained by three methods: tangent to old cables and the Sutong bridge at the end of this paper.
elements, parallel to old elements and original model
coordinate, which are shown in Figure 1, in which
the dashed lines are the initial model positions. The REFERENCES
construction stages analysis for cable-stayed bridge is
performed with the tangent model which is shown in Freire AMS, Negro JHO. 2006. Geometrical nonlinearities
Figure 1a or the the parallel model which is shown in on the static analysis of highly flexible steel cable-stayed
bridges. Computers & Structures 84:21282140.
Figure 1b.
Wang PH, Ling HT. 2002. Study on nonlinear analysis of
The purpose of construction control analysis of a highly redundant cable-stayed bridge. Computers &
cable-stayed bridge in this paper is that the archi- Structures 80:165182.
tectural configuration should be satisfied after con- Wang PH, Tang TY. 2004. Analysis of cable-stayed bridges
struction control analysis is finished. The theory of during construction by cantilever methods. Computers &
construction control analysis is that construction Structures 82:329346.
303
MS14: SmartEN Marie Curie ITN Smart management for
sustainable human environment
Organizers: T. Onoufriou & R. Helmerich
Use of sensors for efficient design of innovative seismic

protection techniques for monuments
C.Z. Chrysostomou, R. Votsis & N. Kyriakides

Cyprus University of Technology, Cyprus
Th. Demetriou
Structural Engineer, Cyprus
A. Stassis
Higher Technical Institute, Cyprus
ABSTRACT
One of the most difficult problems in the design of seis-

mic protection for monuments is the establishment of
their in-situ properties in order to be able to use them
for an as accurate as possible model of the structure.
In addition, any retrofitting method that will be used
should be inconspicuous, and it should neither vio-
late its form, nor change its structural behavior. The
application of SMA and other innovative devices in
protecting monuments are reported by Croci (2000)
and Chrysostomou et al. (2008).
In this work the in-situ dynamic characteristics of
an aqueduct (Fig. 1) are obtained using accelerome-
ters to record the response of the structure to ambient
vibrations, and hence finite element models which are Figure 2. Frequencies for 4 SMA wires.
updated based on those measurements are developed.
The dynamic characteristics of the aqueduct were it is explained, there was a considerable difference in
determined twice: the first in June 2004 and the sec- the behavior of the monument between those two mea-
ond in May 2007. The need for this arose from the surements. The main contributing factor was the effect
fact that the application of the SMA devices on the of the level of the water in the nearby salt lake. In June
structure was performed three years after the char- 2004 the lake was dry while in 2007 the water level
acteristics of the structure were determined, and it was high and it was obvious that the foundation of the
was therefore considered prudent to redetermine those structure was under the ground water table.
characteristics so as to confirm the starting point. As The effectiveness of the use of shape-memory-alloy
wire-dampers for the protection of the monument
from earthquakes is evaluated through their applica-
tion on the monument and in-situ measurement of the
changes in its dynamic characteristics (Fig. 2). It has
been shown that the application of the SMA wires
on a real structure changes significantly the dynamic
characteristics of the structure.
REFERENCES
Chrysostomou, C. Z., Stassis, A, Demetriou, Th. and Ham-
daoui, K.. 2008. Application of shape memory alloy pre-
stressing devices on an ancient aqueduct, Smart Structures
and Systems, v. 4, No. 2: 261278.
Croci, G. 2000. General methodology for the structural
restoration of historic buildings: the cases of the Tower
of Pisa and the Basilica of Assisi, Journal of Cultural
Figure 1. Larnaca aqueduct. Heritage, 1, 718.
307
In-node data processing in structural monitoring with wireless

sensor networks
G. Feltrin, J. Meyer, R. Bischoff & O. Saukh

Empa, Swiss Federal Laboratories of Materials Testing and Research, Duebendorf, Switzerland
ABSTRACT Table 1. Estimations of selected natural frequencies of cable

C1 to C6 in [Hz]. M1 and M2 are measurements made with
The main advantages of wireless sensor networks a wired instrumentation.
are fast deployment, little interference and self- Cable Mode M1 M2 WSN Error (%)
organization. These advantages come into play if an
unattended operation of the network can be achieved C1 3rd 3.91 3.85 3.91 0.06 0.77
for a reasonably long period of time. Since wireless C2 3rd 4.38 4.35 4.37 0.05 0.11
sensor networks are battery powered, a power sav- C3 3rd 5.06 5.02 5.06 0.03 0.40
ing operation is a key issue. Power consumption is C4 3rd 6.55 6.58 6.58 0.05 0.23
reduced by using low power hardware, energy efficient C5 3rd 12.03 11.85 11.91 0.19 0.50
communication protocols and duty cycling. C6 3rd 7.77 8.00 7.78 0.14 1.33
However, in data intensive applications, e.g. vibra-

tion based monitoring, these technologies are not
sufficient for achieving a sustainable system life-
time. Significant power saving is achieved by in-node
data processing with the goal to reduce dramatically
the data that is communicated over a wireless chan-
nel. This data reduction, however, is a challenging
task, since it has to be performed with very limited
computational and memory resources.
This paper investigates the capabilities of this con-
cept with a long term field test on a cable-stayed bridge.
The goal of this deployment was to monitor the tension
force of cable-stays via natural frequency estimations
(Fig. 1). The paper shows that the data processing can
be performed despite the significant restrictions pro-
vided by the low power hardware. Although the raw
data is based on 12 bit data acquisition and the algo-
rithms are performed with 16 bit operations, the results
are accurate (Table 1). Furthermore, in-node data pro-
cessing can be performed with nearly negligible power
consumption. This allows, depending on the measure-
ment interval, to reach system lifetimes of a year or
more with a set of batteries.
REFERENCE
Feltrin, G., Meyer, J., Bischoff, R., & Motavalli, M. 2009.
Wireless sensor networks for long term monitoring of civil
Figure 1. a) Stork Bridge in Winterthur, Switzerland. structures. Structure and Infrastructure Engineering: DOI:
b) Locations of the network nodes. 10.1080/15732470903068573.
308
Capabilities of non-destructive testing of RC structures quality

assurance for crack repair using ultrasonic echo
M. Friese, U. Effner, F. Mielentz & R. Helmerich

BAM Federal Institute for Materials Research and Testing, Department of Nondestructive
Testing and Environmental Measurement Methods, Berlin, Germany
ABSTRACT
Cracking is a common phenomenon in concrete struc-

tures. Depending on their width and depth, cracks can
significantly reduce the durability of the structure. To
prevent the loss of durability and premature failure, the
cracked structural member needs to be repaired. Sev-
eral repair techniques can be applied for this purpose.
One of the most common repair methods is epoxy,
polyurethane or cement injection into the crack.
Having reliable quality control measures is of great Figure 1. Results obtained at a specimen with cracks before
importance to ensure the effectiveness of the repair and after injection with epoxy. 1: SH surface waves, 2: back-
procedures. So far, taking cores has been the only side reflection, 3: artifacts caused by corner reflector effects,
4: multiple backside reflection
way to get reliable information about the quality of
injection.
Experiments have proven that classical non-
crack/notch was filled with cement or epoxy. However,
destructive testing methods to assess cracks, such as
ultrasonic shear waves proved to be unfavourable for
ultrasonic time of flight measurement dont work well
testing cracks filled with polyurethane.
for concrete. Rebars and contact between the crack
The SAFT image-based data analysis revealed a big
surfaces (e.g. caused by grain) can act as acoustic
advantage over the single point measurements in inter-
bridges which makes the interpretation of the data
pretation of the test results. The SAFT reconstruction
more difficult.
process includes many single measurements to calcu-
In the past, best results were achieved by means
late the image of the investigated object. In most cases,
of applying two dimensional surface scanning with
it was possible to distinguish between real indications
an ultrasonic transmitter at one side of the crack and
of the crack and the spurious indications (e.g. caused
a scanning laser vibrometer at the other side, fol-
by edge effects) in the reconstructed image.
lowed by three dimensional reconstruction (3D-SAFT,
Significant differences between empty and filled
Synthetic Aperture Focusing Technique) of the data
cracks could be observed. In a number of cases, the
(Mielentz et al. 2001).
results indicated that the cracks were not completely
Based on these experiences, a new practice ori-
filled. These indications were verified by ground-truth
entated method was developed which uses advanced
data. In summary, the outcome of this project clearly
array techniques (linear array, multistatic array) with
demonstrated the potential for future developments
dry-point-contact shear wave transducers for data
of a non-destructive quality control system for crack
acquisition and 3D-SAFT reconstruction.
repair with cement or epoxy.
To investigate the application of ultrasonic echo
in evaluating the effectiveness of crack repairs, a set
of laboratory measurements was carried out. Several REFERENCE
commercially available ultrasonic devices as well as
an in-house system developed at BAM were used Mielentz, F., Milmann, B., Krause, M. und Mller, W.
to examine composite specimens, specimens with 2001. Zerstrungsfreie Risscharakterisierung in Beton-
notches and specimens with cracks. Specimens with bauteilen mit Ultraschall. in: DGZfP (Hrsg.); Fach-
notches or cracks were investigated before and after tagung Bauwerks-diagnose Praktische Anwendungen
filling/ injecting the notch/crack. Zerstrungsfreier Prfungen, 2001, Leipzig, DGZfP-
Significant differences between empty crack/ notch volume 76 on CD, Berlin.
and the filled crack/ notch could be observed, if the
309
Reference-free health monitoring system using chaos theory
H. Furuta
Y. Nomura
Ritsumeikan University, Shiga, Japan
K. Koichiro, H. Hattori & Y. Teramae

ABSTRACT
In this study, an attempt is made to develop a refer-

ence free health monitoring system that can detect
the damage with ease, using the data collected from
the vibration measurement. In order to investigate the
damage state, it is, in general, necessary to measure
the intact state of the structure in advance. However,
there are many existing structures which have no such
data. Thus, it is very useful to develop a new damage
detection method without using the reference data. The
proposed system can identify the location and intensity Figure 2. The results of Difference of the spread of local
of damage without using the baseline reference data. area of attractor.
Based upon the chaos theory, it is possible to perform
the damage detection by paying attention to the chaos is given to the structure against the direction of x axis.
characteristics of the structure of response displace- The standard deviation of the Difference of Spread of
ment wave. This study considers a simple symmetric Local Area of Attractor (DSLAA) obtained at the 4th
structure shown in Figure 1 and Table 1 as a struc- story is the largest among other stories. Besides the 4th
tural model. This study investigates the damage cases story, remarkable difference of the standard deviation
in which the 20% deterioration occurs below the node of DSLAA is not confirmed. From these trends, it is pre-
14. Figure 2 is the results in which chaotic excitation dicted that the 4th story has something abnormal. By
evaluating the standard deviation of DSLAA, damaged
story can be detected without using any reference
information.
REFERENCES
Nichols, J.M., Todd, M.D., Seaver, M. & Virgin., L.N. 2003a.
Use of chaotic excitation and attractor property analy-
sis in structural health monitoring, Physical Review E,
67:016207 18.
Nichols, J.M.,Virgin, L.N.,Todd, M.D. & Nichols, J.D. 2003b.
On the use of attractor dimension as a feature in struc-
tural health monitoring, Mechanical System and Signal
Figure 1. Structural model. Processing, 17(6):13051320.
Nichols, J.M., Todd, M.D., Seaver, M., Trickey, S.T., Pecora,
Table 1. Structural property L.N. & Moniz. L. 2003c. Controlling system dimension:
A class of real systems that obey the Kaplan-Yorke con-
jecture, Proceedings of the National Academy of Sciences
Moment of inertia Iy,z 0.0108 m4
of the United States of America (PNAS), 100(26):15299
Youngs modulus E 210 Gpa
15303.
Weight/m 27.68 kN/m
Nichols, J.M., Trickey, S.T., Todd, M.D. & Virgin, L.N.
Poissons ratio 0.25
2003d. Structural Health Monitoring Through Chaotic
Damping ratio for the 1st and 2nd mode 0.01
Interrogation, Meccanica, 38:239250.
310
SHM in integrity management of deterioration prone concrete structures
M. Imran Rafiq
University of Surrey, Guildford, Surrey, UK
ABSTRACT management of deterioration prone structures (Rafiq,

2005). This would increase confidence in the predicted
Uncertainties encompassing predictive models of con- performance by reducing associated areas of epistemic
crete bridges make it difficult to decide the timing of uncertainty in the probabilistic models. SHM sup-
management activities. These high asset value struc- ported predictive models can be used effectively to
tures are subjected to a constant increase in loading provide early warning to potential failures, detect and
frequency and severity, and exposed to harsh envi- predict the rate of structural deterioration, and aid a
ronmental conditions, often decay at rates higher decision support tool for the planning of maintenance
than envisaged in design. Bridge management sys- on structural systems.
tems provide a systematic framework for repair and The concept of proactive health monitoring is
maintenance management. Key components of the presented whereby the monitored parameter is only
system include regular assessment of existing con- remotely related to the deterioration process. Its appli-
dition through a comprehensive program of regular cation allows the planning or justification of preventa-
inspections, and predictive models regarding the future tive actions on the bases of increasing risk of certain
condition of the structures. defects being developing. This aspect is exclusive
Visual inspections are subjective and qualitative. to proactive SHM where the structurally important
Occasionally supplemented with testing and monitor- defects have not yet physically initiated.
ing but the need to access the vicinity of structure is The methodology is demonstrated through its appli-
a major hurdle in their use. Hence, the information cation on a concrete element subject to chloride
obtained through the inspections, testing and moni- induced deterioration. The extent of deterioration
toring and cannot be used explicitly and efficiently varies at different locations of the system due to tem-
for performance prediction purposes. This aspect is poral and spatial effects of the deterioration. The use
highlighted in some detail in the paper. of multiple sensors is presented for the above case and
Inspections provide information that is continuous for the case where increased confidence in the sys-
in space but intermittent in time. Structural health tems performance is required at critical locations, or
monitoring (SHM) is typically continuous in time but increased robustness of the SHM is required. It has
spatially limited to certain points on the structure been shown that the proposed methodology is applica-
(Fig. 1). This information can be effectively com- ble for these scenarios and that the overall performance
bined to develop a powerful decision support tool for of the element/system can be obtained by rationally
combining similar data obtained through sensors at
different locations.
Much work is need in this area before the approach
can be implemented in practice. Key challenges in
this regard are discussed in the paper. These are being
addressed and will be presented in near future.
REFERENCE
Rafiq, M.I. 2005. Health monitoring in proactive reliabi-
lity management of deteriorating concrete bridges. PhD
Thesis. School of Engineering. University of Surrey. UK.
Figure 1. Interaction between information in space and time
(Inspection vs SHM).
311
Wireless sensor technology for continuous health monitoring of structures
A. Kounoudes
SignalGeneriX Ltd, Lemesos, Cyprus
A. Kalis
Research and Education Laboratory in Information, Athens, Greece
Centre Tecnologic de Telecomunicacions de Catalunya (CTTC), Barcelona, Spain
T. Onoufriou
Cyprus University of Technology, Lemesos, Cyprus
A.G. Constantinides
Imperial College, London United Kingdom
ABSTRACT integral part of this research is the development of fast,

energy efficient localization algorithms embedded on
This paper gives an overview of continuous monitoring miniaturized sensor node hardware to be developed
and local positioning principles for health monitoring in the framework of the project. This technology is
of structures using wireless sensor networks, includ- intended to develop a powerful remote monitoring tool
ing recent results of the research project WiseSPOT for structural civil engineers that can be used for con-
(Novel wireless sensor nodes with smart antennas for tinuous monitoring of the health of large structures
localization of faults in structures). A novel approach such as bridges, dams, the transportation and utilities
for the development of a new generation of miniature infrastructure.
low-powered wireless sensor nodes that will utilize The paper describes a new architecture for develop-
advanced smart antenna technology for continuous ing a low powered wireless sensor which utilizes smart
monitoring, localization and tracking of events in is antenna technology. The proposed system satisfies the
presented. Furthermore, the results of a first demon- hardware needs for advanced localization capabilities.
strator of an optimized directional antenna add-on Initial results from the directional switching antenna
board are presented. add-on board developed under the WiseSPOT project
Despite this significant increase of the application showed very promising results as far as the commu-
of Wireless Sensor Networks (WSN) in various indus- nication range, power consumption and localization
tries, their application to the infrastructure monitoring capabilities is concerned. To this end, further research
industry remains low, and this is mainly due to the will concentrate on the integration of the new anten-
limitations introduced by the current technology of nas into a new sensor node which will utilise advanced
wireless nodes. More specifically, wireless nodes cur- localisation algorithms for the detection of faults in
rently used in WSN have significant limitations in large infrastructure systems.
energy consumption, communication range, cost, and This two year project which started in January 2009,
their localization efficiency. These limitations impede is co-funded by the Cypriot Research Promotion Foun-
their wide integration in applications such as health dation. Three SmartEN participants are involved in
monitoring of large infrastructures (bridges, tunnels, WiseSPOT and the projects results will be regularly
water supply systems, transport networks, etc.), envi- disseminated to the SmartEN community.
ronmental control and many other industries, requiring
deployment of sensor nodes on a large scale. The
WiseSPOT research project aims to develop a new gen- REFERENCES
eration of miniature low-power wireless sensor nodes
that will utilize advanced smart antenna technology Onoufriou T. Constantinides A., Kounoudes A. SmartEN:
for continuous monitoring, localization and tracking A Research Framework on Smart Management for Sus-
of events in the network environment (i.e, crack detec- tainable Human Environment. Conference on Bridge
Maintenance, Safety and Management IABMAS 2010.
tion in large structures, land movements in landslide
Leang D., Kalis A., Smart Sensor DVBs: sensor network
prediction etc.). Localization is a network function that development boards with smart antennas, in: Proc. of
requires the synergy of both hardware and software IEEE International Conference on Communications, Cir-
mechanisms, and is applied both for finding the rela- cuits and Systems (ICCCAS 2004), June 2004, vol. 2,
tive and absolute locations of sensor nodes, as well p. 14761480.
as for localizing these events on the structure. An
312
Role of monitoring in life-cycle assessment, prediction and management

of deteriorating structures

ABSTRACT
Assessment, prediction and management of deterio-

rating structures are crucially needed practices for
the effective and economic upkeep of deteriorating
structures. The uncertainties in the available infor-
mation for conducting analyses pertaining to these
practices impose a great challenge in the process.
However, with the use of proper probabilistic tools
for handling these uncertainties, more accurate out-
comes do become achievable. In structures, one great
source of uncertainty is related to the load effect. Code
design values are traditionally used to establish maxi-
mum lifetime anticipated limits for these load effects.
In many cases, these limits are highly conservative.
In other cases, the loading demands of the structure
unforeseenly increase over time to even cross these
pre-set limits. This is clearly an undesired situation
that should be avoided. As an effective solution, the
demands of the structure may be monitored either Figure 1. Changes in the coefficient of variation of the
throughout the lifetime of the structure, or during extreme value distribution parameter un with increase in
separate periods over its lifetime. Doing so greatly monitoring information for the exterior girder.
improves the knowledge regarding the real loading
demands on the structure, reduces uncertainties in the (Neal 2003) that can also be used in difficult Bayesian
input information, and provides warnings to structural problems as such.
managers when an overload takes place. As shown in this paper and illustrated in Figure 1
The objective of this paper is to illustrate how data for an exterior girder in the bridge considered, the
obtained from monitoring can be integrated into the first sample monitoring data point produces a large
assessment, prediction and management of deterio- effect on the prior information. This was followed by a
rating structures. The role that monitoring plays in decreasing effect as the sample size grows. Evidently,
enhancing the outcomes of these processes is demon- the reduction of the uncertainty due to incorporating
strated. An example will be conducted over an existing the monitoring data is clear in the figure.
bridge in the state of Wisconsin. Details of the bridge
and its instrumentation are given in Mahmoud et al.
(2005). REFERENCES
Inclusion of the monitoring data in structural per-
Bedford, T. & Cooke, R. 2001. Probabilistic risk analy-
formance evaluation most likely means combining this sis. Cambridge University Press. The Edinburg Building,
data with the prior existing information. A Bayesian Cambridge, CB2, 2RU, UK.
approach is most suited for this situation. In engineer- Mahmoud, H.N., Connor, R.J. & Bowman, C.A. 2005 Results
ing applications, treatment of monitoring data is asso- of the fatigue evaluation and field monitoring of the
ciated with monitoring of extreme events. Extreme I-39 Northbound Bridge over the Wisconsin River. ATLSS
value distributions do not lend themselves easily to Report 05-04. Lehigh University, Bethlehem, PA, USA.
Bayesian updating (Bedford & Cooke 2001). The slice Neal. R.M. 2003. Slice Sampling. The Annals of Statistics,
sampling technique is a recently developed technique Institute of Mathematical Statistics 31(3): 705741.
313
Predicting the life of reinforced concrete structures in severe marine

environments
R.E. Melchers
The University of Newcastle, Australia
C.Q. Li
The University of Greenwich, England
ABSTRACT
It is generally agreed that reinforced concrete struc-

tures exposed to harsh marine environments will,
within a space of one or two decades, start to show
modest or even serious deterioration due to reinforce-
ment corrosion unless special care is taken to prevent
or reduce the rate of entry of aggressive chlorides.
Usually the life prediction is based on the rate of
ingress of chloride ions. However, careful review of
many older reinforced concrete structures shows that
some have survived decades despite very high chloride
levels and little or no protective measures or special
additives. Conversely, there are cases where rein-
forcement corrosion is evident despite considerable
concrete cover and high concrete quality.
A review is given of more than 300 cases for which
corrosion initiation and corrosion progression took
Figure 1. Mean time to initiation of corrosion as a func-
many years to occur. Considerable differences were tion of chloride content in the concrete. The bold corrected
found in the time to corrosion initation and in time to curves exclude structures affected alkali-aggregate reactions.
active corrosion. Figure 1 gives an example.
A small number of brief case studies are given,
together with an extended discussion of the possible
reasons why some reinforced concrete structures show
REFERENCES
much better long-term durability than others.
It is argued that long-term durability depends on Melchers, R.E. & Li, C.Q. 2009a. Reinforcement corro-
pH reserves (i.e. alkalinity) and that this can be sion initiation and activation times in concrete structures
provided by aggregates such as limestone and non- exposed to severe marine environments, Cement and
reactive dolomites. This observation is consistent with Concrete Research, 39: 10681076.
corrosion science principles. Melchers, R.E. & Li, C.Q. 2009b. Time to commence-
It is concluded also that calcium carbonate by itself ment of reinforcement corrosion in concrete structures
(such as caused by carbonation) does not lower the in marine environments, Proc. ACA conference, Corro-
sion & Protection, Coffs Harbour, Australasian Corrosion
concrete pH immediately adjacent to the reinforce-
Association, Paper 037.
ment to permit corrosion initiation. The additional Melchers, R.E. & Li, C.Q. 2009c. Reinforcement corrosion
leaching of alkalies must occur before corrosion can in concrete exposed to the North Sea for over 60 years,
initiate. Corrosion, 65(8) 554566.
These findings potentially have important practical Melchers, R.E., Li, C.Q. & Davison, M.A. 2009. Observa-
implications for the prediction of the life of reinforced tions and analysis of a 63-year old reinforced concrete
concrete structures. This matter is currently under promenade railing exposed to the North Sea, Magazine of
investigation. Concrete Research, 61(4) 233243.
314
Research and training challenges within SmartEN Marie Curie ITN
T. Onoufriou, N. Mousoulides, E.A. Tantele, C. Chrysostomou & D. Hadjimitsis

Cyprus University of Technology, Lemesos Cyprus
A. Kounoudes
SignalGenerix Ltd, Lemesos, Cyprus
A.G. Constantinides
Imperial College, London, United Kingdom
A. Kalis
Research and Education Laboratory in Information, Athens, Greece
ABSTRACT research performed locally and through intersectoral

visits in academic institutions and industry, partici-
This paper addresses the research and training chal- pation in multi-disciplinary scientific workshops and
lenges of the EU funded SmartEN Marie Curie ITN researching skills workshop, and dissemination and
Programme which focuses in the area of Smart Man- exploitation of research results in research conferences
agement for Sustainable Human Environment. The and workshops.
programme aims to train the next generation of Individual multi-disciplinary projects, a core action
research leaders in a subject of paramount importance of SmartEN project aim to provide researchers with
to European competitiveness through the develop- opportunities to: (a) develop expertise beyond their
ment of an initial research and training network that core area through short duration placements, (b) focus
will focus its activities on the development, effec- more on the integrating technologies moving away
tive integration and increased utilisation of emerging from isolated insular research efforts and topics and
technologies in wireless sensors, communications and achieving more effective component methodologies,
proactive management, targeting key issues of current (c) exploit the complementarity between the exper-
interest to the European Union and internationally. tise of the various partners as well as their research
The programme brings together a collaborative and training infrastructure, and (d) promote long term
research network that will focus its activities on networking and establish effective lines of communi-
research and training in the disciplines of Wire- cations.
less Sensor Networks, Sensor Signal Processing, Non In concluding, the aim of the SmartEN project is
Destructive Evaluation, and Smart Proactive Manage- to provide innovative research and solutions in the
ment. Beyond the research challenges in the various above interdisciplinary area through the development
individual disciplines involved in this programme of an initial research and training network. The signif-
there are many additional challenges associated with icance and methods of effective management of these
its multi-disciplinary nature and its trans-national and research and training key aspects are discussed in this
multi-centre research environment. Furthermore, the paper.
programme addresses many vertical and horizontal
research themes, working both at a generic and appli- REFERENCES
cation level. Key application areas to be considered
Onoufriou, T. 2007. Proactive Risk Based Framework for
within SmartEN include structural systems, heritage Integrity Management of Civil Infrastructure Systems.
and infrastructure, transportation infrastructure sys- Keynote Paper, Japan Conference on Structural Safety and
tems, and urban microclimate. Reliability, Tokyo.
The training challenges the SmartEN ITN net- Onoufriou, T. Constantinides, A. & Kounoudes, A. 2010.
work will face, include the development of a high SmartEN: A Research Framework on Smart Manage-
quality comprehensive and multi-disciplinary train- ment for Sustainable Human Environment. Conference
ing programme able to equip the young researchers on Bridge Maintenance, Safety and Management IABMAS
with world-class scientific knowledge and necessary 2010, Philadelphia, Pennsylvania, USA, 1115 July.
Ellinger, F. Eickhoff, R. Gierlich, R. Kounoudes, A. Kalis,
complementary skills in the area of research and devel-
A. Onoufriou, T. & Constantinides, A.G. 2010. Smart
opment of IP. The intended programme consists of a Wireless Sensor Technology for Continuous Health Mon-
number of interconnected and complimentary actions. itoring of Structures. Conference on Bridge Maintenance,
These include the development of individual multi- Safety and Management IABMAS 2010, Philadelphia,
disciplinary projects, training through personalized Pennsylvania, USA, 1115 July.
315
SmartEN A research framework on smart management for sustainable

human environment
T. Onoufriou
Cyprus University of Technology, Lemesos, Cyprus
A.G. Constantinides
Imperial College, London, United Kingdom
A. Kounoudes
SignalGenerix Ltd, Lemesos, Cyprus
ABSTRACT Projects aims will be achieved through: (a) con-

ducting top level research and training and devise inno-
This paper presents a research framework which vative solutions in the areas of monitoring and smart
focuses on developing innovative technologies in the proactive management, (b) educating the next gener-
area of Smart Management for Sustainable Human ation of intersectional and transnational researchers
Environment. This research framework underpins the in the area of smart management of the sustainable
development and implementation of the SmartEN environment and provide them with a unique range
Marie Curie ITN project. of skills and a network that will open up challeng-
This is an area of particular current interest world- ing and attractive career perspectives, and (c) taking
wide, given the increasing concerns regarding the forward the state-of-the-art in wireless sensor commu-
environmental impact of human actions, the use of nications, digital signal processing and non destructive
the environment and climate changes. These are cou- evaluation for the successful application of wireless
pled with ageing infrastructure systems, continuously sensors in the smart proactive management of the built
growing and changing demands on the built and nat- and natural environment.
ural environments and limited financial and depleting To this end, the aim of the SmartEN project is to
natural resources. Until now, research was focused on provide innovative research and solutions in the above
the development of proactive risk-based approaches interdisciplinary area through the development of an
for civil infrastructure reliability and management initial research and training network that will focus its
with benefits in improved performance, safety and activities on the development and effective integration
cost. However, there are significant uncertainties asso- of emerging technologies in wireless sensors, com-
ciated with the various predictive models directly munications and proactive management targeting key
affecting the quality of the decision making. Recently, application areas of current interest to the European
a new generation of miniature wireless sensor plat- Commission and internationally and including moni-
forms which utilize novel digital signal processing toring and smart proactive management of Structural
has emerged. These can be adopted to obtain large Systems, Heritage and Infrastructure, Transportation
quantities of highly diverse sensor data that are con- Infrastructure Systems and Urban Microclimate.
tinuously collected over a long period of time from
multiple locations providing significant insight on REFERENCES
the condition, demands and performance of the sys-
tem. These developments open up a completely novel Onoufriou T. (2007). Proactive Risk Based Framework for
area of multidisciplinary research towards the smart Integrity Management of Civil Infrastructure Systems.
management of sustainable environment. Keynote Paper, Japan Conference on Structural Safety and
The main objective of the SmartEN project is to Reliability, Tokyo.
implement a joint multidisciplinary research train- Onoufriou T., Kounoudes A., Constantinides A., Kalis A.,
Mousoulides N., Tantele E., Chrysostomou C., Had-
ing programme which will be focused on the smart jimitsis D. (2010). Research and Training Challenges
management of sustainable environment. Projects within SmartEN Marie Curie ITN. Conference on Bridge
activities will focus on research and training in the Maintenance, Safety and Management IABMAS 2010.
disciplines of Wireless Sensor Networks, Sensor Sig- Kounoudes A., Kalis A. et al. (2007). Local Positioning
nal Processing, Non-Destructive Evaluation and Smart for Wireless Sensor Networks. IEEE GLOBECOM 2007,
Proactive Management. Washington, DC, USA.
316
Role of structural health monitoring in Pareto optimization of bridge

management strategies
A.D. Orcesi
D.M. Frangopol
ABSTRACT
Appropriate assessment strategies of bridges includ-

ing visual inspections and monitoring programs are
essential for determination of efficient maintenance
strategies (Kim & Frangopol 2010). Nowadays, an
important challenge is to include monitoring concepts
in a general maintenance framework to determine opti-
mal maintenance strategies (Frangopol & Messervey
2009, Orcesi & Frangopol 2010). Monitoring data
provide additional information on the structural per-
formance and can change the intervention strategy.
Figure 1 shows, for three monitoring strategies, the
cumulative monitoring duration that impacts, in turn,
the accuracy of the additional information provided by
monitoring on the structural performance. The effects
of the duration of monitoring on accuracy of the Figure 1. Cumulative monitoring duration for three strate-
prediction results and on the maintenance strategies gies of monitoring.
decisions are investigated. As the monitoring duration
increases, more information becomes available. The process, based on monitoring results. The proposed
knowledge on the structural performance is then more approach is applied to an existing bridge.
accurate, but the monitoring costs are greater. Based
on the accuracy of monitoring information, different
decisions can be made. REFERENCES
The objective of this paper is to determine optimal
assessment and maintenance strategies, considering Deb, K., Pratap, A., Agarwal, S., & Meyarivan, T. 2002. A
multiple criteria. A reliability-based approach is pro- fast and elitist multiobjective genetic algorithm: NSGA-II.
posed to include uncertainties associated with struc- IEEE Transactions on Evolutionary Computation , 6(2):
182197.
tural degradation of bridges. Decisions are expressed
Frangopol, D.M. & Messervey, T.B. 2009. Maintenance prin-
in terms of probabilities. Inspections, monitoring pro- ciples for civil structures. Chapter 89 in Encyclopedia of
grams and maintenance actions are cost-dependent. Structural Health monitoring, C. Boller, F-K Chang, and
Therefore, the associated expected costs are cal- Y. Fujino, eds. John Wiley & Sons Ltd, Chicester, UK,
culated. Since maintenance decisions depend on Vol.4, 15331562.
assessment strategies, the optimal in-depth inspec- Kim S. & Frangopol D.M. 2010. Optimal planning of struc-
tion/monitoring times should be determined by tak- tural performance monitoring based on reliability impor-
ing into consideration all the possible outcomes tance assessment. Probabilistic Engineering Mechanics,
of the performance assessment. Optimal inspection- Elsevier; 25(1): 8698.
Orcesi,A.D. & Frangopol, D.M. 2010. Inclusion of crawl tests
monitoring-maintenance programs are determined by
and long-term health monitoring in bridge serviceability
using a Genetic Algorithm based procedure (Deb et al. analysis. Journal of Bridge Engineering, in press.
2002). Optimal solutions are Pareto solutions that con-
sider multiple criteria such as expected failure cost,
expected owner cost and accuracy of the decision
317
Optimization of life-cycle preventative maintenance strategies

using genetic algorithm and Bayesian Updating
E.A. Tantele & T. Onoufriou

Cyprus University of Technology, Cyprus
ABSTRACT at the rebar obtained from two assumed inspections

(year 5 and year 30) are incorporated in the probabilis-
Bridge maintenance is an effective way of maintain- tic model. In cases 2 and 3 the inspections data showed
ing the safety, condition and efficient operation of that the chloride ingress was less severe than originally
bridges. Preventative maintenance (PM) can play an predicted. By incorporating this observation through
essential part in the structural maintenance of bridges, the Bayesian Updating it was possible to reduce the pf
and if planned effectively, can result in reduced whole and in case 2 to improve the strategy and result in a
life maintenance costs while improving their safety lower overall cost. This is interpreted as finding a strat-
and performance profile. The authors have developed egy with lower cost than the initial that keeps the pf
an optimization genetic algorithm (GA) methodology lower than a target pf at all times. The overall reduction
that enables the optimization of PM strategies applied of cost was 6% which can be a significant amount con-
to reinforced concrete (RC) bridges. sidering the whole structure. It is expected that further
The PM measures are used to delay/prevent the benefits can be obtained by incorporating values from
reinforcement corrosion of bridge beams due to con- regular inspections and more available PM measures
tamination from chloride ions present in de-icing salts. through Bayesian Updating. These can have a signifi-
By combining probabilistic modeling of PM effective- cant effect and benefits in improving the reliability and
ness and GA optimization principles was feasible to cost of optimum strategies.
produce PM strategies that can maintain the reliability The methodology developed although specifically
profile within the acceptable limits while minimizing demonstrated here for RC bridge beams, is generic
the whole life costs. and can be easily adapted for application to other types
A key element in predicting optimum PM strategies of deterioration processes or structural elements and
using the GA methodology is the accuracy of estimat- systems.
ing the degree of deterioration of an element.To further Further work will be carried out to investigate the
improve the reliability of this estimation Bayesian full effect and potential of Bayesian Updating in con-
Updating is utilized. The use of Bayesian updating junction with regular inspections and/or monitoring
enables the updating of the probability of failure based and the GA methodology.
on site inspection data and the adjustment if neces-
sary of the timing of subsequent PM interventions. To
demonstrate the application and the effectiveness of
the proposed updated GA methodology 3 case studies REFERENCES
are presented in this paper. The studies also exam-
ine the influence of applying Bayesian Updating at Tantele, E.A. Onoufriou T, Mulheron M. 2006b. Optimization
of preventative maintenance strategies for bridges. In the
different time frames in reaching the optimum PM
proceedings of the IABMAS06 Third International Con-
maintenance strategy. ference on Bridge Maintenance, Safety and Management,
In case 1 Bayesian Updating has not been used. In Porto-Portugal, July 1619.
case 2 updating takes place at year 5 while in case 3 the Liu, M. Frangopol, D.M. 2004. Optimal Bridge Maintenance
updating occurs at years 5 and 30. In the cases employ- Planning Based on Probabilistic Performance Prediction.
ing Bayesian Updating data on chloride concentration Engineering Structures, Vol. 26, Issue 7, pp. 9911002.
318
MS15: European approach on integrated
infrastructure risk management (IRIS)
Organizer: H. Wenzel
Health monitoring of a smart base isolated benchmark cable-stayed

bridge using symptom approach
G.P. Cimellaro & A. De Stefano

Department of Structural and Geotechnical Engineering (DISTR), Politecnico di Torino, Italy
M. Domaneschi & L. Martinelli

Department of Structural Engineering (DIS), Politecnico di Milano, Italy
ABSTRACT
This paper presents a smart base isolation system

for cable-stayed bridges, which consists of passive
hysteretic devices between the deck and the piers.
The ASCE benchmark cable-stayed bridge developed
in MATLAB is used as case study in an updated
model developed in the commercial finite element
code ANSYS. The ASCE benchmark on cable-stayed
bridges has gathered, in the last years, the interest Figure 1. Finite element model in ANSYS.
of many specialists in the field of the structural con-
trol and the dynamic response of long span bridges.
Starting from the structural model of the original
benchmark statement, a refined version is developed The symptoms considered in the case studies are
in a commercial finite element environment to include vibration related parameters at selected position along
new modeling aspects, in the simulation of the stay the deck. The proposed structural condition monitor-
cables dynamics and in the soil-structure interaction. ing system is applied to detect the existence of different
The model comprises soil-structure interaction using types of damage in the bridge. Symptoms are ordered
impedance functions, the foundation are simulated by in a rectangular Symptom observation matrix (SOM)
lamped masses with equivalent spring and dampers. opportunely normalized. Each column of the matrix
Focusing the attention on the simulation of the struc- describes the change in time of each symptom, while
tural dynamics, the cable model is refined moving from each row corresponds to a given observation at a given
the single rod type representation, used in the bench- time t of all considered symptoms. By successive
mark, to a description with six rope elements for each application of single value decomposition (SVD), it is
cable. The monitoring of the bridge is approached in possible to obtain full extraction of fault-related obser-
the symptom space, assuming that the bridge degrades vations from SOM. In other words, it is possible to
and deteriorates with time. Damage to the longitudinal pass from multidimensional non-orthogonal symptom
isolator between the deck and the pier is considered, space to orthogonal generalized fault space, of much
while 42 observations are taken in account throughout reduced dimensions. These information are useful for
the life span of the bridge. redesign of CM systems.
Excitation is applied in the time domain in the
form of traffic load. This is simulated as translating
white-noise load intensity fields. Particularly, 8 traffic REFERENCES
lanes, with a moderate percentage of heavy vehicles,
are simulated as approaching the bridge from opposite Marco Domaneschi, Luca Martinelli, Passive and Semi-
directions, moving at a moderate speed. The loading active Seismic Protection of the ASCE Cable-stayed
process is carried long enough to assure the low struc- Bridge, The Twelfth International Conference on Civil,
tural modes can be excited. The moving traffic load Structural and Environmental Engineering Computing
(CC2009), 14 September, 2009, Funchal, Madeira, Por-
is applied on the finite element model in the form of tugal. (doi:10.4203/ccp.86.13).
a vertical load, bending and torsional moments at the Ove Ditlevsen, Traffic Loads on Large Bridges Modeled
nodes of deck that are described by individual nodal as White-Noise Fields, ASCE Journal of Engineering
time varying functions. Mechanics, v 120, n 4 1994. pp 681694.
321
Reliability of SHM procedures and decision support

in infrastructure management
A. Del Grosso
Department of Civil Environmental and Architectural Engineering, University of Genoa, Italy
1 INTRODUCTION 2 INFRASTRUCTURE MANAGEMENT, SHM

AND DSS
Infrastructure management is one of the topics that
has received increasing attention from both the aca- All the above considerations render the subject of opti-
demic and the engineering community in the last mal infrastructure management and the transfer of
ten or fifteen years. Indeed, the cost of infrastructure procedures already established in the industrial engi-
maintenance is becoming in many countries an over- neering field a very questionable issue. In the paper,
burden that reaches several GDP points, thus rendering such procedures are briefly reviewed in their adapta-
the implementation of strategies for maintenance cost tion to the civil infrastructure field and the question
optimization a must for infrastructure owners and will be raised concerning how a theoretical model of
governmental organizations. the decisional process involved could be constructed
Consequently, the approaches to systems manage- in view of the formalization of a suitable Decision
ment, risk and maintenance engineering already com- Support System (DSS).
mon in the industrial field are becoming popular also Recognizing that a proper decisional model can-
in the civil engineering field, opening new research not be based other than on observation of the real
lines commonly referred to as life-cycle engineering behavior and conditions, monitoring approaches will
approaches (Frangopol & Liu 2006). be reviewed with special reference to the process of
Significant differences however appear when com- deriving information on potential degradation states
paring the industrial systems to the civil infrastructure. from the monitoring data (Del Grosso et al. 2010).
The main difference appears comparing the operat- Finally, taking into consideration some of the com-
ing life of industrial systems and products and of civil monly used probabilistic reasoning schemes, the effect
infrastructures. This difference implies that the causes of the reliability of the damage identification algo-
of degradation of the safety and usability conditions rithms on the reasoning models will be discussed and
are much wider for the civil infrastructure case. potential lines of research are traced (Del Grosso,
A second but not secondary difference is related 2008).
to the fact that the phenomena that render the behav-
ior of real structures different from the corresponding REFERENCES
design models are very complex and unknown to a sig-
nificant extent. As a matter of fact, industrial products Aktan, A.E., Ellingwood B.R. & Kehoe B. 2007.
are very similar, if not identical, to one another while Performance-based engineering of constructed systems.
every civil structure is built in different environments Journal of Structural Engineering 133(3), 311323.
and at different times, with materials like masonry or Frangopol D.M. & Liu M. 2006. Life-cycle cost and perfor-
concrete the characteristics of which may be different mance of civil structures. McGraw-Hill 2006 Yearbook of
from one case to another, thus rendering the real as- Science and Technology, McGraw-Hill, New York.
built state substantially different and to some extent Del Grosso A., Lanata F., Posenato D. & Mercalli A. 2010.
Validation of SHM procedures for concrete bridges based
unknown. on static strain records. IABMAS; Proc. 5th Int. Conf.,
This latter consideration is very important when Philadelphia, 1115 July 2010.
dealing with already existing infrastructure, the state Del Grosso A. 2008. On the Reliability of Smart Moni-
of which may be very difficult to assess (Aktan et al. tored Structures. 14th World Conference on Earthquake
2007). Engineering, Beijing, China, October 1217, 2008.
322
Crystal clear data fusion in subspace system identification

and damage detection
M. Dhler, B.X. Lam & L. Mevel

INRIA, Centre Rennes Bretagne Atlantique, Rennes, France
ABSTRACT excitation factor to all the setups, then merges the data
and does the system identification in the final step.
Subspace-based linear system identification methods A new strategy called crystal clear is used to clean
have been proven efficient for the identification of the the resulting stabilization diagrams from unstable and
eigenstructure of a linear multivariable system in many noise modes in the identification step.
applications. Moreover, damage detection using null Furthermore, the uncertainty of the obtained modal
space based stochastic subspace detection techniques parameters is evaluated. For this, an existing algorithm
has also been proved to be useful in structural health is adapted to the covariance- and data-driven SSI used
monitoring during the last decade. Our main motiva- in this paper.
tion in this paper is output-only structural identifica- Finally we derive a version of the null space based
tion and damage detection in vibration mechanics. statistical damage detection algorithm that is adapted
The problem consists in identifying the modal to multi-setup measurements.
parameters (natural frequencies, damping ratios and All the presented algorithms are applied to the Z24
mode shapes) of a structure subject to ambient unmea- bridge, being a benchmark of the COST F3 Euro-
sured vibrations, by using accelerometer measure- pean network. The system identification results show
ments or strain gauges. This is output-only system high quality, confidence intervals are computed on the
identification, as the excitation input is unknown and obtained frequencies and damping ratios, and damage
not measured. Examples are, amongst others, offshore is successfully detected when comparing multi-setup
structures subject to swell, bridges subject to wind and measurements of two damage scenarios of the Z24
traffic, etc. bridge.
We analyze how the Stochastic Subspace Identifica-
tion (SSI) can be adapted when several successive data
sets are recorded, with sensors at different locations in REFERENCES
the structure. For doing this, some of the sensors, called
the reference sensors, are kept fixed, while the others Basseville, M., Abdelghani, M. & Benveniste, A. 2000.
Subspace-based fault detection algorithms for vibration
are moved for a so-called multi-setup measurement.
monitoring. Automatica, 36(1):101109.
Like this, we mimic a situation in which lots of sen- Mevel, L., Basseville, M., Benveniste,A. & Goursat, M. 2002.
sors are available, while in fact only a few are at hand. Merging sensor data from multiple measurement setups
However, there is one unpleasant feature of structural for nonstationary subspace-based modal analysis. Journal
identification of structures subject to ambient exci- of Sound and Vibration, 249(4):719741.
tation, namely that excitation is typically turbulent Overschee, P.V. & Moor, B.D. 1996, Subspace Identification
in nature and nonstationary. Like this, the excitation for Linear Systems:Theory, Implementation,Applications,
factor can change from setup to setup and has to be Kluwer.
normalized for the data fusion. Peeters, B. & De Roeck, G. 1999. Reference-based stochas-
tic subspace identification for output-only modal analy-
We describe a new merging algorithm for multi-
sis. Mechanical Systems and Signal Processing, 13(6):
setup measurements that is common for both 855878.
covariance-driven SSI and the data-driven SSI using Reynders, E., Pintelon, R. & De Roeck, G. 2008. Uncertainty
the Unweighted Principal Component (UPC) algo- bounds on modal parameters obtained from stochastic
rithm. This merging approach called PreGER (Pre subspace identification. Mechanical Systems and Signal
Global Estimation Re-scaling) introduces the same Processing, 22(4):948969.
323
Experiments for damage detection by subspace identification

on a tied arch bridge
C. Ebert
Wlfel Beratende Ingenieure GmbH & Co. KG, Hchberg, Germany
A. Lenzen
University of applied sciences Leipzig, Germany
ABSTRACT Results from experimental measurements in our

laboratory on a cantilever bending beam will show
In this lecture it is proposed to identify the dynamic that the presented methods are able to localize
characteristic of a structure by vibration measure- changes of stiffness and mass. A rectangular steel
ments. In the case of damage this characteristic will be pipe (80 40 2.9 mm) with a length of 2.45 m was
alternated. At the present time the experimental modal used as test object. For vibration measurements eight
analysis is often used to determine eigenfrequen- one-dimensional acceleration sensors were attached
cies, modal damping and mode shapes. For damage equidistant. The mechanical structure was excited by
detection finite element models will be optimized by impulse loads. Furthermore experiments on a pre-
experimental datas through model updating. The cor- stressed concrete tied-arch bridge in Hnxe (Germany)
rect choice of parameters which will be adapted in the will be presented. The bridge (built in 1952) had a span
finite-element model is crucial for this procedure. of 62.5 meters and was deconstructed in 2005. Main-
Here another method is to be presented for dam- and cross-girder, track-slab and the hanger consisted
age detection and localization. Black-box state space of prestressed concrete, the arch was built in reinforced
models can be identified by subspace method from concrete. On the verge of deconstruction it was possi-
measurement datas. These identified state space mod- ble to accomplish numerous vibration measurements.
els represent the transfer function between input and For the experiments two states as a variation of the
output. The black box model for the intact system is structure were induced. First an additional support near
compared with the black box model of the damaged the bridge bearing of one main girder was set-up. In a
monitored system. Variations of the structure can be second experiment one hanger from one tied arch was
detected by evaluating special damage indicators for cut through.
instance by the static or dynamic influence coefficients
([Link et al. (2007) Link, Sthr, and Weiland] and REFERENCES
[Lenzen and Waller(2003)]).
Additionally we will present a first step to transfer [Lenzen and Waller(2003)] Armin Lenzen and Heinz Waller.
black-box state-space models into white-box models Deterministische und stochastische systemidentifikation
in this lecture. White-box models are physically inter- mit methoden der linearen algebra zur formulierung
pretable and permit direct damage localization. The von mathematischen modellen im lebensdauerzyklus von
possibility of extracting mechanical properties like bauwerken. In 16th International Conference on the
mass, stiffness or damping direct from identified state Applications of Computer Science and Mathematics in
Architecture and Civil Engineering (IKM 2003). Univer-
space models will be shown by theoretical mechani-
sittWeimar, 2003.
cal equations. Because of differences between theory, [Link et al. (2007) Link, Sthr, and Weiland] Michael
simulation and experiment this is more difficult by Link, Stefan Sthr, and Matthias Weiland. Identifica-
real measurements. Nevertheless the identified model tion of structural property degradations by computational
parameters (e.g. Markov-Blocks) are able to detect and model updating. In International Conference on Damage
localize variances of mechanical properties. Assessment of Structures, 2007.
324
Web based monitoring and assessment of bridges and structures
M. Fritz & M. Stger

VCE-Holding GmbH, Vienna, Austria
R. Berger
mageba SA, Blach, Switzerland
ABSTRACT sophisticated systems can also offer additional features

like dynamic weight registration, to enhance the capa-
Structures like bridges are exposed to a variety of envi- bilities of the system depending on local circumstances
ronmental influences like wind, rain or traffic load and clients requirements.
which induce corrosion and material fatigue. Own- Modern monitoring systems feature also web-based
ers of structures want to or even have to know exactly control and monitoring capabilities. Clients can access
the condition of the structure due to different reasons. the measured data via a web-interface either in graphi-
Therefore over the last 10 years numerous monitoring cally prepared format or tabular format. Measurement
systems for bridges and structures have been installed. data provided for download afford the opportunity
Permanent monitoring systems in remote areas often to use it in calculations afterwards. Thresholds or
have to work without connection to the local power and more complex events can be configured for measured
telecommunication networks. Moreover site visits to parameters to induce an alarm via SMS, email or
pick up measured data and automatically generated simply on the web-interface.
reports are costly and impose a certain time delay Reconfiguration of the monitoring system like sam-
from measuring until notification of the owner. This ple rate, measuring intervals or changing the algo-
fact emphasizes the need of energy autonomous and rithms for assessment is possible through the web-
web-based remote monitoring and assessment. interface. Remote reconfiguring is reserved to the
Modern monitoring systems can be designed to con- monitoring system implementers because of their
stantly measure changes in parameters such as length, expert knowledge of hard- and software.
position, force, pressure, temperature or eigenfrequen- Hardware design and installation are mature; how-
cies. This information is measured at desired time ever awareness of the benefits of a modern web-based
intervals and recorded for analysis.The potential appli- remote monitoring system for structures still has the
cations of automated monitoring systems are almost potential to increase. Such system provide an exciting
unlimited, but would underlie the following general future for clients, for research and development as well
areas among others: as for special applications (e.g. modal shape analysis
or dynamic weight registration).
Safety concerns: ensuring immediate notification
of the occurrence of a change in a structures condi-
tion that may indicate that the structure is becoming REFERENCES
or has already become unsafe to use.
Engineering data: supplying records of loading and Wenzel H. (2009): Health Monitoring of Bridges; J. Wiley
movements to which a bridge is subjected, and the and Sons Ltd, ISBN0470031735; Chichester England
Wenzel H. Veit-Egerer R. (2009): Measurement based traffic
structures response to the conditions.
loading assessment of steel bridges a basis for perfor-
Usage data: providing records, such as weight and mance prediction in International Journal of Structure
speed of traffic using a bridge. Cameras can be and Infrastructure Engineering, Taylor & Francis Group,
integrated to gather visual proof of traffic events. New York, (submitted and already finally reviewed final
editing)
Basically a modern monitoring system consists of a
Veit-Egerer R., Wenzel H. (2007): Monitoring based weak
central processing node, at least one sensor, and some point determination of a steel bridgess torsional brac-
kind of connection to the central processing node. A ings with regard to fatigue threat, in Proceedings of the
power source local power supply system or some 2nd International Conference Experimental Vibration
kind of independent power supply system as well Analysis for Civil Engineering Structures (EVACES07)
as enough memory to store data is required, too. More ISBN 978-972-752-094-4, Porto, Portugal
325
Practical implementation of SHM with a special focus

on the end users needs
P. Furtner & H. Wenzel

VCE Holding GmbH, Vienna, Austria
ABSTRACT Curiosity driven motivations comprise those cases

where clients would like to know more about their
Health monitoring for civil engineering structures is important and complicated structures. Results can
a challenge. Our structures are a prototype each and also lead to better planning for future structures.
show small safety margins and a great exposure to
The usual life cycle assessment methods are not
the public. Bridges for example were the backbone of
very reliable and therefore tend to predict towards the
powerful empires from China to Rome and the Incas in
safe side. Differences between theoretically calculated
America. The transportation infrastructure is directly
and observed lifecycles may differ in an order of mag-
related to economic success of a nation.
nitude. The lifecycle assessment by means of parallel
Structural health monitoring of civil engineering
structural health monitoring can considerably improve
structures is more difficult than any monitoring of a
the accuracy of these predictions.
well defined mechanical structures. The major num-
If we close all structures where the predicted life-
ber of uncertainties in geometry, material properties
time has ended, our transportation network would
and the influence of the environment might have a
break down completely. The pressure on the SHM
higher impact on monitoring results than any minor
community to produce prediction of longer lifecycles
damage. Therefore only complex approaches under
will dramatically increase in the near future. Struc-
consideration and compensation of the already known
tural health monitoring has the burden to identify those
phenomena will be successful. This requires a more or
structures where less maintenance input will not lead to
less scientific approach.
unacceptable deterioration and secondly to accurately
Nevertheless an appropriate focus on the end-
assess those structures where critical stages have been
users needs must not be forgotten to achieve a more
or will be reached any moment.
widespread implementation and to increase the accep-
Actually the main deficiencies of a widespread
tance by the clients. End-users require robust and
implementation of SHM are:
reliable structural health monitoring concepts with full
automatic data analysis, interpretation of results and Closed solutions, which due to the complexity of
notification in case of problems. These concepts have the subject are almost impossible to be identified.
to cover the whole availability period of the structures The step from laboratory to the field has not been
starting form the design and construction phase and successfully managed. In many approaches environ-
ending with the demolition. The economic benefits mental conditions govern the response in the field.
have to be shown clearly. This completely invalidates the assessment results.
The motivation to apply and order services based A clear focus on the non scientific end users needs.
on the new technologies can be:
Responsibility driven, which means the new meth- REFERENCES
ods to become standard applications supported by
WENZEL, H., Health Monitoring of Bridges, J. Wiley
codes, standards and guidelines.
and Sons Ltd, ISBN0470031735; Chichester England,
Economically driven motivations, such as situations
December 2008.
where a ranking of structures to be rehabilitated is WENZEL H., PICHLER D., Ambient Vibration Monitoring,
necessary because of insufficient budget available or J. Wiley and Sons Ltd, ISBN 0470024305; Chichester
the need to use a structure for a certain time period England, April 2005.
longer than designed. www.brimos.com.
326
Recording and simulating environmental effects upon Tamar

Suspension Bridge
R.J. Westgate, K. Koo & J.M.W. Brownjohn

University of Sheffield, Sheffield, United Kingdom
ABSTRACT Instrumentation installed upon the Tamar Suspen-

sion Bridge, in south west England, UK, has been
It has been noted from long term monitoring results collecting modal data since 2006. Typical monitoring
that the static and modal data fluctuates over time, equipment has been used, such as accelerometers and
which is shown in Figure 1. It is noted that this corre- strain gauges, as well as the recent use of total posi-
lates with measured wind speeds and warming of the tioning station (TPS) to collect global displacements.
bridge elements. For damage detection, an understand- This has been presented in an accessible program for
ing of the environmental effects is essential, since they ease of reference.
will be the major contributor of noise to data otherwise Accompanying the research, a finite element (FE)
indicative of structural damage. model of the bridge has been created, as shown in
Figure 2, to provide theoretical results and further the
understanding of effects from environmental loading.
The accommodation of the hybrid structural system
of suspension cables with stay cables has been con-
sidered, as well as geometric non-linear effects of the
forces in the wires, and the catenary of the suspension
cable.
REFERENCES
Brownjohn, J. M. W. & Carden, P., Real-time operation
modal analysis of Tamar Bridge, 26th International Modal
Analysis Conference (IMAC XXVI), Orlando, 2008.
Cornwell, P., Farrar, C. R., Doebling, D. W. & Sohn, H., Envi-
ronmental variability of modal properties, Experimental
Techniques, Vol. 23, No. 6, 4548, 1999.
Liu, C. & Wolf, J. T. D., Effect of temperature on modal
variability of a curved concrete bridge under ambient
Figure 1. Modal frequencies of the Tamar Bridge through- loads, Journal of Structural Engineering, Vol. 133, No.
out a week in March. 12, 17421751, 2007.
Figure 2. Finite element model of Tamar Suspension Bridge.
327
A statistical method for masonry arch bridges damage detection
G. Ruocci & A. De Stefano

Department of Structural and Geotechnical Engineering, Politecnico di Torino, Italy
ABSTRACT
This work deals with the application of a vibration-

based damage detection method to masonry arch
bridges.The motivations which lead to the introduction
of SHM methodologies and non-destructive assess-
ment methods in the maintenance and preservation of
the historical bridges are presented. The critical issues
and the challenges of the implementation of automatic
diagnostic monitoring systems are addressed too.
The reliability of a monitoring system design
relies on the identification of the sources of weak- Figure 1. The experimental model of the twin-spans
ness and vulnerability. Therefore, the most common masonry arch bridge built in the laboratory.
causes of damage for masonry arch bridges are illus-
trated. The foundation settlements are identified as the the occurrence of a change in the system which can
most threatening damage causes for historical bridges ascribed to damage after the exclusion of the depen-
because of their accidental nature and their tragic dency with variation of the environmental conditions.
consequences on the structural integrity. In order to The results of the analyses carried out employing the
investigate this damage scenario a scaled model of a natural frequencies proved the reliability and robust-
masonry arch bridge was built in the laboratory and ness of these features. The observed increase of the
subjected to an extensive campaign of vibration tests. Mahalanobis squared distance with the testing sets
The application of an experimental modal analysis confirmed the suspects of the evolution of damage
technique on the results of the dynamic acquisitions characterised by an increasing extent in time. On the
allowed to identify the modal parameters and to assess contrary, the damping ratios resulted to be too scat-
their evolution in time. In order to investigate the tered and prevented to build a statistical model of the
influence of the environmental factors on the varia- unfaulted condition. This influenced the ability of the
tion of the identified modal parameters some specific method to distinguish the observations deriving from
environmental tests were arranged. different states of the structure.
A novelty detection algorithm was implemented
to statistically prove the occurrence of the change in
the state of the structure suggested by the decreas- REFERENCES
ing variation of the identified natural frequencies in
time. The concept of the vibration-based damage iden- Farrar, C. R. & Worden, K. 2007. An introduction to structural
tification expressed in the terms of the Statistical health monitoring, Phil. Trans. R. Soc. A. 15, 365: 303
Pattern Recognition paradigm is introduced and its 315.
methods are classified according to the required level Ruocci, G., Ceravolo R. & De Stefano A. 2009. Modal Iden-
tification of an Experimental Model of Masonry Arch
of the damage assessment, the pursued approach to Bridge, Key Engineering Materials, 413414: 707714.
the problem and the availability of the damaged and ISSN: 1013-9826, doi: 10.4028.
undamaged data. Worden, K., Manson, G. & Fieller, N.R.J. 2000. Damage
The outlier analysis based on the modal parameters detection using outlier analysis. Journal of Sound and
identified from the vibration tests statistically proved Vibration, 229(3): 647667. Elsevier.
328
Proposal of a workers and scaffolds monitoring and risk mitigation system

for building sites
A. Sassi, L. Gioanola & P. Civera

ChiLab Integrated Microsystems Applications Laboratory, Electronic Department, Politecnico di Torino, Turin, Italy
ABSTRACT Several different types of sensors can be involved

into the networks. For instance, into the BAN network
Building site is a place where workers are exposed to can be installed one or more accelerometers for falls
many risks, and moreover it is characterized by one of monitoring, one or more pressure sensors embedded
the higher death-rate into the entire industrial sector. in the shoes to check if the worker is carrying heavy
The main reasons are related to a wrong use or even loads, one or more acoustic sensors for monitoring if
to the missed use of the safety equipment by workers, the worker is exposed to an high rate of acoustic pol-
falls form a height or into a dig, falling objects, the lution, other types of sensors (temperature, humidity,
wrong assembly and use of scaffolds and their aging. biometrical sensors ).
In particular, falls from a height has been identified On the other hand, a SAN network can be composed
as the main cause of workers death, while the scaffold by strain gauges for axial forces monitoring, LVDT
collapse has been identified as the main risk factor for lateral displacements monitoring, gyroscopes for
during the erection stage of building site. structure tilt monitoring, high bandwidth accelerom-
To face such a scenario, the design of a sensor eters for structure vibrations monitoring and, finally,
network, both wearable by the worker and dislocated low bandwidth accelerometers for structure low fre-
in strategic places into the building site, could be a quency oscillations monitoring.
proper proposal to monitor the workers safety. Several
sensors can be organized in one or more networks,
with variable dimension and range, in order to real-
REFERENCES
ize a monitoring and warning system able to perform
punctual and continuous operations (otherwise not A.A.V.V. Labour Department of Hong Kong, 2008. Accidents
reachable in any other way) and to quickly alert the in the Construction Industry of Hong Kong (19982007).
staff in case of an abnormal checked condition. A.A.V.V. Ministry of Manpower of Singapore, 2006. ProBE:
The ChiLab proposal concerns a network composed Specific application to scaffold industry.
by two sub-networks: a Body Area Networks (BANs), A.A.V.V. Health & Safe Commission, 2009. Statistics of
one for each worker, with the main task of monitor- workplace fatalities and injuries Fall from a height.
ing the worker status and in particular if he is fallen Chan, S.L., Zhou, Z.H., Chen, W.F., Peng, J.L. & Pan,
A.D. 1995. Stability analysis of semirigid steel scaf-
down; a Scaffold Area Network (SAN) able to mon-
folding. Engineering structures vol. 17, issue 8: pag.
itor some important parameters such as stability of 568574.
scaffolds erected into the building site. Farringdon, J., Moore, A.J., Tilbury, N., Church, J. &
These two sort of sub-networks can interact to each Biemond, P.D. 1999. Wearable Sensor Badge & Sen-
other in order to improve the management of the sor Jacket for Context Awareness. Digest of Paper. The
worker safety in building sites. 3rd International Symposium on Wearable Computers,
Any single network (BAN or SAN) has an own San Francisco, CA, 2002, October 1819, 1999: pag.
coordinator able to manage the data collected from 107113.
the sensors, and any coordinator is able to directly or Huang, Y.L., Chen, W.F., Chen, H.J., Yen, T., Gao, Y.G. &
Lin, C.Q. 2000. A monitoring method for scaffold-frame
indirectly reach a gateway. There is just one gateway
shoring systems for elevated concrete formwork. Com-
for each building site that has the purpose of forward- puter and Structures 78: 681690.
ing the data collected to a remote device (for instance Nyan, M.N., Tay, F.E.H. & Murugasu, E. 2008. A wear-
a personal computer), where a decisional algorithm able system for pre-impact fall detection. Journal of
generates alarms and alert messages. Biomechanics vol. 41: pag. 34753481.
329
Observed dynamic characteristics of an overpass bridge during

destructive testing
D.M. Siringoringo, T. Nagayama, Y. Fujino, D. Su & C. Tandian

Department of Civil Engineering, The University of Tokyo, Tokyo, Japan
BRIDGE DESCRIPTION AND DAMAGE 2. Damage also alters the mode shapes locally. Modal
SCENARIOS displacements at the pier-girder node for damage
cases increase significantly suggesting immediate
Vibration measurement and analysis of dynamic char- effect of constraint-losing at the boundary condi-
acteristics of an overpass bridge during a full-scale tion. The changes are evident from bending and
destructive testing are described. The tested bridge torsional low-order modes and are well predicted
is the S101 Overpass Bridge located in Reibersdorf, by FEM. Effects of damage on mode shapes are
UpperAustria, west side of Vienna,Austria. The bridge more obvious in torsional modes than in bending
crossed over the national highway A1 Westautobahn modes as indicated by larger changes in modal dis-
Austria. It is a post-tensioned concrete bridge with the placement of pier-girder node of torsional modes
main span of 32 m, side spans of 12 m, and the width than that of bending modes.
of 6.6 m (VCE 2009). Measurement system consists 3. In general damping increases as the damage level
of six triaxial accelerometers. During two days mea- increases. Estimations from system identification
surement, ambient vibration of bridge was measured (Juang and Pappa 1985) indicate that damping
using six sensor configurations. in damage stages increase up to 2.53% from
Damage was introduced to the structure by cut- previously 1.5% in undamaged stage.
ting the pier column just above the pier footing. A 4. Multivariate outlier analysis (Worden et al 2000) is
hydraulic jack was placed on bottom of the steel col- used to detect the presence of damage. The results
umn to provide a temporary support. Immediately show that using Mahalanobis distance of auto-
after the cutting process was completed, the tempo- spectra the presence of damage at the earliest stage
rary steel column was lowered gradually by releasing can be detected with reasonable accuracy. When
the pressure in hydraulic jack. This caused the ver- damage has significantly changed the characteris-
tical settlement of the bridge at the location of pier tics of structure such as the case, all detection points
column. are unambiguously detected as outliers indicating
the clear presence of damage. Also the distance
between threshold line and damage points in outlier
detection are increasing as damage becomes larger.
RESULTS OF VIBRATION ANALYSIS This distance can be used further as indicator of
damage severity.
Through systematic data analysis using spectrogram,
output-only modal analysis, and multivariate outlier
analysis, important results of the study are summarized REFERENCES
as follow:
Juang JN, Pappa RS,1985.An Eigensystem RealizationAlgo-
1. A non-uniform pier settlement simulated as rithm For Modal Parameter Identification And Model
damage in this study, affects global stiffness of Reduction, Journal of Guidance, Control, and Dynamics,
structure significantly. This is evident by the signif- Vol. 8(5): 620627
VCE 2009. Progressive damage test S101 Flyover Reibes-
icant change in frequency of low-order modes. The
dorf,Vienna Consulting Engineers (VCE) Internal Report,
effects are more obvious in torsional modes than in Report Nr.08/2308, Vienna Austria, May 2009
bending modes, as indicated by larger changes in Worden K., Manson G., and Fieller N.R.J. 2000. Damage
frequencies of torsional modes than that of bending Detection Using Outlier Analysis, Journal Sound and
modes. Vibration. Vol. 229(3): 647667
330
Case based reasoning systems for comparative assessment
R. Stumptner, B. Freudenthaler & J. Kng

FAW Institute for Applied Knowledge Processing, Johannes Kepler University of Linz, Austria
M. Aburahma
Computer Science Department, Faculty of Media Engineering and Technology, German University of Cairo, Egypt
ABSTRACT and from matters of fine tuning of the system. Nev-

ertheless, for some situations the prototype provides
Aging and corrosion increasingly affects our infras- remarkable results. To predict if simple structures are
tructure, what makes regular and systematic inspec- in excellent condition showed an error rate below 1%.
tions indispensable, especially of critical structures To overcome the issue of subjectivity of experts, it is
(e.g. bridges as a part of the traffic infrastructure). An advantageous to keep the user informed on the current
inspection normally consists of (vibration-) measure- performance of the case base. Performance in this con-
ments and visual classification as a basis for further nection relates to the efficiency of clusters within the
analysis and assessment done by experienced experts. set of cases. This means that a user is informed how
It is obvious that this generally is a time-consuming clear the clusters in the case base are before includ-
process and besides that a certain amount of subjec- ing new information. Relying on that, he can decide
tivity has influence on the results. In this contribution that cases, which cannot be classified unambiguously,
a system is described which is based on the methodol- should not be included used as reference cases for
ogy Case-based Reasoning (Aamodt & Plaza 1994) future problems. Not administrated case bases can
for providing computer aid to experts working in the become unsystematic and can lose in performance.
field of measurement analysis and structure assess- The application of systems for automated classifica-
ment. The Case-based Reasoning (CBR) paradigm can tion of measurements of simple structures or structure
be described as a drawing of conclusions from compa- components promises good results and in certain cases
rable historical cases to solve new problems including even offers the chance of a fully automated process.
the integration of newly found solutions to step-by-
step improve the future problem-solving-ability of the
system, which consequently is based on a cooperative
approach (human-computer) of knowledge acquisition REFERENCES
(Freudenthaler et al. 2008). A solution in this con-
Aamodt, A. & Plaza, E. 1994. Case-Based Reasoning. Foun-
text can be a single piece of information or even a dational Issues, Methodological Variation and System
kind of algorithm to solve an initial problem. Apart Approaches. AI Communications 7(1): 3959.
from numerous advantages of Case-based Reasoning, Freudenthaler, B. Gutenbrunner, G. Stumptner, R. & Kng,
purely case-based approaches still have some short- J. 2008. Case-based Decision Support for Bridge Mon-
comings and the definition of basic rules can lead to itoring. Proceedings of the Third International Multi-
higher performance and precision. Conference on Computing in the Global Information
The current state of the CBR system for the inter- Technology: 3136.
pretation of measurements could be termed research Freudenthaler, B. Fritz, M. Stumptner, R. & Kng, J. 2009.
Case-based Reasoning for the Assessment of Simple
prototype. Results of first experiments, carried out
Structures in Terms of Structural Health Monitoring. Pro-
on measurement data of simple structures, reaching ceedings of the 7th International Workshop on Structural
an overall error rate below 10%, were quite promis- Health Monitoring.
ing (Freudenthaler et al. 2009). This error rate mainly Klein, M.R. & Methlie, L.B. 1995 Knowledge-Based Deci-
results from the experts subjectivity, particularly in sion Support Systems. With Applications in Business. New
marginal cases, from shortcomings in data acquisition York: John Wiley & Sons, Inc.
331
European approach on integrated infrastructure risk management
H. Wenzel & H. Allmer

VCE, Vienna, Austria
ABSTRACT
At present the European Industry recognized their obli-

gation to reconsider risk and safety policies, having a
more competitive industry and more risk informed and
innovation accepting society in vision. Therefore the
large collaborative project IRIS is proposed to iden-
tify, quantify and mitigate existing and emerging risks
to create societal cost-benefits, to increase industrial
safety and to reduce impact on human health and
environment.
Motivation Figure 1. The current practice line shows preventive main-

Current practices in risk assessment and management tenance actually not necessary. A firm knowledge of the
for industrial systems are characterized by its methodi- baseline will enable the omission of unnecessary works und
such reduce costs, without sacrificing safety! Design can
cal diversity and fragmented approaches. In retrospect make use of the information and save wherever over design
these risk and safety paradigms resulted from diverse exists!
industries driven and limited by available knowledge
and technologies. A change based on industry driven
R&D work is needed. The Partnership (>1 million workers) has been
selected out on the principles: complementarities, non
Project Outline competition and commitment to the subject of indus-
The project is led and driven by industry to consoli- trial safety. The top management is committed to
date and generate knowledge and technologies which take a big step towards considerably improved safety
enable the integration of new safety concepts related to performance in the European Industry.
technical, human, organizational and cultural aspects.
The partnership represents over 1 million workers.
The proposed project integrates all aspects of indus- REFERENCES
trial safety with some priority on saving human lives
Report on User Requirements supporting Risk and Vulnera-
prior cost reductions and is particular underpinning
bility Management of Systemic and Industrial Risks. 2006.
relevant EU policies. EC/JRC.
Sohn H., Farrar, C.R., Hemez, F.M., Shunk, D.D.,
Basic Concept Stinemates, S. W., Nadler, B.R. and Czarnecki, J.J. 2004.
In short the concept of IRIS is to focus on diverse A Review of Structural Health Monitoring Literature
industrial sectors main safety problems as well as to form 19962001. Los Alamos National Laboratory report
transform its specific requirements into integrated and LA-13976-MS. http://www.lanl.gov/projects/ei/shm/pubs/
knowledge-based safety technologies, standards and LA_13976_MSa.pdf.
services. Moon F.M., Aktan A.E., Furuta H. and Dogaki, Masahiro.
2007. Governing Issues and Alternate Resolutions for
The project covers a broad range of industrial
a Highway Transportation Agencys Transition to Asset
sectors and involves main stakeholders. Furthermore Management. SIE Journal, Taylor and Francis.
significant demonstration elements, training activities Frangopol. 2003. Life-Cycle Performance and Cost of Civil
and technology transfer, also on international level, Infrastructures. Keynote Address at the International
create a leverage effect for the acceptance in industry Structural Health Monitoring Workshop, Stanford, CA,
and society. Proceedings pp. 5363, DEStech Publications, Inc. 2003.
332
MS16: Safety & management of bridges in Mexico
Organizer: D. De Leon
Basis for risk management of bridges exposed to seismic loading
D. De Leon
Autonomous University of Mexico State, Toluca, Mexico State, Mexico
ABSTRACT
Bridges located on seismic and highly populated zones

involve a great amount of risk that should be mitigated.
Quantitative descriptions of the seismic hazard and the
risk involved due to the bridge importance constitute
valuable components for the management of mitiga-
tion measures to optimize limited resources for civil
protection and bridge maintenance in Mexico. Figure 1. Family of conditionals and unconditional initial
According to Meli (1994), the reliability of a struc- cost curves for a bridge on the zone III, Mexico City.
ture is associated to a certain cost which should be
minimized to balance safety with cost. Therefore, an intensities. The bridge designs were analyzed under
optimization process should be performed where the given maximum seismic coefficients c/g, using the
objective function must include the initial cost of the typical spectral form for Mexico City, and according
work and the cost of the potential damages and other to the range of intensities as reported in Mexican seis-
consequences in case a failure occur. mic hazard and failure rates studies (Esteva and Ruiz,
By recognizing the uncertainties inherent in the 1989).
design process, especially the seismic hazard, it has All the curves in the family shown in Fig. 1 are
been proposed, (Frangopol et al, 2001), to appraise conditional to the ocurrence of the indicated intensity.
bridge performance by using the expected life-cycle In order to obtain the unconditional curve, the ordi-
assessment. nates of the conditional curves need to be weigthed by
In the offshore technology (Stahl, 1986) the the ocurrence probabilities according to the seismic
expected life-cycle cost E[Ct], is expressed in terms hazard curve for Mexico City.
of the initial cost Ci and the expected failure/damage
cost E[Cd].
The acceptable (optimal) failure probability may REFERENCES
then be calculated by minimizing the expected life-
AASHTO, (2002), Standard Specifications for Highway
cycle cost respect the failure probability. Bridges, American Association of State Highway and
From well known structural reliability theory, the Transportation Officials, US.
bridge reliability is calculated, (Ang and Tang, 1984). Ang, A. and Tang, W., (2007), Probability concepts in
The structure is a vehicles bridge built on the Ben- engineering, John Wiley, US.
ito Juarez International airport area, in the transition Esteva, L. and Ruiz, S. (1989) Seismic failure rates of mul-
seismic zone III, in order to improve the traffic condi- tistory frames. Journal of Structural Engineering, ASCE,
tions. The bridge has a 400 m total span divided into 115: 2, pp. 268284.
16 segments of 25 m each. The structural modeling Frangopol, D.M., J.S. Kong, and E.S. Gharaibeh (2001).
was made through a finite element-based commercial Reliability-based life-cycle management of highway
bridges.
software (RAM Advanse, 2006). Meli, R., (1994), Diseo estructural, Limusa, Mxico.
A family of bridge designs were obtained RAM Advanse, (2006), Structural Analysis Software,
(AASHTO, 2002) by varying the original design 8th Version, Bentley Systems, US.
dimensions and steel areas. These designs allowed for Stahl, B., (1986), Reliability Engineering and Risk Analy-
a series of alternative designs to measure the varia- sis, in Planning and design of fixed offshore platforms,
tion of reliability with cost under specified seismic McClelland., B. and Reifel., M. (Eds), VNR, New York.
335
Seismic vulnerability of free toll bridges in Michoacan state, Mexico
H. Hernandez
Engineering School, Universidad Michoacana de San Nicolas de Hidalgo, Michoacan State, Mexico
D. De Leon
Engineering School, Universidad Autnoma del Estado de Mexico, Mexico State, Mexico
ABSTRACT deterioration makes mandatory an extensive revision

and generation of cost-effective strategies to protect
In Mexico, the free toll federal highway network has and maintain those bridges.
a length of more than 45000 Km and it is being In this paper, a bridge classification based on con-
operated by the Communications and Transportation struction year, substructure type, number of spans,
Secretary throughout the General Direction of High- viaduct width and approaches and substructure mate-
ways conservation. Usually, highways bridge design rial is presented. In addition, their epicentral distances
follows the AASHTO seismic specifications which, to the near seismic faults, are considered as a simplified
before 1971, were based on the SEAOC buildings lat- measure of seismic risk. Fundamental vibration peri-
eral force requirements but, after the San Fernando ods and the influence of the superstructure concrete
earthquake, CALTRANS code (1973) established new modulus and, in case it applies, masonry mechani-
seismic design criteria. These criteria set the basis for cal properties are also taken into account to state the
the AASHTO code from 1975 to 1992. seismic vulnerability of the bridge system.
However, and in spite of the fact that Michoacn Visually inspected damages are used for a pre-
State is located on a potentially seismic zone, many liminary assessment of safety conditions and repair/
bridges there have been designed without any seis- maintenance actions are proposed for each class of
mic consideration. Michoacn State has 485 highway bridge importance and damage level.
bridges and it is the Mexican State with most of the In the future, a probabilistic damage assessment and
bridges in the country. More than half of these bridges life-cycle economic consequences appraisal will serve
were designed between 1940 and 1965, with a struc- to improve the evaluation procedures and to set optimal
tural type and design loads typical for those years. maintenance strategies.
In addition, increased traffic demand and physical
336
New technologies to rehabilitate an old Mexican pier

in a harsh environment
A.A. Torres-Acosta
Instituto Mexicano del Transporte, Pedro Escobedo, Quertaro, Mxico
P. Castro-Borges
Centro de Investigacin y Estudios Avanzados del IPN, Mrida, Yucatn, Mxico
M. Len-Irola
Administracin Portuaria Integral de Progreso, Progreso, Yucatn, Mxico
ABSTRACT
A 70-year old concrete pier (a bridge-viaduct with

2 km long, 9.5 m wide, and 146 concrete arches), con-
structed with stainless steel bars, and exposed to a
harsh tropical marine environment (>C-5, ISO), has
shown good performance during its service life. Visual
and detailed inspections have shown structural deteri-
oration symptoms, but little corrosion was found on
the stainless steel bar. Several studies performed on
the old pier during the past five years, and using recent
technology advances in bridge monitoring, have shown
structural degradation due to the increase of load mag- Figure 2. Recent picture of the pier.
nitude and frequency, mainly longitudinal cracks on
some of the 146 arches. The paper describes the his-
torical information obtained from the pier construction REFERENCES
procedure and detailed inspection tests performed to
the pier. Brief discussion about durability issues on Castro, P., Troconis, O., Moreno, E.I., Torres-Acosta, A.A.,
this pier, based on the inspection results, and the reha- Martnez, M. & Knudsen, A. 2002. Performance of a
bilitation procedure performed, to increase the service 60-Year-Old Concrete Pier with Stainless Steel Reinforce-
life of this 70-year old structure, is given in this paper. ment. Materials Performance, 41(10): 5055.
Moreno, E.I., Torres, A. & Castro, P. 2004. Construction of
the Progreso pier (in Spanish). Revista Acadmica de la
Facultad de Ingeniera de la Universidad Autnoma de
Yucatn, 8(1): 6166.
Torres-Acosta, A.A., Fabela, M.J., Vzquez, D., Martnez,
M., Castro, P., Moreno, E.I. & Cuadros, H.D. 2005. Struc-
tural Evaluation and Rehabilitation of Concrete Arches in
the Progreso Pier. In E. Alexander (ed.), ICCRRR 2005
Concrete Repair, Rehabilitation and Retroffiting, Cape
Town, 2005. London: Taylor & Francis Group.
Figure 1. Picture of the pier during its construction.
337
Corrosion damage evaluation and diagnosis of bridges in the Mexican

highway network
A.A. Torres-Acosta, M. Martnez-Madrid, A. del Valle-Moreno,

J.T. Prez-Quiroz & M.A. Backhoff-Pohls
Instituto Mexicano del Transporte, San Fandila, Quertaro, Mxico
ABSTRACT
This work presents an innovative and original method

to determine the degree of corrosion in bridges built
in the Mexican Federal Road Network (MFRN), as
a function of environmental factors. These factors
include prevailing climate where the bridge is located,
accurate bridge geographic positioning (e.g. coastal
line, river distance) and its distance from industrial
corridors. This information was obtained and analyzed
using the last generation of Global Positioning Sys-
tems (GPS), incorporating the current structural data
produced during the last bridge evaluation available
by the national bridge administration management
program (SIPUMEX). This work present the results
obtained so far of the corrosion damaged bridges in Figure 2. Map of chosen bridges for immediate inspection.
the Mexican Federal Bridge Inventory.
REFERENCES
Torres-Acosta, A.A. & Martnez-Madrid, M. 2003. Residual
life of corroding reinforced concrete structures in marine
environment. Journal of Materials in Civil Engineering.
Vol. 15 (4): 344353.
Torres, A., Martnez, M., del Valle A. y Prez, J. 2004. El uso
de nuevas tecnologas en el mantenimiento, conservacin
en la infraestructura de puentes en Mxico que presenta
daos por corrosin. Revista Ingeniera de Construccin.
Vol. 19 (2): 6572.
Red DURAR. 1997. Manual de inspeccin, evaluacin y
diagnstico de corrosin en estructuras de hormign
armado. Maracaibo: CYTED.
Figure 1. Inventory of all concrete bridges in the MFHN.
338
MS17: COWI Group mini-symposium Cable supported bridges
Organizer: J.S. Jensen
Fatigue monitoring systems Great Belt Bridge
N. Bitsch, J.L. Jensen & J.E. Andersen

ABSTRACT The extraordinary traffic growth and in particu-

lar an increased share of heavy goods vehicles, has
The Great Belt Bridge is a major Suspension Bridge, called for a reassessment of the fatigue capacity of the
located in Denmark. The main span of the bridge has orthotropic steel deck and in particular of the welded
a length of 1624 meters and carries 3 lanes in both joints of the trapezoidal stiffeners and the trough-to-
directions.The bridge girder is a closed steel box girder deck plate welds. The results of various load tests and
with an orthotropic steel deck. The suspenders of the monitoring programs carried out during the period
cable system are locked coil cables with a spacing of 2003 to 2007 will be presented in the paper. Further
24 meters and a maximum length of 180 m. the paper will highlight the influence of the temper-
Since inauguration of the Great Belt Link in June ature of the asphalt pavement on the steel/pavement
1998 the traffic growth has been significantly above composite effect.
the forecasts and in the same period large vibrations
of the suspender cables has been discovered. In order to
monitor the effect of the increased traffic on the fatigue SUSPENDER CABLES
capacity of the orthotropic steel deck, and monitor the
effect of the cable vibrations on the fatigue capacity The large cable vibrations, with amplitudes on up to
of the suspender cables, a structural health monitoring 2 meters for the longest suspender cables, was evalu-
system was set up. This paper will present the results ated to cause fatigue failure in the locked coil cables
from the structural health monitoring system. at the upper socket entrance after a few years. A moni-
toring program was therefore set up in order to get
more detailed information about the size and num-
ORTHOTROPIC STEEL DECK ber of vibrations, and hereby be able to carry out a
more detailed evaluation of the impact on the fatigue
For year 2007 it was expected that the average daily capacity. The paper will present the result of the cable
traffic (ADT) would be more than 27,000 vehicles, vibration monitoring programme.
and according to the initial forecasts this level of ADT
should not be reached before year 2025.
341
Dehumidification of suspension bridge main cables
M.L. Bloomstine & O. Srensen

COWI A/S, Denmark
ABSTRACT A sealing system for the main cables, including

cable bands, saddles and other connected compo-
Corrosion of main cables on suspension bridges is nents.
a major problem on a worldwide basis. It has been A control and monitoring system.
proven by experience that the traditional system devel-
These components are designed as an integrated
oped about 130 years ago for corrosion protection of
system to suit the individual bridge and fulfill the
main cables generally does not provide sufficient pro-
specific requirements.
tection. Very serious corrosion has been detected on
A dehumidification system for main cables has so
the main cables of older bridges and somewhat surpris-
many clear advantages as compared to a traditional
ingly on younger bridges as well. In order to eliminate
system that it is obvious that it should be chosen instead
this serious problem, proven dehumidification tech-
of traditional system. These advantages include virtu-
nology has been further developed for application to
ally 100% safety against corrosion, low construction
suspension bridge main cables. The dehumidification
cost, short construction period, maintenance friendli-
method ensures that the atmosphere inside the main
ness, low maintenance costs, excellent documentation
cables is kept sufficiently dry so that corrosion can
for effectiveness, environmental friendliness and good
not occur. The full paper includes a history of dehu-
health and safety for workers.
midification, a description of systems for main cables,
The latest trend in dehumidification systems for
design considerations, information on the latest trend
bridges is integrated systems, i.e. systems that incor-
integrated systems and information on the current
porate as many steel bridge elements as possible in
status of existing systems.
one integrated system. An integrated system could for
The main principle of dehumidification is that steel
example provide corrosion protection for the bridge
does not corrode when the relative humidity (RH)
box girder, the towers and the main cables with just
is below 40%. Between 40% and 60% corrosion
one dehumidification plant in a tower or a box girder.
can occur, though at a very low rate. When relative
Several such systems have already been designed and
humidity exceeds 60% the rate of corrosion increases
will soon be constructed.
dramatically.
The current status for main cable dehumidifica-
Corrosion protection of steel bridge structures by
tion systems includes a total of 28 suspension bridges
dehumidification has been practiced for about 40
with completed or planned dehumidification of main
years. This started on the Little Belt Suspension
cables in a total of 11 different countries, including 20
in Denmark, where dehumidification systems were
systems in service.
installed in the box girder and the anchor chambers
As the main cables of a suspension bridge are
during construction in 1965 to 1970. Since the Lit-
virtually irreplaceable, it is essential that they are pro-
tle Belt Bridge, dehumidification has spread to other
tected from corrosion by the best means available. A
bridges, first in Denmark and then worldwide. It
dehumidification system is in all regards the optimal
has virtually become a worldwide standard to apply
method for protecting main cables from corrosion.
dehumidification to bridge box girders and anchor
This is the only system which completely prevents
chambers, as it is recognized as the most effective and
corrosion, whereas other systems at best can only
economical means of corrosion protection.
slow it down. Suspension bridge owners should take
Dehumidification systems for main cables are com-
advantage of this technology and make plans to pro-
posed of the following three main components:
tect their cables and bridges for many generations to
A dry air system capable of producing and blowing come.
dry air through the main cables.
342
Imagination, good engineering and saving money
K.F. Kirkwood, K. Deng & P.G. Buckland

ABSTRACT surveyed (at night during an overcast sky with no

wind) and the profile of the bridge as surveyed should
Imaginative but soundly based engineering can lead to match the profile predicted by the model. In virtually
considerable cost savings over the long term for own- all of the twenty suspension bridges treated this way,
ers and operators of major bridges. Examples are given the surveyed profile has not matched the predicted.
of such savings engineered by the Authors company This means that something was not modeled because
for a variety of bridge owners. The authors have used it was not known. The detective work then begins to
innovative engineering to proactively address both discover what has happened that was not known, and
rehabilitation and maintenance issues on a variety of is it serious? An early example of this was the discov-
bridges, always with an eye on keeping bridge or lane ery of stretch of the main suspension cables made with
closures to a minimum in order to minimize impacts structural strands.
on bridge users and/or maintain toll revenues as high The Canadian Highway Bridge Design Code has
as possible. The goal has always been to develop good introduced the ability to safely reduce load factors on
engineering solutions to minimize both rehabilitation the basis that more is known about an existing bridge
capital costs and ongoing maintenance costs. than is the case with a bridge still on the drawing board,
Buckland & Taylor Ltd. has pioneered structural and not all things have to be equally safe. This can
health studies of suspension bridges, whereby the provide considerable cost savings.
bridge as it was built is modeled in the computer, Examples are given of rehabilitation projects where
and all known changes to the bridge since it was built innovative but careful engineering has provided cost
are incorporated into the model. The bridge is then savings to the owner/operator.
343
When major bridges need to carry more and heavier loads
K.F. Kirkwood & P.G. Buckland

ABSTRACT Examples are given from the authors experience,

and all have been executed successfully. They encom-
As traffic loads increase, and more lanes of traffic are pass suspension bridges, cable-stayed bridges, arch
required, what can be done at an affordable cost to bridges, truss bridges and steel plate girder bridges,
increase the load carrying capacity of major bridges with concrete, steel and grating decks. Some were
that are already in service? Some of the key issues smaller bridges that can be regarded as prototypes for
are discussed, and examples are given of some origi- larger bridges.
nal methods of increasing the load capacity of major It is emphasized that the key ingredients for success
bridges, that have been proven in practice. are a clear understanding of structural behavior, careful
A key element is the need to keep traffic flowing study of the expected traffic loads, derivation of suit-
during any alterations, both for the benefit of the public able load factors, and, above all, creative thought pro-
and, if applicable, to maintain income at the toll booths. cesses applied with great care. High-powered analysis,
And if alterations are to be made, a perfect opportu- while often essential, should serve the process, not
nity presents itself of improving access for inspection drive it. Constructibility and practical considerations
and maintenance. are more important.
344
Structural health monitoring for Bosphorus Bridge
D. Mackenzie, N. McFadyen & S. de Neumann

Flint & Neill Limited, United Kingdom
ABSTRACT system for the bridge with the purpose of providing

KGM with information on the operation and condi-
The Bosphorus Bridge in Istanbul, Turkey, is one of tion of the bridge and providing vital health monitoring
the original streamlined steel box girder bridges. The information in the event of a major earthquake or wind
bridge spans 1074 m between Europe and Asia over storm.
the Bosphorus Straits, an important shipping chan- The volume of traffic restricts the bridge owners
nel between the Black Sea and the Marmara Sea. The ability to close parts of the bridge to traffic, thus limi-
bridge has now been closed to heavy goods vehicles, ting the ability to perform maintenance and repair
however it currently serves 200,000 vehicle passages operations. The sheer weight of traffic reduces the life
on a daily basis. of many of the replaceable components, placing some
In 2004, during strong winds and icy conditions, a elements at risk of premature failure.
hanger stool near to one of the towers failed in fatigue. The paper identifies the requirements for the sys-
It had also been noted that a number of hangers were tem and the development of the design. It looks at the
slack. The stool failure and the slack condition of a clients primary concerns and how these were dealt
number of hangers prompted a review of the structure. with by monitoring. The system build and operating
Review into the failure of the hanger stool identified constraints are described along with the clients longer
the primary cause as fatigue resulting from wind- term monitoring objectives.
induced vibrations. Appropriate mitigation measures The facility is expected to become a vital tool to
were thus installed on the hangers. the owner of the bridge in developing maintenance
Consequently, KGM sought advice on the state of strategy as well as providing rapid-response structural
the bridge and commissioned Flint & Neill to carry out assessment of this lifeline following extreme load-
a detailed assessment of the structure, a position survey ing events such as earthquakes. The Structural Health
of the towers and decks and a full inspection of the pri- Monitoring System presents a succinct tool to assist
mary structural components. From this information, with the management of the structure.
Flint & Neill developed a structural health monitoring
345
Bridge hanger deterioration
D. Mackenzie
Flint & Neill Limited, London, United Kingdom
N. Adamson
Flint & Neill Limited, Gloucestershire, United Kingdom
M. Bloomstine
COWI A/S, Lyngby, Denmark
ABSTRACT The inspection of hangers is therefore a critical

issue, where there is a desire to understand the condi-
The design of modern suspension bridges recognises tion of the hangers and the remaining life of damaged
that the deck hangers have a finite life, and one that hangers. Methodologies such as magnetic flux, x-ray
is likely to be less than the target design life of the and magnostriction systems are examined to assess
bridge. The ability to replace hangers in service there- their effectiveness in determining corrosion and wire
fore becomes a design case. This is a relatively new breaks
design constraint, specified in documents such at the The effectiveness of different protection strategies
Eurocodes1 ; for older bridges the replacement of deck such as galvanising, internal blocking and external
hangers becomes a major maintenance exercise. sheathing is investigated based on cases studies of
The performance of bridge hangers has shown that hangers thought out Europe, America and Asia.
design lives of 40 years have been the norm on bridges Finally a review of hanger replacement strategies
built prior to the 1980s but that modern bridges should is provided, showing methods of construction and the
have hanger life expectancies greater than this. The risk mitigation processes required to deliver a contract
paper opens with an examination the problems that successfully. Often the enabling works to gain safe
have arisen in the performance of deck hangers and access require as much effort as the hanger changing
ways in which maintenance strategies can be devel- itself.
oped to extend the design life of hangers. Steel wire
hangers are susceptible to corrosion and fatigue with REFERENCE
damage manifesting itself both externally, which is
easy to spot, and internally, where detection becomes BSI Eurocode BSEN 1993-1-11: 2006, Design of structures
critically in ensuring the safety of the structure. with tension components 2006
346
Bridge barriers and parapets
D. MacKenzie & J. Tarrell

Flint & Neill, London, Great Britain
O. Srensen
ABSTRACT The work follows a number of assessments of

bridges in Europe and Australia where earlier designs
The design of bridge barrier systems on long span failed to recognise that the greatest vulnerabilities to
bridges must strike a balance between providing safe loss of service and indeed complete collapse arise from
containment of traffic while achieving an economic errant vehicles. The results of these studies are pre-
cross section with low transverse wind load demands. sented along with the methods of retrofit that have
The problem exists as a solvable design problem in new been employed. In some instances the addition of
bridges, where flexibility exists in striking such a bal- safety fences can be extended to include wind shield-
ance and the economies of delivering all-year round ing around sharp obstructions, thereby improving the
wind protection can be explored. But with existing operational efficiency of the bridge in high winds.
structures built without consideration of traffic impact Chief among the design issues are the need to
effects, the consequences can be more significant. provide sufficient containment within restricted car-
The paper looks at existing vulnerabilities in bridges riageway widths and the ability to disperse large
where steel towers, cables systems and overhead impact forces into thin plate deck sections. The design
gantries are at risk from impact from heavy goods of such systems must also look at the effect of impacts
vehicles and the methods employed to assess the risk from light vehicles to ensure that a greater hazard has
of damage and loss of service to the bridge as well not been introduced.
as the methods of providing protection to the vital
components.
347
Surfacing for orthotropic bridge decks
N. McFadyen
Flint & Neill Limited, Stone, Gloucestershire,GL13 9LB, UK
J. Blumensen
COWI A/S, Parallelvej 2, DK-2800 Kongens Lyngby, Denmark
ABSTRACT The paper addresses both design issues and practical

considerations for the owner, specifier and applica-
Due to increasing traffic demand, the need to find tor; however, lessons learnt extend beyond the cited
improved surfacing systems for the UKs stock of systems.
lightweight steel box bridges arose in the mid 1980s
and continues to the present day. These structures, built
during the 1960s and 1970s were originally provided REFERENCES
with a thin (38 mm) mastic asphalt surfacing system
and rubberized waterproofing membrane. Evans J.E. 1992 The Strengthening and Refurbishment
The resurfacing of five major U.K. bridges with of Severn Crossing, Proceedings Institution of Civil
orthotropic steel decks is described, giving the back- Engineers Structures & Buildings,. 94 Feb., 2336.
Smith J.W. 1988 Surfacing materials for orthotropic bridge
ground to the research into replacement systems and decks. 1st Oleg Kerensky memorial conference, Institu-
their development towards enhancing the life of the tion of Structural Engineers, Session 5.
underlying steel deck. Following the research, a stan- Cullimore M.S.G., Flett I.D., Smith J.W., Feb. 1983 Flexure
dardized method of system evaluation was developed of steel bridge deck plate with asphalt surfacing. IABSE
for tendering purposes. Proceedings pp 57/83, 115.
The paper describes practical aspects of system Cuninghame J.R , Beales C., 1992 Strengthening and refur-
application, including deck inspection and repair. With bishment of Severn Crossing Part 4: TRRL research
reference to research findings and experience gained on Severn Crossing Proceedings Institution of Civil
on a number of bridges, the paper will provide useful Engineers Structures & Bldgs,. 94 Feb., 3749.
Smith B.W., Huband M.V. 1993 Resurfacing Severn Cross-
information on specifying and overseeing resurfacing: Development of the chosen system, Surrey Confer-
ing contracts with long-term performance firmly in ence (2), pp 774 to 786.
mind. Flint A.R., Smith, B.W. 1992 Strengthening and refur-
Two markedly different system types are discussed; bishment of Severn Crossing, Part 5: Other background
a development of the original UK mastic asphalt research and development. Proceedings Institution of
system incorporating lake asphalt and an all-epoxy Civil Engineers Structures & Bldgs,. 94 Feb., 5160. Paper
system. Conclusions are drawn on the effectiveness 9849.
of the systems from the point of view of asphalt dura- Rebbechi J.J. 1980 Epoxy asphalt surfacing of West Gate
bility, enhancement of the performance of underlying Bridge. Proceedings, Australian Road Research Board,
V10, part 3, pp 136146.
steelwork and the reliability of the application process.
Many papers on surfacing composite action and
fatigue in orthotropic decks have been published over FURTHER READING
the years, whilst others have covered the design of
specialized asphalts; it is hoped this paper will help Gurney T. 1992 Fatigue of Steel Bridge Decks. TRL State of
to focus on tying these issues to effective technical the Art Review 8. HMSO Publications Centre, London,
delivery ISBN 0 11 551141 5.
348
Bridge bearings and joints
O. Srensen
D. MacKenzie
Flint & Neill, London, Great Britain
E. Lauersen
A/S Storebltsforbindelsen, Denmark
ABSTRACT design philosophy should therefore be based on: As

few as possible movement structures, long durability,
Many years of experience of design and maintenance easy to maintain and replace. This paper describes:
of bridges shows that movement structures are among Typical articulation systems on major suspension
the more expensive as far as maintenance costs are con- bridges and minor urban cable stay bridges are
cerned. These are mechanical structures and require described among others how wear and durability
regular maintenance, or part or full replacement. The problems can be overcome.
349
Replacement of suspension bridge suspenders and main cables
O. Srensen & M.L. Bloomstine

P.G. Buckland
Buckland & Taylor Ltd, Vancouver, Canada
ABSTRACT replaced. The new main cables are protected from

corrosion by dehumidification. At the same time the
Projects for replacement of suspenders and/or main deck was widened to include six lanes instead of
cables are described. The lvsborg Suspension Bridge four. The rehabilitation work was executed with pass-
crosses the Gta River at the entrance to Gothenburg ing traffic of about 70.000 cars a day. Two examples
in Sweden and is an essential link between the city are given of taking advantage of suspender replace-
and the industrial area to the north. The inspections, ment to gain other benefits. In the case of the Lions
investigations and condition history of the suspender Gate Bridge temporary suspender extensions were
cables leading up to the replacement and the replace- used to control stresses during reconstruction of the
ment are described. The Aquitaine Bridge crosses the bridge, and in another case the opportunity arises
river Garonne at Bordeaux in France. The total sus- of permanently increasing the vertical clearance for
pension system, main cables and suspenders, were shipping.
350
MS18: Lifetime design of super long span bridges
Organizer: H.-M. Koh
The prediction of long term operation and maintenance costs

of long span bridges
N. Bitsch & E.S. Larsen

J.V. Thomsen
Danish Road Directorate, Copenhagen, Denmark
ABSTRACT elements it is possible to assess their service life and

a list of experience based service life is set up.
More than 40 years of operation and maintenance The overall picture is very similar for each type of
experience of major structures is hereby presented the bridges; namely there will within the first 10 years
as costs on electric and mechanical components and of service life be a limited level of maintenance costs.
structural elements on the structures. A number of This level will raise and reach maximum after approx-
structures is represented such as suspension, cable imately 30 years depending on the type of bridge and
stayed, bascule and cantilevered concrete bridges. the structural material. After 30 years the costs again
Analyzing the data gives a clear picture of the dis- will deviate between a relative high level and the max-
tribution of both operation and maintenance costs as imum level depending on the repetition in replacement
function of time. This distribution is depended on the and reinvestment.
type of structure and the quality of the components Based on insight in unit costs, the actual bridge
used in the structure. and the service life of the components and structural
The overall importance is the investment and the elements, it is possible to predict average costs for
service life (performance) of the said component or operation and maintenance with time. Examples on
structural element. For the components and structural the 30-60 years budget will be given.
353
Deterministic and probabilistic durability design methods

and their application to super long span bridges
S.-W. Cha
Department of Civil and Environmental Engr., University of Ulsan, Korea
K.-H. Kim
Research Institute of Energy and Resources, Seoul National University, Korea
J.-Y. Cho
Department of Civil and Environmental Engr., Seoul National University, Korea
ABSTRACT solutions of the Ficks second law of diffusion. Life-

365 service life prediction model is an example of this
Many concrete structures including breakwaters, tidal approach.
barriers, offshore drilling platforms, have been con- Deterministic durability design approach, how-
structed in marine environment. In Korea of nowadays, ever, can not consider the variability of parameters
a number of long span bridges have been constructed concerning the chloride penetration. On the contrary,
and are also being built in marine environment as probabilistic method, such as DuraCrete methodology,
shown in Figure 1. Corrosion of reinforcement reduces can be used in the durability design. This method uses
the service life of concrete structures, thus durability the error-function solution of chloride penetration and
design against chloride attack is needed to guarantee considers the variability of parameters.
a long service life of the concrete members of super In this paper, the fundamentals of chloride diffusion
long span bridges such as concrete pylons, piers and are dealt first. Two approaches of durability design are
footings. presented and some design cases of long span bridges
Durability design approach for concrete structures in South Korea are introduced. Diffusion coefficient
in marine environment is divided into two categories: is dealt next. An improvement of diffusion coeffi-
deterministic and probabilistic approaches. Simple cient model is made based on the consideration of test
deterministic durability design method against chlo- method and mathematical expression. For probabilis-
ride attack is to calculate the service life using so- tic design, first and second order reliability analysis
called error-function solution with respect to the Ficks methods are applied to calculate the probability of cor-
second law of diffusion. This method is simple and rosion initiation in order to compare the accuracy of
relatively easy to use. A more sophisticated way to results.
consider the changes of environmental temperature
and surface chloride concentration is to use numerical
REFERENCES
Bentz, E.C. & Thomas, M.D.A. 2008. Life-365 Service Life
Prediction Model and Computer Programe for Predict-
ing the Service Life and Life-Cycle Costs of Reinforced
Concrete Exposed to Chlorides. Lovettsville: Silica Fume
Association.
Cha, S.W. et al. 2009. Durability Estimation and Opti-
mum Mix Proportion of Marine Concrete Bridge. Seoul:
Daelim.
Collepardi, M. et al. 1972. Penetration of Chloride Ions into
Cement Pastes and Concrete. American Ceramic Society
55(10): 534535.
Haldar, A. & Mahadevan, S. 2000. Reliability Assessment
Using Stochastic Finite Element Analysis. NY: John
Wiley & Sons.
Maage, M. et al. 1996. Service Life Prediction of Existing
Concrete Structures Exposed to Marine Environment. ACI
Materials Journal 93(6): 602608.
Tang, L. 1996. Chloride Transport in Concrete Measure-
ment and Prediction. Ph.D. Dissertation. Gothenburg,
Figure 1. Long span bridges in Korean seaside. Sweden: Chalmers University of Technology.
354
Analysis and comparison of recent bridge failures in China and the rest
of the world
Y.-J. Ge and Z. Sun

State Key Laboratory for Disaster Reduction in Civil Engineering,
Department of Bridge Engineering at Tongji University, Shanghai, China
ABSTRACT failures occurred disproportionally high in China com-

pared with the rest of world, some further detailed
As the most important requirement for bridge sustain- study is thus required to tell the reason; 3) the most
ability, structural safe reliability attracts much concern frequent causes of bridge failures were attributed to
of bridge engineers. Although many efforts have been construction and maintenance, and these two causes
put on the related scientific research and technology constitute 37.8% in China and 35.9% in the rest of
development, recent bridge failures reveal that the the world, of the total bridge failures, which tells that
efforts are still not enough. This study is thus moti- the more efforts of bridge engineering community
vated to investigate bridge failures that have occurred should be put on the study related to bridge con-
in China and the rest of the world recently, i.e., the struction, maintenance and dismantle, the process of
period between 2001 and 2009. According to informa- bridge design can meet the nowadays bridge construc-
tion collected, the types and causes of bridge failures tion requirement; and 4) the most frequent triggering
is analyzed in this study. causes of bridge failures were attributed to overload-
Collective studies conducted for bridge failures ing and ship collision, which is expected to be avoided
revealed that 37 and 39 bridges of various types failed by strict vehicle and ship management.
in this period in China and the rest of the world, respec-
tively. The age of the failed bridges ranged from 0 year
(during construction) to 68 years in China and 140 ACKNOWLEDGEMENT
years in the rest of the world. Based on the informa-
tion collected, the following phenomenon is observed: The study is partially supported by the NSFC Grant
1) beam/girder bridges are most frequently failed since 90715039, and the MOST Grants 2008BAG07B02
they are most popular bridge type; 2) arch bridge and 2006AA11Z109.
Table 1. Causes of Bridge Failures.
Number of collapses Percentage
Principal causes Specific causes China World China World
Enabling Design 1 2 2.7 5.1

Detailing 1 0 2.7 0
Construction 8 9 21.6 23.1
Maintenance/dismantle 6 5 16.2 12.8
Material 1 0 2.7 0
Triggering Overload 9 5 24.3 12.8
Collision 3 2 8.1 5.1
Flood 3 3 8.1 7.7
Strong wind 0 2 0 5.1
Explosion 0 1 0 2.6
Fire 0 1 0 2.6
Landslide 1 1 2.7 2.6
Terrorist 0 1 0 2.6
Unknown 4 7 10.8 17.9
Total 37 39 100 100
355
Terminology for treating disproportionate collapse
M. Haberland & U. Starossek

Hamburg University of Technology, Germany
ABSTRACT A performance-based framework for designing

against disproportionate collapse applicable to any
Disproportionate collapse is a complex problem for kind of structure, and thus also to bridges is pre-
which the existing terminology and procedures are sented A set of design criteria is presented. These
inadequate. A multitude of terms is used to describe include design requirements, design objectives, design
structural characteristics and concepts in the context strategies, and verification procedures.
of disproportionate collapse. Some of them, namely Design requirements are the specification of
collapse resistance, structural robustness and vulner- whether collapse resistance is required for a structure
ability are discussed in this paper. This paper dis- and, if yes, the specification of the level of require-
tinguishes these terms and the associated structural ments on design objectives, design methods, and
characteristics, and suggests working definitions. verification procedures. The requirements depend on
A disproportionate collapse is characterized by a the significance of the structure with respect to the con-
pronounced disproportion between a relatively minor sequences of a collapse and on the structures degree
event and the ensuing collapse of a major part or of exposure to abnormal events. Collapse resistance
the whole of a structure. Disproportionate collapse will not be required for every structure.
is prevented by ensuring collapse resistance, a pro- Design objectives are the basis of a performance-
perty defined as the insensitivity of a structure to based design. They comprise hazard scenarios, per-
abnormal events. An abnormal event is an event that formance objectives, and applicable combinations of
is unforeseeable or occurs with very low probabi- actions and safety factors. Hazard scenarios are the
lity and is not considered in the ordinary design of abnormal conditions to be assumed in the design
a structure. Collapse resistance can be achieved by to affect the structure during construction and life-
reducing the exposure of a structure or by reducing time. In a threat-specific approach, they are specific
its vulnerability two measures that aim at preventing abnormal events. In a non-threat-specific approach,
failure initiation or by increasing its robustness a they are notional actions or notional damage, without
measure that aims at preventing failure progression. regard to the cause. Performance objectives specify
The exposure results from the abnormal events that the acceptable response of the structure to the hazard
possibly affect a structure during construction and scenarios.
lifetime. The vulnerability of a structure is its sus- Design methods are methods that enhance the col-
ceptibility to suffer immediate initial damage in case lapse resistance of a structure; available methods are
it is affected by abnormal events. Robustness is the event control, protection, increased local resistance,
structures insensitivity to such initial damage. alternative load paths, and segmentation.
The basis of current reliability-based design codes Verification procedures are used to demonstrate that
for general structures is reviewed with particular a structure meets specified performance objectives
regard to their suitability to prevent disproportionate when subjected to specified hazard scenarios. Verifi-
collapse. Their inadequateness regarding the preven- cation procedures of varying degrees of accuracy can
tion of disproportionate collapse is outlined. be performed, which are selected depending on the
To give immediate guidance to the practicing engi- requirements and the design objectives.
neer, a pragmatic design approach is proposed in which Codification is moving towards such an approach
probability-based design procedures as described in already. This paper shows how these efforts can be
the codes are complemented by an additional assess- continued to lead to a clearer description of design
ment and design measures with particular regard to criteria and to a precise and consistent use of language.
disproportionate collapse.
356
Identification of aeroelastic parameters for cable-supported

bridges using measured accelerations
Y.H. Hong, K. Jung, H.K. Kim & H.S. Lee

ABSTRACT Though EEE results in less compatible results

between predicted and measured displacements than
After the flutter derivative-based aeroelastic formula PEM, the authors believe that the flutter derivatives
had been by Scanlan, great number of efforts have been identified by EEE contain more physically reason-
made to estimate the flutter derivatives from the test of able meanings based on the governing equilibrium
bridge model in wind tunnel. Scanlan and Tomko pro- equation. Moreover, the EEE is free from the con-
posed the extraction scheme for flutter derivatives from vergence of ill-posed problem, and give very stable
2DOFs coupled motion tests (1971). Sarkar developed identification results.
the Modified Ibrahim Time Domain (MITD) to esti-
mate cross flutter derivatives along with direct flutter
derivatives (1994).The procedures of these approaches ACKNOWLEDGEMENT
are generally based on the response error estimation,
which minimizes the relative error between measured This research was supported by the grant (09CCTI-
displacement and predicted displacement. A052531-02-000000) from the Ministry of Land,
However, the flutter derivatives are conceptually Transport and Maritime of Korean government
more closely related with the aerodynamic force equi- through the Core Research Institute at Seoul National
librium than with the predicted displacement itself University for Core Engineering Technology Develop-
in formulas. Therefore, this paper proposes a new ment of Super Long Span Bridge R&D Center.
approach to identify flutter derivatives based on the
equation error estimation (EEE) which minimizes
the relative error between the resisting forces (stiff- REFERENCES
ness, damping and inertia) and the aeroelastic self-
excitation forces. Scanlan H. R., & Tomko J.J. 1971, Airfoil and bridge deck
flutter derivatives. Journal of the Engineering Mechanics
The EEE approach requires not only displacement
Division 97(6): 17171737.
response but also velocity and acceleration history Sarkar P. P., Jonse P. N. & Scanlan H. R. 1994, Identification
for system identification. In this approach, a displace- of Aeroelastic Parameters of Flexible Bridges. Journal of
ment and velocity reconstruction scheme is used to Engineering Mechanics 120(8): 17181742.
calculate displacement and velocity history from mea- Lee H.S., Hong Y.H. and Park H.W., Design of an FIR Fil-
sured acceleration. Hence, both these reconstructed ter for the Displacement Reconstruction Using Measured
responses and the measured acceleration is used for Acceleration in Low-frequency Dominant Structures,
EEE method. 2009, doi: 10.1002/nme.2769, International Journal for
The validity of the proposed method is demon- Numerical Methods in Engineering.
Lee H.S., Hong Y.H. and Kim H.K., Reconstruction of
strated through the experimental free vibration test.
Dynamic Displacement and Velocity from Measured
The flutter derivatives of a section model representing Accelerations Using the Variational Statement of an
a bridge deck system are evaluated using the mea- Inverse Problem, under review after submitted in Journal
sured acceleration and reconstructed displacement and of Sound and Vibration.
velocity using proposed method, and compared with
the result of the prediction error minimization method
(PEM) with measured displacement.
357
Reliability-based durability design and quality control of long span

bridge of Busan-Geoje Fixed Link project in Korea
S. Kim, S. Jeong, J. Kim & I. Koo

DAEWOO E&C, Seoul, Korea
ABSTRACT
The Busan-Geoje Fixed Link comprises a 8.2 km

motorway link from Busan, Koreas southernmost and
second largest city, to the island of Geoje. The connec-
tion includes two cable-stayed bridges, respectively 1.6
and 1.9 km in length and a 4 km immersed tunnel with
50 metres water depth. The project is scheduled for
completion in 2010. DAEWOO E&C is the leading
contractor.
The durability design criteria for the Busan-Geoje
Fixed Link Bridges and Immersed Tunnel specify
that the contractor must verify that the concrete in
combination with the selected concrete cover will
prohibit initiation of corrosion during the design life Figure 2. The distribution of tensile stress and strength in
(100 years) with a probability of 90%. pylon wall which is used to calculate the crack width.
The DuraCrete approach developed during an Euro-
pean Research project (19961999) and internation-
place in early aged concrete by restraint to the volume
ally the only available probability-based service life
changes due to combinations of e.g. hydration heat,
approach, has been adopted as service life design
thermal movements, early shrinkage, creep and settle-
methodology. Durability design using DuraCrete was
ments. Thus it is difficult to control the crack width
fufilled by the main designer of Busan-Geoje Fixed
because the evaluation of the strength and stress of
Link project, COWI. Concrete mixes determined by
early aged concrete is not simple. Until now no method
pre-testing showed that they fully satisfy the criteria
of calculating the crack width in early-aged concrete
suggested by DuraCrete design.
is suggested.
For concrete with a good performance for 100 years,
But in Busan-Geoje Fixed Link project cracking of
the cracking of concrete which gives an effect to the
early-aged concrete was controlled by calculation of
durability of concrete should be controlled. There-
crack widths based on the stresses through hydration
fore most of specifications have the requirement of
heat and stress analysis. Crack inspection showed that
the allowable crack width. Many cracks however take
cracking in early-aged concrete can be controlled by
calculating the crack width.
REFERENCES
DAEWOO E&C, Institute of Construction Technol-ogy
(DICT) (2005). Internal report on Concrete Mixing
Design Results for the Busan-Geoje Fixed Link (private
document).
DuraCrete Final Technical Report. (2000). Proba-bilistic
Performance Based Durability Design of Concrete
Structures. Document BE95-1347/R17, European Brite-
EuRam Programme. Published by CUR, The Netherlands.
EuroCode-2 Concrete Structures, Section 4.4.2.4 Calculation
of Crack Width [ENV 1992-1-1:Dec 1991].
Nordtest method NT Build 492 (1999). Concrete, mortar
Figure 1. Chloride diffusivity coefficient of core taken from and cement-based repair materials: Chlo-ride migration
actual structure. coefficient from non-steady migra-tion experiments.
358
Field surveying and FE model updating of a suspension bridge in service
H.-K. Kim & H.-S. Chung

N.-S. Kim
Department of Civil and Environmental Engineering, Pusan National University, Busan, Korea
ABSTRACT Table 1. Selected structural parameters.
A suspension bridge is a structure that sustains dead Upper/Lower Weighting

and live loads through a main cable and stiffening Parameter Bounds factor
girders or trusses. The main cable is a main structural
Area lower chord 15% 1
member for a suspension bridge and, as a result, the
Area upper chord 15% 1
introduced tension in a main cable has to be close to Area diag. member 15% 1
the designed one. However, the tension of the main Area of K truss 15% 1
cable can only be estimated indirectly by measuring Area of main cable 5% 3
the sag. Since the structural parameters of an as-built Tension of cable 5% 1
bridge show some variations, the estimating procedure Mass of main cable 5% 3
should utilize every possible informations provided Mass of Stiff. truss 10% 2
from field. This study proposes a technique for the Rot. mass of Stiff. 10% 2
establishment of an analytical model based on field-
measured data and applied to the Gwangan Bridge in
Korea, to propose an analytical model reflecting the
as-built state of the bridge for the purpose of mainte- Table 2. Natural frequencies of baseline and updated model.
nance. The exploited field-measured data are surveyed
Updated model
coordinates of the target points located in the main
Baseline
cables, stiffening trusses and towers, thermal measure- Identified model Unbounded Bounded
ments at the time of survey, tensions of the hangers, Vibration from
and measured dead loads during construction. Mode AVT (Hz) (%) (Hz) (%) (Hz) (%)
Fig. 1 shows the measured sag and length of center
span of the Gwangan Bridge. 1st lateral 0.1460 0.115 21.48 0.125 14.23 0.125 14.38
An ambient vibration test has been also carried out 1st vert. 0.2435 0.224 8.21 0.244 0.02 0.241 0.90
to identify the dynamic characteristics of the bridge. 2nd vert. 0.2519 0.235 6.73 0.252 0.03 0.253 0.00
Based on the identified natural frequencies, a FE 3rd vert. 0.4630 0.398 13.97 0.432 6.62 0.435 6.13
model updating procedure has been applied to enhance 1st torsion 0.6227 0.576 7.52 0.623 0.02 0.614 1.45
the dynamic reproducibility of the model. The FE
model updating produced closer results in terms of
the frequencies (Table 1 & 2). However, the struc-
tural parameters have undergone greater change than REFERENCES
expected. For this reason, a further consideration is Kim H-K, Lee M-J, Chang S-P. Non-linear shape-finding
being followed in the view point of manual tuning of analysis of a self-anchored suspension bridge. Engrg
a structural system. Struct 2002; 24(12): 154759.
M. I. Friswell and J. E. Mottershead. 1994. Finite Ele-
ment Model Updating in Structural Dynamics. Kluwer
Academic Publishers.
Busan Metropolitan City Facilities Management Authority.
Technical report for the tension measurement of hanger
rope of the Gwangan Bridge, 2004 in Korean.
Figure 1. Measured sag and length of main span.
359
Forth replacement crossing design for safe maintenance

and management
S. Kite
Arup, London, UK
M. Carter & N. Hussain

Arup, Hong Kong
ABSTRACT ACCESS FACILITIES
The Forth Replacement Crossing will be built across The main maintenance access entry to the bridge is at
the Firth of Forth in Scotland to maintain and enhance a the south abutment. A secondary entrance is available
vital transport link in the country.The wide estuary will via the north abutment.
be crossed by a 3-tower cable stayed bridge support- The full length of the deck including the approach
ing a pair of navigational main spans each 650 m, and viaduct and cable stayed bridge is accessible by a pair
an approach viaduct. The scheme design of the cross- of internal deck shuttles. Openings are provided in the
ing, completed by the Jacobs Arup JV, aims to provide top of the cable stayed bridge deck at regular inter-
a fitting 21st century icon, standing alongside exist- vals to gain access to the central zone between the
ing 19th and 20th century Grade A listed bridges. The carriageways.
specifications for the Design and Build contract to be The towers are accessible from the deck and an inter-
let by the client, Transport Scotland, include enhanced nal rack and pinion lift in each tower is the primary
measures to ensure durability and a full suite of access means to gain access above deck level with a stairway
facilities to allow safe maintenance and management provided both below and above deck level to give inter-
of the bridge. nal access over the full height of each tower. Similarly
the piers are accessible from within the deck and inter-
nal ladders / stairs provide access through the height
DESIGN FOR DURABILITY of the pier shafts. Abseiling is envisaged to reach the
external faces of the towers and piers. Suitable struc-
Durability is not an absolute property of a material tural inserts will be provided in the external faces to
but can be affected by both design and construction aid abseiling.
factors. To achieve a satisfactory working design life Permanent moveable access gantries are provided
regular maintenance and some repair work will be below the deck to provide access to the deck sof-
required. Both the required protection (by durability fit. Because of the configuration of the piers, some
design) and mitigation (by maintenance) have been areas of the soffit will not be fully accessible from the
assessed in order to ensure that the design life of 120 gantries, so here the soffit will be accessed by lorry
years can be achieved with a good degree of confi- mounted underbridge inspection vehicles. It is also
dence. For the certain elements, where replacement recommended that a stay cable inspection gantry is
is feasible and cost effective, a shorter service life as provided.
appropriate has been adopted in the design.
Figure 1. General arrangement elevation.
360
R&BD programs on the lifetime design technology for super

long span bridges in Korea
H.-M. Koh & W. Park

Seoul National University, Seoul, Korea
ABSTRACT
Table 1. Super Long Span Bridge R&BD Program.
The need for sea-crossing long-span bridges in Korea
has been recently revitalized by the ambitious plan of Center Korea Expressway Corporation
the government to link some of the 3,000 islands of the Programs 1. Engineering Technology
peninsula to the mainland. For example, the invest- (institute) (Seoul Natl University)
ment of ongoing projects intend to link some major 2. High Performance Material & Application
islands of the southwestern coast with the mainland Technology (Research Institute of
reaches approximately $10 billion until 2010 for the Industrial Science & Technology)
3. High Efficiency Construction Technology
Province of Jeollanamdo. Additional investments for (Korea Institute of Construction Technology)
the construction of bridges are also foreseen until 2025 4. Test Bed & Operating Technology (Korea
to promote economical and social balanced regional Expressway Corporation)
development all over the peninsula. Sponsors Ministry of Land, Transport and Maritime
Thanks to this unprecedented bridge construction Affairs Korea Institute of Construction &
projects, remarkable technological accomplishments Transportation Technology Evaluation and
for long-span cable supported bridges have been Planning (KICTEP)
achieved in Korea. In order to support such enterprise Duration 7 years (2008.122015.12) 3 stages (V3):
and sustain the next generation of bridges to be con- Verification Validation Value Creation
structed in the peninsula, the Korean R&D community Budget 94.6 billion KRW (about 80 million USD)
agreed with the necessity to develop and construct Government (MLTM): 63.6 billion KRW,
a new generation of high performance facilities by Private: 31 billion KRW
means of enhanced materials, advanced structural sys-
tems and technologies as well as upgraded or improved
specifications or standards in a lifetime perspective.
Accordingly, the Korea Ministry of Land, Trans- lifetime design technology (Table 1). Other important
port and Maritime Affairs (MLTM), formally the R&D programs are also presented.
Korea Ministry of Construction and Transportation
(MOCT), launched the Program for 5 Years Plan for
Construction Technology Innovation (20032007) to REFERENCES
strengthen and systematize R&D programs as well as
improve R&D management system. Koh, H.M. 2006. Recent research and development of bridge
This paper introduces recent research and business technology in Korea, Proceedings of International Sym-
development programs for super long span bridges in posium on Sea-Crossing Long-Span Bridges, Invited Lec-
ture, Mokpo, Korea , 1517 February: Korean Group of
Korea. Among the currently ongoing programs, Super
IABSE, 81101.
Long Span Bridge center, a 7 year national R&BD pro- Koh, H.M. and Choo, J.F. 2004. Preparing for the future:
gram launched in 2009, is the largest single program National research programs for the next generation of
dedicated to the development of cable-supported long bridge design and maintenance in Korea, Second Intern.
span bridge technology. Overview and some details Conf. on Bridge Maintenance, Safety and Management,
of the program are summarized with the emphasis on IABMAS 04, Kyoto, Japan, October 1822.
361
Multi-scale system reliability analysis of bridge structures using dominant

failure modes identified by selective searching technique
N. Kurtz & J. Song

University of Illinois at Urbana-Champaign, Urbana, IL, USA
D.-S. Kim
Korea Railroad Research Institute, Uiwang, Kyeonggi-Do, Korea
S.-Y. Ok
ABSTRACT the component events as well as among the identified

failure modes. Lower-scale MSR analyses evaluate the
Bridge structures often have innumerable potential probabilities of the identified failure modes and their
system failure modes, i.e. combinations or sequences statistical dependence. A higher-scale MSR analysis
of local failures. For efficient reliability analysis evaluates the system failure probability based on the
of such complex redundant systems, many research results of the lower-scale analyses. A graphical rep-
efforts have been made to identify dominant fail- resentation of this multi-scale effect can be seen in
ure modes with significant likelihoods, mostly based Figure 2.
on the probabilities of failure modes found dur- This paper presents this decoupled approach in
ing an event-tree search; however, this approach is detail and tests its applicability to complex bridge
time-consuming due to repeated calculations of the structural systems. The efficiency and accuracy of the
probabilities of innumerable failure modes, which method are demonstrated through comparison with
eventually necessitates using heuristic assumptions or Monte Carlo simulations. The results show that the
simplifications. proposed search method skillfully identifies the dom-
Recently, a decoupled approach was proposed: inant failure modes contributing most to the system
dominant failure modes are first identified in the space failure probability, and the multi-scale MSR method
of random variables without system reliability anal- accurately evaluates the system failure probability
yses, then an efficient system reliability analysis is with statistical dependence fully considered. Due to
completed to compute the system failure probability the decoupling between the failure mode identifica-
based on the identified modes. In order to identify tion and the system reliability evaluation, the proposed
dominant failure modes in the decreasing order of their method is expected to be effective for larger structural
relative contributions to the system failure probability, systems.
a simulation-based selective searching technique was
developed by use of a genetic algorithm (Figure 1). The
system failure probability is then computed by a multi-
scale matrix-based system reliability (MSR) method
that can account for the statistical dependence among
Figure 1. Three failure modes identified in the Figure 2. Hierarchical structure of the multi-scale MSR
two-dimensional standard normal space. method.
362
Elasto-plastic behaviors and ultimate strength of continuous

super long-span suspension bridge
K. Nogami
Tokyo Metropolitan University, Tokyo, Japan
A. Okubo
Kanagawa Prefecture, Kanagawa, Japan
Y. Morizono
Chodai Corporation, Tokyo, Japan
M. Nagai
Nagaoka University of Technology, Niigata, Japan
ABSTRACT Table 1. Load factor at the ultimate state of LC2.
Cross-section
In order to realize strait crossing bridge projects of the
next generation, the development of new materials and Single box Two-box + grating
new structural systems for super long span bridges, as
well as economical, durable, rational design and con- Non initial imperfection, 1770 MPa 2.106 2.210
struction method are required. From this background, Initial imperfection, 1770 MPa 2.041 2.157
the consecutive continuous structural system of a Initial imperfection, 2000 MPa 2.090 2.195
3-span suspension bridge was considered first based
on previous experience. However, this form requires
the installation of an intermediate anchorage, and the Table 1 shows the load factor at the ultimate state
number of tower foundations also increases, so this is of loading condition LC2 which gives lowest ultimate
not economical. Hence a multi-span suspension bridge strength based on the analysis results.
is focused as an economical alternative. Since the dead load can be reduced by employing
At present, a wide range of studies has been two-box and grating section, it is clear that the load
reported as part of the planning for the straits cross- factor at the ultimate state increases by 45% com-
ing project, focusing on the structural characteristics, pared to that of the single box section model. The load
the economics, the methods on applying the live factor at the ultimate state of the high strength cable
loads, the buckling characteristics of the central towers model increased by around 2% compared to that of the
(O.Yoshida and T.Moriya 1997), etc., for 4- and 5-span conventional cable model.
suspension bridges. In four super long-span suspension bridge exam-
In the research and development work as summa- ined in this study, an application of the stiffening girder
rized above, there has been no studies investigating having the cross-section with two-box and grating and
the elastic-plastic behavior or ultimate strength of the high strength cable are effective to improve the
the overall structure of multi-span suspension bridges load carrying capacity of the whole bridge system and
(K.Nogami and M.Nagai 2002). Hence, in this study, can maintain the value more than 2.0 in conventional
analytical study was carried out to make clear the required load factor at the ultimate state. Hence, four
elastic-plastic behavior and ultimate strength of the super long-span suspension bridge is able to secure
overall structure of a 4-span suspension bridge having enough safety.
a central span length of 3000 m (K.Nogami A.Someya
and T.Yamasawa 2006). This bridge employed a con- REFERENCES
ventional single box section and a new structural type
consisting of two boxes and the grating as the stiffening O.Yoshida and T.Moriya 1997. Structural characteristics and
girder and of using 1770 MPa and new high-strength feasibility of four suspension bridge, Journal of structural
2000 MPa cables. engineering, Vol. 43A (in Japanese).
The numerical calculation takes into account K.Nogami and M.Nagai 2002. Effect of initial imperfections
both geometric and material non-linearity. The of towers on safety factors of main cables, hangers and
towers in a super long-span suspension bridge, Journal of
materialstress-strain relationship used the incremen- JSCE, No. 717/I-61: 161166 (in Japanese).
tal method in accordance with incremental plasticity K.Nogami A.Someya and T.Yamasawa 2006. Elasto-plastic
theory. Also, the numerical analysis is adopted the behaviors of four long-span suspension bridges and prac-
Newton-Raphson method in combination with the tical rigidity of towers, Journal of structural engineering,
incremental displacement method. Vol. 52A: 901912 (in Japanese).
363
Probabilistic kernel principal component analysis for monitoring

a suspension bridge under environmental variations
C.K. Oh & D. Bae

Kookmin University, Seoul, Republic of Korea
ABSTRACT and unmeasured environmental and operational vari-

ability, and abnormality caused by installation of
Structural health monitoring (SHM) concerns with additional railways is detected in a reliable fashion by
damage detection as well as damage assessment based incorporation of novelty index with GEV distribution.
on the obtained measurements. Since damage causes 10% of randomly selected baseline data are used
changes in the dynamic characteristics of the struc- for computing nonlinear principal components. Then,
ture, it can be considered to be one of abnormality, the entire baseline data are used to compute novelty
which is detected by identifying those changes from index and the associated best-fit GEV distribution. A
the measurements and/or damage-sensitive features statistically meaningful decision boundary is estab-
extracted from the measurements. When the structure lished based on the best-fit GEV distribution. When a
under monitoring experienced abnormality, for exam- new data point is obtained from the structure that may
ple, installation of additional sub-structures, the dam- experience abnormality, the corresponding nonlinear
age detection strategies can also be applied to detecting principal components are computed. Those nonlin-
the abnormality which is not caused by damage. ear principal components are employed to calculate
In reality, however, time-varying environmental and novelty index from the baseline data whose princi-
operational conditions can also affect the dynamic pal components have the minimum Euclidean distance
properties of the structure, and slight changes caused from those of new data. Finally, the number of outliers
by abnormality may be blurred or undetected. There- whose values are beyond the pre-decided threshold
fore, it is necessary to develop a technique to sepa- is counted, and a sudden increase of the number
rate the effects of abnormality from those caused by of outliers is utilized to indicate the occurrence of
environmental and operational variations. abnormality.
Data normalization is a procedure to separate the This study demonstrates that application of a prob-
effects caused by time-varying environmental and abilistic data normalization technique, i.e., KPCA
operation conditions from measured signal changes, incorporated with novelty index and GEV distribution,
so that possible abnormality can be detected in the to an in-field Yeongjong grand bridge demonstrates its
presence of environmental and operational variation feasibility to detect abnormality under time-varying
(Sohn & Oh 2008, Oh & Sohn 2009). environmental and operational conditions without
In this study, kernel principal component analysis explicitly measuring them.
(KPCA) incorporated with novelty index and gen-
eralized extreme value distribution is proposed to
detect abnormality caused by installation of additional REFERENCES
railways in the presence of environmental and oper-
Oh, C.K. & Sohn, H. 2009. Damage diagnosis under envi-
ational variations. The proposed method is applied
ronmental and operation variations using unsupervised
to an in-service Yeongjong grand bridge in Korea to support vector machine. Journal of Sound and Vibration
investigate its feasibility. For the baseline and test 325: 224239.
data, hanger tensions measured on 2003 and 2004, Sohn, H. & Oh, C.K. 2009. Statistical Pattern Recognition in
respectively, are employed. KPCA characterizes the Structural Health Monitoring. Encyclopedia of Structural
hidden relationship between measured hanger tensions Health Monitoring. Wiley.
364
Integrated geometry control system for cable stayed bridge: Application

to Incheon bridge
H.Y. Shin, D.K. Im, J.K. Yoo & C.H. Kim

Samsung Engineering & Construction, Seoul, Republic of Korea
ABSTRACT The integrated geometry control system integrated

the analysis data by structural analysis program and
The integrated geometry control system in the long- measured data collected by wireless system at each
span steel deck cable-stayed bridge has been developed construction stage.The engineer could analyze the data
and applied to Incheon Cable Stayed Bridge. This sys- efficiently and make the proper decision for adjustment
tem consists of 4 systems which are data integrated at an earliest time at each stage by using this system.
management system, structural analysis system, error The integrated geometry control system made it
adjustment system and measurement system. The data possible to minimize the numbers of cable retension
integrated management system stores the numerous and controlled the final geometry and tension error
analysis data and measured data into its system and within the target ranges.
shows them graphically to the engineer. It allows the
engineer to understand and compare the necessary
data more easily and quickly. The structural analysis REFERENCES
systems is developed for the structural analysis with
considering the geometric nonlinearity and modeling Jung, I.H. and Park, C.M., 2005. Incheon Bridge Design Stan-
dards. Magazine issued by Korean Society of Steel Con-
the cable as elastic catenary cable element. The error
struction, Korean Society of Steel Construction, Vol. 17,
adjustment system is developed to analyze the differ- No. 3, Serial Number 64: 816.
ences between the designed values and measured ones Kim, K.S. and Lee, H.S., 2001. Analysis of target geometry
of the geometry and cable tension and then to minimize under dead loads for cable supported bridges. Comput-
the differences by adjustment of the cable length. The ers & Structures, Vol. 79, Issue 2930: 26812692.
measurement system is combined to collect and show Wang, P.H. and Tseng, T.C., 1993. Initial Shape of
the required measurement data for the geometry con- Cable-stayed Bridges. Computers & Structures, 46(6):
trol such as temperature, displacements of girder and 10951106.
pylon, cables tensions promptly. Each system is inte- Kaneyoshi, M.M. and Tanaka, H., 1990. New System Iden-
tification Technique Using Fuzzy Regression Analysis.
grated and managed in order to combine the systems
Uncertainty Modeling and Analysis; proc. 1st intern.
organically. symp. IEEE., Maryland: 528533.
In incheon bridge construction, 4 numbers of cables Jung, K.J. and Lee, H.S., 2008. Development of Geometry
in the side-span and center-span have been erected and Control System for Cable Stayed Bridge and Application
adjusted at the same time along with the cantilevered to the Incheon Bridge. IABMAS 2008, Korea.
deck erection in the center span under 67 days of
cycle time.
365
Redundancy analysis for a cable-stayed bridge using fibre model
K. Sugioka
Osaka Business and Maintenance Department, Hanshin Expressway Company Limited, Osaka, Japan
H. Kanaji
Construction Management Headquarters, Hanshin Expressway Company Limited, Osaka, Japan
S. Matsumoto
Planning Division, Hanshin Expressway Management Technology Center, Osaka, Japan
K. Magoshi
Seismic Analysis Research Inc., Nagoya, Japan
M. Nagai
Nagaoka University of Technology, Niigata, Japan
ABSTRACT
Tempozan Bridge is a 3-span continuous double-plane

fan-system multiple-cable-stayed steel bridge with a
centre span of 350 m and a total length of 640 m, as
shown in Figure 1. The cables and some members are
highly redundant, while damage to some other mem-
bers like pendel bearings is highly likely to lead to
reduced safety performance of the whole structure. Figure 1. General view of the bridge under analysis.
To achieve a rational maintenance of a bridge within
a limited budget with the life cycle cost and life-
time expected risk of the bridge taken into account,
it is necessary to determine an optimum inspection
program through comprehensive investigation on vari-
ous factors including impacts of damage of individual
members on safety performance of the whole struc-
ture and costs required for repair, strengthening and
inspection.
For the purpose of determining limit states of a Figure 2. Analysis model.
long-span bridge and causal phenomena engineer-
ingly, the authors conducted static elasto-plastic finite
displacement analysis. Sensitivity analysis was per-
formed under incremental loads with a certain level of
corrosion given to some members, and aging analy-
sis was performed under constant loads with sectional
area of individual members progressively reduced. The
model used in the analysis was a global model con-
sisting of shell elements for the main girder and fibre
elements for the main towers and end piers, as shown
in Figure 2. The analysis also used a technique of
including the effects of local buckling in the material
constitutive laws so that the dead load state was suc-
Figure 3. Aging analysis results (Vertical displacement at
cessfully reproduced with the effects received during the centre of the centre span).
the erection taken into account.
As a result, it was quantitatively demonstrated that influence of damage in the main girder or cables was
damage in the main girder and cables had most signif- significant on the vertical displacement at the centre of
icant impacts on the safety performance of the whole the centre span, although yield points were not reached
structure. The aging analysis (in Figure 3) revealed that at a corrosion of 9 mm.
366
Analytical study on detection of structural damages of a long-span

suspension bridge by wind-induced response
N.D. Thang, H. Katsuchi, H. Yamada & E. Sasaki

Yokohama National University, Yokohama, Japan
ABSTRACT This paper presents the analytical study on wind-

induced responses of a long-span suspension bridge in
Nowadays, a great number of long-span bridges were case of some kinds of structural damage were occur.
constructed all over the world. Most of long-span In this paper, a very detailed finite element model of
bridges which are common over the sea are partic- a long-span bridge was developed and verified using
ularly difficult to maintain because of their specific field data, making this model as accurate as possible
conditions: a severe natural environment including in representing the actual structural behavior. Using
strong winds, strong tidal currents and salt air, a large this finite element model, the reliability analysis of the
degree of continuing deformation of structures, an bridge is performed considering dead and wind loads.
extremely large variety of structural members and After that, based on the realistic deteriorations, vari-
materials, and the need to cope with the fatigue of ous types of structural damages of a long-span bridge
structural steel especially in the case of any bridges are simulated to facilitate the discussion. All of the
which carry trains as well as road traffic. However, dynamic data for comparing damaged with undam-
their service time have to be more than 100 years aged cases were generated numerically from the finite
because these bridges are very expensive to design, element model. Lessons archived from this study are
construct and maintain. As a result, the health, dura- expected not only to maintain this bridge but also to
bility, and safety of these bridges in a long-term improve our understanding of the real bridge perfor-
service period are now attracting a lot of scientists mance as well as to provide useful feedbacks for future
and engineers. An issue arising will be a methodology design.
how the structural damages can be detected from the
monitoring data. REFERENCES
Besides, with increasing of span and slenderizing
of structure, long-span bridges become more and more Boonyapinyo, V., Miyata, T. & Yamada, H. 1999. Advanced
sensitive to wind. For a long-span bridge, with limited Aerodynamic Analysis of Suspension Bridges by State-
torsional stiffness, wind-induced forces, such as self- space Approach. Journal of Structural Engineering,
excited force and buffeting force, can cause destructive ASCE, Vol. 125(12): 13571366.
phenomena and need special attention. Self-excited Katsuchi, H., Jones, N. P. & Scanlan, R. H. 1999. Multimode
couple flutter and buffeting analysis of the Akashi Kaikyo
forces causing flutter are in general dependent on the
Bridge. Journal of Structural Engineering, ASCE, Vol.
geometric profile of the bridge deck section, angle of 125(1): 6070.
wind attack and wind velocity expressed as reduced Li, Z.X, Zhou, T.Q, Chan, T.H.T & Yu, Y. 2006. Multi-scale
frequency. Meanwhile, buffeting is defined as the numerical analysis on dynamic response and local damage
unsteady loading of a structure by velocity fluctuations in long-span bridge. Journal of Engineering Structures 29
in the oncoming flow. (2007), Elsevier: 15071524.
Although many advances in design, construction as Simiu E. & Scanlan R H. 1996. Wind Effects on Structures
well as maintenance have been developed day to day, Fundamentals and Applications to Design, Third Edition.
many problems of structure still remain unknown or John Wiley & Sons, Inc., New York.
Yanaka, Y. & Kitagawa, M. 2002. Maintenance of steel
unsolved. The ability to detect structural damages in
bridges on Honshu Shikoku crossing. Journal of Con-
a bridge before it endangers the structure has been structional Steel Research 58: 131150.
of interest to engineers for many years. However, few Yeung, W.T. & Smith, J.W. 2005. Damage detection in bridges
studies were carried on to investigate how the damages using neutral networks for pattern recognition of vibration
of structure affect on wind-induced response of a long- signatures. Journal of Engineering Structures 27 (2005),
span bridge. Elsevier: 685698.
367
New high performance steels for long-span bridges
T.-Y. Yoon, J.-K. Lee & K.-S. Kim

Research Institute of Industrial Science & Technology, Hwaseong, Korea
J.-Y. Cho
POSCO Technical Research Laboratories, Pohang, Korea
ABSTRACT girders, and a pin-connecting steel modular bridge

system investigated in the HiperCONMAT, are pre-
This paper presents developing process, material prop- sented. HSB800 (tensile strength of 800MPa) which
erties, application and economic benefit of newly are under development for long-span bridges is also
developed High-Performance Steel for Bridges (HSB) introduced.
in Korea.
There were remarkable improvements of steel pro-
duction which brought developments of high per- REFERENCES
formance steel for bridges (HPS) in US, Japan, AASHTO, 2006. AASHTO LRFD Bridge Design Specifica-
Europe and Korea from 1990s. Although they have tion, 2006 Interim.
different names, i.e., HPS50W/70W/100W in US, British Standard Institution, 2005. Eurocode 3: Design
BHS500/700 in Japan, S460M/S690Q in Europe and of Steel Stuctures, Part 1-10: Material toughness and
HSB500/600/800 in Korea, they have possessed sim- through-thickness properties, BS EN 1993-1-10:2005.
ilar improved strength, weldability, higher toughness, British Standard Institution, 2005. Guide to methods for
and better weathering and fabrication characteristics. assessing the acceptability of flaws in metallic structures,
HSB has been developed through the research of BS 7910: 2005.
Burdekin F.M. and Hamour W., 2002. Partial safety fac-
the High Performance Construction Material Research
tors for SINTAP procedure, Off-shore Technology Report
Center (HiperCONMAT) sponsored by the Korea 2000/020, Health and Safety Executive.
Government. The targeting specifications of HSB were Cho JY, Um KK, Choi JK, 2009. Development and Applica-
derived from the opinions of the steel users and vari- tion of High Performance Steel, HSB600, Asia steel 2009,
ous experts, such as designers, academia, fabricators May 2427, Busan, Korea.
of steel bridges, who require higher strength, higher Coldren A. P., Biss V. and Oakwood T.G., 1981. Thermome-
toughness, and better weldability. HSB is produced by chanical Processing of Mi-croalloyed Austenite, AIME,
applying thermo-mechanical control process (TMCP) Pittsburgh, p.591.
and optimizing chemical compositions resulted higher Coldren A. P., Cryderman R.L., Semchyshen M., 1967.
Transformation and Hardenability in Steels, Climax Moly.
tensile strength and no reduction of yield strength
Comp., Ann Arbor, MI, p.15.
for thick plate. High toughness of 47J at 20 or Cota A.B., 2000. Materials Characterization, Vol. 44, p.291.
40 reduces occurrence of brittle fracture at low tem- Kasamatsu Y., Takashima S., Hosoya T., 1979. Tetsu-to-
perature and improves cold-formability. In addition, Hagane, 1979, Vol. 65, p.92.
the low Pcm and Ceq of HSB enhance weldability, Korean Agency for Technology and Standards, 2007. Rolled
that is, reduce cold cracking and lower preheating Steels for Bridge Structures (KS D 3868), Korea Industrial
temperature. Standards Commission (in Korean).
HSB500 (tensile strength of 500 MPa) and HSB600 Ministry of Land, Transport and Maritime Affairs, 2004.
(tensile strength of 600 MPa) were successfully devel- Korean railway bridge design code, KSCE (in Korean).
Ministry of Land, Transport and Maritime Affairs, 2005.
oped and registered in Korean Industrial Standards
Korean roadway bridge design code, KRTA (in Korean).
(KS D 3868) and added to the new Korean roadway Sedlacek G., Stranghner N., Schfers M., Dahl W., Langen-
bridge design code. Until now 46,800 tons of HSB has berg P., Kalinowski B., Brozzetti J., Nussbaumer A., 1998.
been used for bridge construction from January 2010. Composite Bridge Design Improvement for High Speed
In order to use HSB to bridge structures more effec- Railways Determination of permissible plate thicknesses
tively and widely, innovative steel bridge systems, to avoid brittle fracture, ECSC Project 7210/SA/128,
such as hybrid girders, new steel-concrete composite Institute of Steel Construction, RWTH Aachen.
368
Special Sessions
SS1: Use of health monitoring for life-cycle
cost analysis & optimization
Organizers: D.M. Frangopol & A. Orcesi
Optimization of the amount of structural health monitoring via probabilistic

system assessment
C. Albrecht, D. Hosser, C. Klinzmann & R. Schnetgke

Institute of Building Materials, Concrete Construction and Fire Protection,
Technische Universitt Braunschweig, Braunschweig, Germany
ABSTRACT methodology is tested for different monitoring periods

to identify variations of the extreme value distributions
Nowadays, the structural condition of newly erected among the monitoring datasets. The different distri-
important elements of the infrastructure is monitored butions found were tested for their influence on the
using complex structural health monitoring (SHM) reliability index when being used in the LSF.
systems. SHM systems are as well used to guaran- In an exemplary application, a steel truss bridge is
tee the safety of old structures that show significant modeled as a serial system, so that the overall system
defects. A continuous SHM leads to a large amount reliability is highly influenced by the component with
of data which makes the evaluation and interpretation least reliable component, which can be identified by
very difficult. looking at the highest sensitivity.
One possibility to optimize the SHM process is A way to identify and locate potential damage is to
to adapt the amount of monitoring according to the compare the sensitivities of the components in subse-
condition of the structure. A continuous monitoring quent evaluations. A strongly decreasing sensitivity of
procedure is only reasonable if a further increase of a component can be an indicator of load relocation,
damage would lead to serious problems up to a struc- which in conclusion might indicate that a damage has
tural collapse. The amount can be reduced to a periodic occurred in that component or in the vicinity.
monitoring in case the structure shows only minor Load relocation away from damaged components
damage or still has enough reserves. can lead to a calculated increase of the system reliabil-
Normally, no quantitative information about the ity. Hence, this increase can also be used as an indicator
condition of the structure is directly gained from for damage assessment. The methodology introduced,
the measurements. This paper suggests a methodol- and especially the calculation of the reliability index of
ogy that utilizes a probabilistic method to quantify the structural system and the sensitivities of impor-
the actual condition of the structure based on the tant parameters, can be used to adapt/optimize the
time-history of collected monitoring data. SHM process in terms of monitoring periods/amount
The structural system is described in a risk- of monitoring, and to evaluate the structural con-
orientated probabilistic model based on identified dition (i.e. damage) of different components in an
potential weak points. The limit state functions for infrastructure network.
these weak points incorporate measurable parameters,
e. g. strains or deformations, which makes the model
easy to update in case new data is available or damage REFERENCES
is identified. The structural condition is quantified by
applying appropriate reliability methods (First/Second Fischer, L., 1999: Sicherheitskonzept fr neue Normen ENV
und DIN-neu: Grundlagen und Hintergrundinformationen
Order Reliability Method, Monte Carlo Simulation) to
Teil 11 A, Bautechnik 76(1999), Nr. 12.
the model of the structural system. Klinzmann, C., 2008: Methodik zur computergesttzten,
The time-dependent variables (mainly the monitor- probabilistischen Bauwerksbewertung unter Einbeziehung
ing data) can be accounted for by a statistical distribu- von Monitoring, Heft 203, Braunschweig:iBMB TU BS.
tion that is fitted to extreme values that occur within a Melchers, R.E., 2002: Structural Reliability Analysis and
reference period. Hence, the time-dependent problem Prediction, Second Edition, Chichester:Wiley.
is traced back to time-independence and the relia- Spaethe, G., 1992: Die Sicherheit tragender Baukonstruktio-
bility methods mentioned above are applicable. This nen, Vienna (Austria):Springer-Verlag.
371
The potential link between bridge management systems, structural health

monitoring and bridge weigh-in-motion progress and challenges
R.M. Ellis
P.D. Thompson
ABSTRACT BMS with a view to help bridge owners better manage

risk on the bridge network.
Bridge management systems (BMS) have been devel- This discussion paper presents a synopsis of the
oped in many parts of the world. The majority of potentials for linking bridge management systems
systems manage inventory and inspection information (BMS), bridge health monitoring systems (BHM),
and many also provide analysis capabilities to fore- and bridge weigh-in-motion (BWIM) systems. Sug-
cast deterioration, recommend preservation activities gestions are provided for BMS, BHM and BWIM
(such as maintenance, repair, and rehabilitation), func- developers to enhance and develop new ideas. It is
tional improvements (strengthening and widening), hoped that this research will further advance the con-
and structure replacement. These advanced BMS offer cepts of the integration of these systems and provide
powerful predictive and cost-benefit analysis tools that suggestions to BMS, SHM and BWIM system devel-
help determine the most cost-effective activities on opers for future development, and benefit agencies
individual bridges and over a network. With few excep- who are already implementing monitoring solutions
tions, these BMS all rely on visual bridge inspection or who are considering monitoring projects.
techniques and inspector judgment, are heavily depen-
dent on adequate inspector training, and are performed
at discrete intervals.
Structural health monitoring (SHM) is being REFERENCES
applied on bridge structures around the world. These Antoniou, C. 2005. Integration of Weigh-in-Motion Tech-
systems vary from large specialized monitoring sys- nologies in Road Infrastructure Management. ITE Jour-
tems on signature bridges, to more basic systems nal. Institute of Transportation Engineers.
designed to monitor a specific problem on a particular AASHTO. 1992. AASHTO Guidelines for Bridge Man-
structure, or systems having various other purposes. agement Systems. Washington: Transportation Research
Advances in data acquisition, storage, remote sensing, Board of the National Academies.
and structural instrumentation have made it possible Fu, G., Feng, J., Dekelbab, W., Moses, F. Cohen, H., Mertz, D.
to conveniently and remotely monitor many things & Thompson, P. 2003. Effect of Truck Weight on Bridge
Network Costs. National Cooperative Highway Research
including strains and deformations; cracking occur- Program Report 495. Transportation Research Board of
rence and propagation in concrete as well as steel the National Academies.
structures, accelerations \ vibrations, and of course Gastineau,A., Johnson,T. & Schultz,A. 2009. Bridge Health
environmental conditions such as wind speed, temper- Monitoring and Inspections A Survey of Methods.
ature, and humidity. In the last few years bridges are Minnesota Department of Transportation Report MN/RC
being instrumented for weigh-in-motion (BWIM) to 2009-29.
record truck axle loads and frequencies. Honefanger, J., Strawhorn, J., Athey, R., Carson, J., Con-
Despite obvious advantages, there has been little ner, G., Jones, D., Kearney, T., Nicholas, J., Thurber, P. &
progress in the integration of SHM and BMS. This Woolley, R. 2007. Commercial Motor Vehicle Size and
Weight Enforcement in Europe. FHWA Report PL-07-
paper explores the synergies that exist in merging the 002, US Federal Highway Administration.
capabilities of BMS with the information made possi- Thompson, P.D., Ellis, R.M., Merlo, T. & Hong, K.,
ble with SHM. Challenges are presented to researchers 2003. Implementation of the Ontario Bridge Management
and industry to develop means to provide the type of System. 9th International Bridge Management Confer-
condition based data that BMS need, and to develop ence 2003. Washington, D.C.: TRB, National Research
ways to utilize existing available SHM data to improve Council.
372
Probabilistic optimal bridge monitoring planning
S. Kim & D.M. Frangopol

Lehigh University, Bethlehem, Pennsylvania, USA
ABSTRACT
Structural performance cannot be assessed and pre-

dicted accurately due to various epistemic and aleatory
uncertainties. Recently, structural health monitoring
(SHM) has been considered as a significant tool to
reduce the uncertainties associated with structural
performance assessment and prediction. Under lim-
ited financial resources, optimal monitoring planning
should be considered.
In this paper, probabilistic methodologies to estab-
lish optimal monitoring planning with uniform and
non-uniform monitoring time intervals are introduced.
For the monitoring plan with uniform time interval,
the statistics of extremes and availability theory are
used. A bi-objective optimization based on minimiz-
ing the total monitoring cost and maximizing the
availability of the monitoring data for performance
prediction is formulated. Results of this formulation
are shown in Figure 1. For the non-uniform time inter-
val monitoring, the optimal monitoring planning is
formulated as the minimization of the expected dam-
age detection delay. The two proposed approaches
can be used as effective tools for integration of SHM
into cost-effective lifetime maintenance strategies of
deteriorating structures.
Figure 1. (a) Pareto solution set; (b) monitoring plans of
solutions A, B, C, and D in (a).
REFERENCES
Ang, A.H.-S. & Tang, W.H. 1984. Probability Concepts in
Engineering Planning and Design. Vol. II, John Wiley & policies. International Journal of Quality & Reliability
Sons. Management, Emerald, 12(3): 1931.
Enright, M.P. & Frangopol, D.M. 1999. Condition prediction Kim, S. & Frangopol, D.M. 2010. Optimal planning of struc-
of deteriorating concrete bridges using Bayesian updating. tural performance monitoring based on reliability impor-
Journal of Structural Engineering, ASCE, 125(10): 1118 tance assessment.Probabilistic Engineering Mechanics,
1125. Elsevier, 25(1): 8698.
Frangopol, D.M., Lin, K.Y., & Estes, A.C. 1997. Life- MathWorks Inc. 2009. Optimization ToolboxTM 4 Users
cycle cost design of deteriorating structures. Journal of Guide. The MathWorks, Inc., USA.
Structural Engineering, ASCE, 123(10): 13901401. Mori, Y. & Ellingwood, B. R. 1994. Maintaining reliability of
Huang, B.-S. & Chiu, H.-N. 1995. The quality management concrete structures. I: Role of inspection/repair. Journal
of the imperfect production process under two monitoring of Structural Engineering, ASCE, 120(3): 824845.
373
Optimization of bridge maintenance actions considering combination

of sources of information: Inspections and expert judgment
L.C. Neves
UNL/FCT Universidade Nova de Lisboa, Lisbon, Portugal
D.M. Frangopol
ABSTRACT with different life-cycle costs, from which the deci-

sion maker can choose, considering available budget
Due to the high costs associated with maintenance and performance alternatives.
of existing structures, but also the significant uncer- The methodology is applied to a set of bridge com-
tainty present in any prediction of future performance, ponents. The results show the significant improvement
the use of advanced probabilistic life-cycle deterio- in prediction quality obtained by combining differ-
ration models is fundamental in bridge maintenance ent sources of information, even if the quality of the
and management. However, such models must be obtained data is limited. Moreover, the impact of the
based on direct and indirect information on structural new information on the optimal maintenance strategies
performance, including results of inspections, health is also evaluated.
monitoring, but also expert judgment and information Overall, preventive maintenance actions prove to be
on other similar structures. In the proposed approach, an extremely useful tool for reducing life-cycle cost,
a probabilistic deterioration model, considering both and essential maintenance actions are paramount in
condition and safety as indicators of performance, is limiting the effects of deterioration, keeping structures
used as a decision aid in bridge management. The data fit for use.
considered for defining the performance profiles over
time is based on the expert opinion, resulting from
observation of similar structures.
In order to reduce uncertainty and correct the exist- REFERENCES
ing predictions, the results of direct information, in the
form of visuals inspections, are combined with initial Frangopol, D.M., Kong, J., and Gharaibeh, E. (2001).
prediction. Visual inspections are not free of uncer- Reliability-based life-cycle management of highway
tainty. Therefore, a probabilistic framework must be bridges. Journal of Computing in Civil Engineering,
developed. The combination of these two sources of 15(1):2747.
Neves, L.C. and Frangopol, D.M. (2005). Condition, safety,
information (i.e., initial prediction and visual inspec-
and cost profiles for deteriorating structures with empha-
tion) is performed considering Bayesian updating sis on bridges. Reliability Engineering & System Safety,
techniques, using Monte-Carlo simulation. 89(2): 185198.
In this manner, it is possible to improve the knowl- Frangopol, D.M. and Neves, L.C. (2008). Structural perfor-
edge on the performance of the structure, but also mance updating and optimization with conflicting objec-
reduce the uncertainty in several parameters, including tives under uncertainty. In 2008 ASCE-SEI Structures
initial performance level, deterioration rate and effect Congress - Crossing Borders. Vancouver, Canada, April.
of maintenance actions. Neves, L.C., Frangopol, D.M., Petcherdchoo, A. (2006).
Based on this new information, it is possible, con- Probabilistic lifetime-oriented multi-objective optimiza-
tion of bridge maintenance: Combination of mainte-
sidering multi-objective optimization using genetic
nance types. ASCE Journal of Structural Engineering,
algorithms, to define the optimal lifetime maintenance 132(11):18211834.
strategy for a structure. The three objective functions Neves, L.C., Frangopol, D.M., Cruz, P. (2006). Probabilis-
considered are: life-cycle cumulative cost, condition tic lifetime-oriented multiobjective optimization of bridge
index, and safety index. The optimization procedure maintenance: Single maintenance type. ASCE Journal of
will provide a set of optimal solutions, associated Structural Engineering, 132(6):9911005.
374
Structural health monitoring: from data acquisition to optimum life-cycle

management under uncertainty
A.D. Orcesi
D.M. Frangopol
ABSTRACT
Cumulative expected failure cost ( $105 )

50 50
= 4% Cumulative expected maintenance cost
The objective of the proposed framework is to pro-
maintenance cost ( $104 )

40 Cumulative expected 40
vide an efficient decision making tool for optimization
Cumulative expected
failure cost
of maintenance strategies of bridges based on struc- B
tural health monitoring (SHM) information. Recent 30 30
improvements of sensor technologies have made pos- A

sible the use of monitoring data in structural perfor- 20 20
mance analyses. There is a huge interest in including A
maintenance and SHM information in life-cycle cost 10 10
analysis (Okasha & Frangopol 2009, Kim & Frangopol B
2010). Indeed, by providing additional structural infor-

0 0
mation, parameters of the random variables associated 1 5 10 15 20 25 30 32
with resistance and loading characteristics can be bet- Time after monitoring at year 2004 (years)
ter quantified (Orcesi & Frangopol 2010, Orcesi et al.
2010). Therefore, best maintenance strategies can be Figure 1. Comparison of cumulative expected failure cost
used. By improving the knowledge on structural per- and cumulative maintenance cost between two optimum
formance, SHM should enable to control maintenance maintenance solutions ( = discount rate of money).
cost and allocate financial funds without compromis-
ing structural reliability.A multi-state reliability-based terms of cumulative costs. The cumulative expected
analysis process is proposed by considering fatigue, failure cost is larger for solution A. However, the
serviceability and ultimate limit states. Controlled load cumulative expected maintenance cost is larger for
tests and long-term monitoring information are used solution B.
to update some parameters in the structural reliability
analysis.
REFERENCES
An optimization procedure that integrates SHM
information and deals with various limit states is Kim S., & Frangopol D.M. 2010. Optimal planning of struc-
used to determine optimal maintenance strategies. The tural performance monitoring based on reliability impor-
design variables are times and types of maintenance tance assessment. Probabilistic Engineering Mechanics,
actions and the objective is to minimize both failure Elsevier; 25(1): 8698.
and owner costs such that reliability constraints are Mahmoud, H.N., Connor, R.J. & Bowman, C.A. 2005. Results
satisfied. It is shown that optimal solutions cannot be of the fatigue evaluation and field monitoring of the
seen as a juxtaposition of optimal results for different I-39 Northbound Bridge over the Wisconsin River. ATLSS
Report No. 0504, Lehigh University, Bethlehem, PA.
limit states taken individually and that lifetime man-
Okasha N.M., & Frangopol D.M. 2009. Lifetime-oriented
agement has to be considered as a whole at the scale of multi-objective optimization of structural maintenance,
the civil infrastructure and not for each failure mode considering system reliability, redundancy, and life-
considered individually. cycle cost using GA. Structural Safety, Elsevier, 31 (6):
The I-39 Northbound Bridge over the Wisconsin 460474.
River is used as an example to show how the monitor- Orcesi, A.D., & Frangopol, D.M. 2010. Inclusion of crawl
ing information (available in 2004 after the instrumen- tests and long-term health monitoring in bridge service-
tation and long-term monitoring of the bridge during ability analysis. Journal of Bridge Engineering, in press.
four months, see Mahmoud et al. (2005)) can be inte- Orcesi, A.D., Frangopol, D.M., & Kim, S. 2010. Optimization
of bridge maintenance strategies based on multiple limit
grated in a general framework for optimization of
states and monitoring, Engineering Structures, in press.
maintenance strategies.
Figure 1 presents the comparison between two
Pareto optimum maintenance solutions, A and B, in
375
The role of structural identification in asset management
J. Weidner, J. Prader, N. Dubbs, F. Moon & E. Aktan

Drexel University, Philadelphia, PA, US
ABSTRACT It is in obtaining quality, reliable data, that structural

identification (St-Id) can play a role in an AM program.
A large portion of the nations bridges has exceeded its It is difficult for an engineer to associate the qualitative,
original design life. The loads to which these bridges visual appearance of a structure with hard numbers like
are subject to are often much larger than the design- capacity. Often there is too much uncertainty resulting
ers anticipated. Todays typical State DOT is juggling from the complexity and redundancy typical to civil
a massive number of structures with limited person- structures.
nel and inadequate budgets on a daily basis. As such, St-Id is by no means the solution to the nations
the maintenance of bridges cannot be focused on aging infrastructure problem. The in-depth St-Id pro-
prevention. Problems are addressed as they appear cess is far too expensive and time consuming to apply
based on urgency. The combination of large popula- in a global manner to a large population of structures.
tions alongside budgetary and personnel constraints It follows that full-scale St-Id is most applicable to a
put considerable importance on every decision made single or small population of assets.
by the bridge owner. There are tools available to help The application to single assets will be presented in
inform decisions, such as structural identification (St- the context of a review of five bridges tested over the
Id) and life-cycle cost analysis. Individually these tools past several years. The details of these tests including
are effective, but problems still arise with application modeling, experimental design, and testing are pre-
as well as candidate selection. It is therefore beneficial sented in various conference papers (Prader et al 2008,
for a bridge owner to consider an asset management Weidner 2009a,b). The results of these tests, in terms
(AM) system, which can utilize both St-Id and life- of the decisions made by the bridge owner are sum-
cycle cost analyses to help optimize where and when marized and discussed briefly herein as an example of
funds are allocated to projects in an effort to get the the benefits of St-Id.
most out of the both the budget and the bridge popula- The St-Id process is viable for large populations,
tion. AM systems are entirely dependent on the quality with adjustment to each step. This sparse, rapid version
of data which is utilized to influence decision making. of St-Id can provide excellent performance metrics for
The process of St-Id as a means of mitigating uncer- incorporation into an AM system.
tainty in the analysis and basic decision-making for a
single structure is an established and reliable tool for
an engineer or bridge owner. The authors have been
utilizing St-Id over the past 20 years for structures REFERENCES
varying from highway overpasses to signature long-
span bridges. However, the application of St-Id as a Prader, J & Moon, F.L. 2008. Load Testing, Analysis and
tool for management of a large population of assets is Structural Health Monitoring for Evaluating the Load
a relatively new concept. This paper will address how Capacity of Aged RC Bridges, International Bridge Con-
St-Id can best be utilized within the umbrella of AM ference, Pittsburgh, PA, June 2008.
for a population of similar bridges; a common situa- Weidner, J. et al. 2009a. Structural Identification of Bridges
to Assess Safety and Performance, ASCE Structures
tion a State DOT may face. The foundation of using Congress, Austin, TX April 2009.
St-Id in AM is to estimate vulnerability of a popula- Weidner, J. et al. 2009b. Safety Evaluation and Load Capacity
tion of structures in order to assess risk and facilitate Rating of Aged and Deteriorated Concrete Bridges with
the determination of appropriate candidates for further Missing Plans, ASNT Fall Conference, Columbus, OH,
investigation from a cost-benefit perspective. October 2009.
376
SS2: Safety monitoring & maintenance
strategy for long span bridges
Organizer: A. Chen
Safety monitoring of the cable stayed bridge in the Commercial Harbor

of Venice, Italy
B. Briseghella
College of Civil Engineering, Fuzhou University, Fuzhou, China
E. Siviero, C. Lan & E. Mazzarolo

University IUAV of Venice, Italy
T. Zordan
College of Civil Engineering, Tongji University, Shanghai, China
ABSTRACT
Theoretical and experimental investigation (ambient

vibration tests and permanent monitoring system)
on the structural behavior of a curved cable-stayed
bridge is presented in the paper. The quoted bridge
was erected in the Commercial Harbor of Venice-
Marghera, Italy, and opened to traffic on January 2007.
The bridge is characterized by six curved spans, of
which the main two of 105 m and 126 m are cable-
stayed, with the stays arranged on a single plane and
connected to the centerline of the composite steel and a dynamic point of view. Specific parameters able to
concrete deck. Furthermore, the pre-stressed concrete describe the evolution of the structural response related
pylon is single and inclined. As a consequence of the to the presence of a certain level of damage were
spatial structural arrangement, the dynamic behavior introduced. With reference to the dynamic analysis,
of the deck appears very complex both in bending and the variation of natural frequency, the Modal Assur-
torsion. ance Criterion (MAC) and the Normalized Modal
The peculiar architectural layout of the bridge added differences (NMD) were considered as characterizing
significant difficulties in both accurate structural anal- parameters for the structural health assessment.
ysis and assessment of as built behavior. Hence Concerning the static analysis the variation of dis-
ambient vibration modal testing were carried out after placement levels was evaluated. Direct comparison
erection and a detailed FE model was tuned and val- of results emphasized that, for the considered case,
idated on such data. The special geometric layout of static monitoring appears more sensitive to damage
the bridge and its infrastructural role provided also a propagation inside pylon and cables, compared to
strong motivation for the installation of a permanent dynamic one.
monitoring system. The paper describes the character- For the case presented, cable damage is not clearly
istics of the monitoring system, based on the use of detectable: for this reason, the correct and reliable
the SOFO fiber optic sensor family and conceived instrumentation of each cable with a dedicated sensor
for both static and dynamic monitoring. A number of plays a fundamental role.
considerations are carried out on the capability of the
quoted monitoring system to assess in due time the
REFERENCES
presence of unexpected damage on the most critical
zones of the bridge, namely inclined pylon and stay- Inaudi, D. 1994. Low-coherence Deformation Sensors for the
cables. Specific defectiveness index for describing the Monitoring of Civil-engineering Structures; Sensors and
possible loss of pretension in such bridge components Actuators A, 44: 125130.
have been defined. Data provided from cables sensor Brincker, R., Zhang, L.M. & Andersen, P. 2000. Modal
are not considered in such analysis. This assumption identification from ambient responses using Frequency
derives from the fact that some sensors in the cables Domain Decomposition, Proc. 18th Int. Modal Analysis
have proved not to work correctly due to their wrong Conference (IMAC), San Antonio 2000.
Gentile, C. & Siviero, E. 2007. Dynamic characteristics of
placement during the construction phases. the new curved cable-stayed bridge in Porto Marghera
Static and modal analysis were performed with a (Venice, Italy) from ambient vibration measurements;
tuned FE model in order to evaluate the performance of Proc. 25th Int. Modal Analysis Conference (IMAC),
the installed monitoring system, both from a static and Orlando 2007.
379
Iterative buckling analysis for steel cable-stayed bridges

Hanyang University, Seoul, Korea
D.H. Yoo
ENVICO Consultants Co., Ltd., Seoul, Korea
H. Yoo
Hyundai Institute of Construction Technology, Yongin, Korea
ABSTRACT
In general, cable-stayed bridges are composed of three

major structural members: cables, girders and towers.
The cable-stayed structural system of these bridges
has shown itself to be an economical, efficient, aes-
thetical appealing and durable solution for the length
of central span in the range of 200800 m (Gimsing,
1983; Poldony and Scalzi, 1976). But the behavior
of cable-stayed bridges is completely different with
other typical highway bridges because these bridges
are usually composed of three main structural mem-
bers: cables, girders and towers. Cables must resist
tension forces that are induced from dead loads and live
loads. Girders and towers must resist bending moments
as well as compression forces that occurs at the inter-
section between girders and towers. For this reason,
buckling behavior of main members is very impor- Figure 1. Example cable-stayed bridges.
tant in the design of cable-stayed bridges. In addition,
Cable-stayed bridges exhibit nonlinear behaviors con-
siderably due to cable sag, interaction between axial
forces and moments, and large displacements (Wang example bridges are computed using the proposed
et al., 2002; Shu and Wang, 2001; Xi and Kuang, method and compared with those found using sys-
2000). tem buckling analysis. Figure 1 shows the example
This paper illustrates a modification that elimi- cable-stayed bridges for 600 m, 900 m and 1,200 m
nates the problem inherent to system buckling anal- models.
ysis when used to assess the stability of girder and
tower members in cable-stayed bridges.The theoretical
REFERENCES
background of system buckling analysis is presented
briefly, and the root of its inherent problem is clarified Gimsing, N.J. 1983. Cable-Supported Bridges, Concept and
by describing a common cable-stayed bridge exam- Design. Wiley, Chichester.
ple. The axial force term in the geometric stiffness Poldony, W.J. and Scalzi, J.B. 1976. Construction and Design
matrix is modified by adding a fictitious axial force to of Cable-Stayed Bridge. Wiley, New York.
make the member more critical to the overall buck- Shu, H.S. and Wang, Y.C. 2001. Stability Analysis of
ling of the structure. The converged effective lengths Box-Girder Cable-Stayed Bridges. Journal of Bridge
of girder and tower members are determined based on Engineering-ASCE 6(1): 6368.
Wang, P.H., Lin, H.T. and Tang, T.Y. 2002. Study on Nonlin-
an iterative eigenvalue analysis with a continuously ear Analysis of a Highly Redundant Cable-Stayed Bridge.
modified geometric stiffness matrix. The proposed Computers and Structures 80: 165182.
iterative system buckling analysis is applied to exam- Xi,Y. and Kuang, J.S. 2000.An EnergyApproach for Geomet-
ple cable-stayed bridges that have center spans of rically Non-Linear Analysis of Cable-Stayed Bridges Pro-
600 m, 900 m and 1200 m, and different girder depths. ceedings of the Institution of Civil Engineers-Structures
The effective lengths of individual members in these and Buildings l(140): 227237.
380
Rehabilitation planning and stay cable replacement design for the Hale
Boggs Bridge in Luling, Louisiana
A.T. Ciolko & S.T. Wyatt

CTLGroup, Skokie, Illinois, USA
C.A. Ligozio
CTL Engineers & Construction Technology Consultants, PC, Rochester, New York, USA
ABSTRACT evaluated alternatives. The design of the complete

72-cable array replacement, completed in December
The Hale Boggs Bridge across the Mississippi River 2008, is the first occasion on which this process was
opened in October, 1983. It was the first cable- attempted in North America. This paper describes
stayed span on the Interstate Highway system. An efforts made to improve the cables and extend the
all weathering-steel structure, the bridge serves the service life of this historic river crossing.
public well. However, its stay cables corrosion protec-
tion sheathing performance made the stay cable array REFERENCES
vulnerable to both corrosion and service life deple-
tion. Repairs were attempted during and following Ciolko, A.T., and Yen, W.P. 1999. An Immediate Payoff from
construction, to correct sheathing defects. The repairs FHWAs NDE Initiative. Public Roads. Washington, DC.
performed poorly, and failed to protect the cablesmain Ligozio, C.A., and Mehrabi, A.B. 2005. Development and
tension element. Verification of a NDT Method for Detection of Flaws in
Condition assessments and a rehabilitation study PE Sheathing of Stay Cables. Proceedings of the 2005
concluded that 39 out of 72 cables were in critical need International Bridge Conference, Pittsburgh, PA.
Mehrabi, A.B., and Ligozio, C.A. 2007. Cable-Stayed
of repair. Timely action was necessary. To address the
Bridges Discovering Alarming Distress and Damages.
damage, and to assure the structural integrity of the Proceedings of the 2007 International Bridge Conference,
bridge structure, several strategies involving a range of Pittsburgh, PA.
repair and replacement options were evaluated using Post-Tensioning Institute. 2007. Recommendations for Stay
life cycle cost analysis. It was concluded that the strat- Cable Design. Testing and Installation, Fifth Edition,
egy to replace all cables presents the best value among Phoenix, AZ.
381
Reliability assessment of Yonghe Bridge based on structural

health monitoring
S. Li
School of Transportation Science and Engineering, Harbin Institute of Technology, China
H. Li
J. Ou
School of Civil and Hydraulic Engineering, Dalian University of Technology, Dalian, China
ABSTRACT load effects can be determined by the strain measure-

ments. For components without FBG strain sensors,
This paper presents the reliability estimation of the dead and unit temperature load and wind load
Yonghe Bridge based on structural health monitor- effects of the bridge can be evaluated by the finite
ing technology. The data collected by structural health element model, updated and calibrated by monitor-
monitoring system can be used to update the assump- ing data. By applying measured truck loads and axle
tions or probability models of random load effects, spacing data from weight in motion (WIM) sys-
which would give potential for accurate reliability tem to the calibrated finite element model, the live
estimation. The reliability analysis is based on the load effects of components without FBG sensors can
estimated distribution for dead, live, wind and tem- be generated. The stochastic process of Live Load
perature load effects. For the components with FBG effects can be described approximately by a Filtered
strain sensors, the dead, live and unit temperature Poisson Process and the extreme value distribution
of live load effects can be calculated by Filtered
Poisson Process theory. By combination temperature
and wind pressure distribution in the present design
code and unit temperature and wind load effects, the
extreme value distribution of temperature and wind
load effects can be achieved. Then first order reliability
method (FORM) is employed to estimate the reliability
index of main components of the bridge (i.e. stiffen-
ing girder). Furthermore, practical damage situations
are considered to evaluate the influence of certain
elements failure on the reliability of undamaged
components.
REFERENCES
Neves, A.C. & Frangopol, D.M. 2006. Probabilistic life-time
oriented multi-objective optimization of bridge mainte-
nance: combination of maintenance types. Journal of
Structural Engineering 132(11): 18211834.
Enright, M.P. & Frangopol, D.M. 1999. Condition prediction
of deteriorating concrete bridge using Bayesian updating.
Journal of Structural Engineering 125(10): 11181125.
Estes, A.C. & Frangopol, D.M. 1999. Repair optimization
of highway bridges using system reliability approach.
Figure 1. PDF comparison between measured and simu- Journal of Structural Engineering 125(7):766775.
lated vehicle load moment at closure segment.
382
Risk management of newly-built bridge vibration effects

on surrounding buildings
X. Shi, S. Geng, X. Ruan, Z. Wang & Z. Qu

Department of Bridge Engineering, Tongji University, China
ABSTRACT
A Concomitancy with economy development require-

ment and urban road traffic perfection, bridge con-
struction demand faces a new upsurge in China. Space
and distance between bridge structure and surrounding
buildings are becoming smaller and shorter. Newly-
built bridge, the real traffic situation of which is
unknown before its service, will probably have bad
vibration effects on surrounding buildings safety.
Evaluating reasonably newly-built bridges vibration
effect, which changes with vehicle flow, is a problem
for existing surrounding building and is also an impor-
tant measurement for risk management and reducing
overall risk losses. The similar or the same bridge type Figure 2. Similar bridge test point selected in Shanghai.
for similar for vehicle load is relatively easily found
while the similar or the same surrounding buildings are
difficult to determinate in China. The similar bridges
dynamic effect analysis and quantitative calculation Cast-in-place driven pile vibrations on construc-
are adopted in this paper. Based on the reality research tion, vehicle-bridge coupling vibration on operation,
topic, this paper will present the solution to evaluate are considered as the most important dynamic influ-
vibration effect. encing factors. Driven pile is treated as point vibra-
tion source moving down process. Hammer vibration
energy propagates from head to all directions. Com-
bining bridge location geological data with basic
construction technology of other similar bridge, a 4ton
diesel hammer with 2 m drop height is used to analyze
driven pile dynamic effect.
Accordingly, different risk management and sug-
gestions are put forward; a risk management hand-
book is formed to serve the bridge in the real topic
successfully.
REFERENCES
Xin Ruan & Airong Chen & Xuefei Shi, 2008. Risk assess-
ment of bridge engineering. Beijing: China Communica-
tions Press.
Guohao Li,1996. Stability and vibration of bridge structures
(Revised Edition). Beijing: China Railway Publishing
Figure 1. Surrounding buildings 3D location of real House.
research topic in Wenzhou.
383
Bridge safety assessment based on field test data with SORM method
X.M. Wang, X.F. Shi & X. Ruan

T.Y. Ying
Architectural Design & Research Institute of Tongji University (Group) Co., Ltd. Shanghai, China
ABSTRACT The comparison of Algorithms shows that:

(1) During the process of the existing bridges assess-
If the structural reliability index is close to the thresh-
ment, if the reliability index is close to the tar-
old value, the different algorithms may lead to different
get value, the more accurate calculation method
assessment conclusions due to the diversity of calcu-
SORM needs to adopt, and evaluation index is
lation results. By probabilistic methods ,the existing
essential to change from the reliability index to
structures safety assessment not only need to consider
the failure probability. It is particularly suitable for
the actual resistance, load effects and other field infor-
the mid-span cross-section subject to the greatest
mation, a reasonable choice of probabilistic algorithms
positive moments.
is also necessary.
(2) The difference of failure probability is almost
The paper reviews the FORM and SORM algo-
depended on the negative normalized gradient
rithm, their advantages, disadvantages and the cor-
component with respect to live load effect at the
responding origins are pointed out for the limit state
design point in the standard normal space. If the
equation used in bridges assessment.
normalized gradient component is higher than
According to the state inspection information and
0.2, the difference of failure probability exceeds
weighing vehicle data in motion, a large-span pre-
5%. In this case, a more accurate algorithm such
stressed continuous beam bridge is assessed using the
as SORM is suggested to calculate evaluation
probabilistic method. The design model is updated by
index.
the different kind of field information respectively.
FORM and SORM are adopted to calculate these
models. The reasons for the differences between two
methods and the variation law are studied.
REFERENCES
The safety assessment of the example bridge shows
that: Abraham Getachew. 2003, Traffic Load Effects on Bridges.
(1) The actual state of the existing bridge usually Doctoral Thesis, Structural Engineering Royal Institute of
Technology, Stockholm, Sweden,
deviates from the design; state inspection can Breitung, K. 1984. Asymptotic approximation for integrals.
effectively identify the true state to establish an Journal of Engineering Mechanics, ASCE, 110(3).
actual evaluation model; Chang, D.M. & Jiang, K.B. 1995. Probability analysis and
(2) With the increase of vehicles, the code parameters, design for bridge structural. Beijing: China railway pub-
based on the findings of 90s of the last century, lishing house. (in Chinese)
underestimate the vehicle load effect gradually. Yang D.X., Li G. & Cheng G.D. 2006. Convergence anal-
It is essential to establish the probability distri- ysis of first order reliability method using chaos theory,
bution and statistical parameters in accordance Computers and Structures 84: 563571.
with the current traffic, and forecasting methods Fiessler B. et al.1979. Quadratic limit states in reliability.
Journal of Engineering Mechanic, ASCE, 105(4).
to consider upgrading vehicle class; Guo, X.W. 1997. Determination of standard value for the
(3) Direct reliability methods have the higher suitabil- self-weight of highway bridge member. China Journal of
ity for the important long-span bridge, whose live Highway and Transport 18(1): 3541. (in Chinese)
loading characteristics may differ markedly from Haldar, A. & Mahadevan, S. 2000. Reliability Assessment
the descriptions contained in the code. Using Stochastic Finite Element Analysis, New York.
384
SS3: Bridge condition assessment
Organizers: A. Miyamoto & F. Tondolo
Use of vibration parameters for evaluating structural degradation

in R.C. elements and structures
R. Ceravolo, G. Mancini & F. Tondolo

Politecnico di Torino, Department of Structural and Geotechnical Engineering, Torino, Italy
ABSTRACT aim is to obtain experimental information that is useful

for more complex structures. Degradation due to cor-
Degradation due to corrosion of reinforcing bars is rosion is obtained through an electrochemical process
one of the principal deterioration phenomena affect- (Giordano et al. 2009), which simulates the damage
ing reinforced concrete structures (Bertolini et al. caused by about 27 years of exposure to environmental
2004). It is widespread all over the world; every year a conditions XC4 (class for carbonation induced cor-
great amount of funds is assigned to maintenance and rosion) with 25 days of test, during the propagation
repairing works for deteriorated structure and infras- period. The simultaneous effect of load action and
tructure. Therefore the analysis of structural behavior corrosion attack is established in order to take into
of corroded structures, such as bridges, is becoming account the interaction between stresses generated by
an urgent need for both risk evaluation and prior- tension load and the degradation mechanism. Crack
ity assignment of intervention. Corrosion of steel in opening due to both load ands corrosion are monitored
concrete deeply modifies the structural behavior of during the tests. Furthermore, electrical strain gauges,
reinforced concrete structures (Tondolo 2006). In fact, permanently positioned on the external surface of the
the oxide formation on the surface of the rebar in con- structural element, make it possible to measure ambi-
tact with concrete generates a significant volumetric ent vibrations of the specimen. Deformation signals
expansion of this new layer of granular material. The obtained from the vibration setup are then analyzed
radial pressure determines tensile stress orthogonal in order to extract modal quantities, including natural
to the rebar axis and generates radial cracks on the frequencies and modal damping, whose evolution in
surrounding concrete. Due to corrosion, a reduced per- time is associated to structural damage, such as fatigue
formance of concrete elements is observed especially crack growth and corrosion.
in service conditions. Bond mechanism is activated
by cracking of concrete; its reduction generally is
followed by an increase of crack opening. As ser- REFERENCES
viceability and durability of concrete structures can
Bertolini L, Elsener B, Pedeferri P, Polder R. Corrosion
be seriously affected by the aforementioned effects
of steel in concrete, prevention, diagnosis, repair. 2004,
(Bhargavaa et al. 2006), control and monitoring of Wiley, New York.
corrosion assume a significant importance in safety Bhargavaa K, Ghoshb AK, Moric Y, Ramanujama S, Model
reassessment, as well as in the evaluation of the over- for cover cracking due to rebar corrosion in RC structures.
all economic utility of an intervention (Ceravolo et al. 2006, Engineering Structures 28:10931109.
2009). Changes in the structures dynamic properties Ceravolo R, Pescatore M, De Stefano A, Symptom based
may often provide quantitative evidence that damaging reliability and generalized repairing cost in monitored
phenomena are underway. bridges. 2009, Reliability Engineering and System Safety
This paper presents the results of laboratory tests 94:13311339.
Giordano L, Mancini G, Tondolo F, Experimental evaluation
conducted on concrete ties subjected to static load and
of corrosion effect on bond between steel and concrete
simultaneous accelerated corrosion. The environmen- in presence of cyclic action. 2009, Key Engineering
tal attack due to corrosion of the rebar is simulated Materials 417418:345348.
in order to evaluate the structural condition of simple Tondolo F, Comportamento strutturale di elementi in calces-
concrete elements. The loading condition is derived truzzo armato soggetti a corrosione. 2006, PhD thesis,
from a bridge element under service load. The final Dep. of Struct. Eng., Politecnico di Torino.
387
Structural condition assessment of long-span suspension bridges using

long-term monitoring data
Y. Deng & Y.L. Ding

Key Laboratory of Concrete and Prestressed Concrete Structures of Ministry of Education, Southeast University,
Nanjing, Jiangsu Province, China
A.Q. Li
College of Civil Engineering, Southeast University, Nanjing, Jiangsu Province, China
ABSTRACT variability of the modal frequency and displace-

ment can be effectively quantified.
This paper focuses on developing an on-line structural (3) The mean value control chart is further employed
condition assessment technique using the long-term to monitor the time series of condition index e
monitoring data measured by structural health moni- for real-time condition assessment. The proposed
toring system. The seasonal correlation analyses of method can effectively detect the damage-induced
frequency-temperature and beam-end displacement- 0.2% variances of the modal frequencies and the
temperature for Runyang Suspension Bridge are firstly damage-induced 2.0% variances of the displace-
performed, respectively. Then the statistical technique ments of the Suspension Bridge
using a 6-order polynomial is further applied to for- (4) The results show that the proposed method can
mulate the correlations of frequency-temperature and effectively eliminate temperature complications
displacement-temperature, on the basis of which the from frequency and displacement time series
abnormal changes of measured frequencies and dis- and exhibit good capability for detecting the
placements are detected using the mean value control minor abnormal changes of measured modal fre-
chart. The analysis results reveal that: quencies and displacement, which is suitable for
online health monitoring for long-span suspen-
(1) The daily averaged values have better capacity to
sion bridges.
describe the correlation of frequency-temperature
and displacement-temperature than the short-
time values by eliminating the random variations
due to ambient excitations. The minimum vari-
REFERENCES
ance, maximum variance and mean variance of
daily averaged frequencies are 0.649%, 2.186% Ni,Y.Q., Hua, X.G., Fan, K.Q., et al., 2005. Correlating modal
and 1.403%, respectively. And the seasonal properties with temperature using long-term monitoring
changes of the northern and southern displace- data and support vector machine technique. Engineering
ments are 37.4 cm and 38.8 cm, respectively. Structures, 27(12): 17621773.
(2) A simple 6-order polynomial regression model Ding, Y.L., Li, A.Q., and Liu, T., 2008. Environmental Vari-
will be developed to be a quantitative description ability Study on the Measured Responses of Runyang
of the seasonal correlations in the present study. Cable-stayed Bridge Using Wavelet Packet Analysis. Sci-
ence in China Series E: Technological Sciences, 51(5):
The reproduction and prediction values of modal 517528.
frequency and displacement favorably agree with Ni, Y.Q., Hua, X.G., Wong, KY., et al., 2007. Assessment of
the measured values, which indicate the satis- Bridge Expansion Joints Using Long-Term Displacement
factory reproduction and prediction capability of and Temperature Measurement. Journal of Performance
the regression model. So the temperature-caused of Constructed Facilities, 21(2):143151.
388
Time-dependent reliability of PSC bridge box-girders exposed

to aggressive environments
T. Guo
College of Civil Engineering, Southeast University, Nanjing, P.R.China
R. Sause & D.M. Frangopol

ATLSS Center, Department of Civil and Environmental Engineering, Lehigh University, Bethlehem, USA
A.Q. Li
College of Civil Engineering, Southeast University, Nanjing, P.R.China
ABSTRACT
Bridge structures undergo time-varying changes in

response when exposed to aggressive environments.
While much has been done on reliability of time-
dependent corrosion, little is known to the influence
of concrete creep and shrinkage that can undermine
the reliability of bridge box-girders. In this study, an
MC90-based creep, and shrinkage model is depicted
by using the finite element (FE) codes in conjunc-
tion with time-dependent corrosion. More specifically,
time-dependent assessment is made for a composite
pre-stressed concrete (PSC) box-girder bridge under Figure 1. Composite degenerated shell element.
aggressive chloride attack. This assessment is real-
ized via an advanced stochastic finite element method
(SFEM). The thin-walled structure is described by
the composite degenerated shell element, see Fig.1,
in which the embedded reinforcement grids are used
to simulate the distributed reinforcements, and an
automatic generation scheme is adopted to facilitate
the finite element modeling of the curved prestress-
ing tendons. The post-cracking behavior of concrete
is depicted by the smeared cracking model with
the Hordijks tension softening relation. To improve
the efficiency of reliability analyses, an approximate
Importance Sampling arithmetic is proposed, in which
a number of Monte Carlo simulations are used to
get the proximate design points for the Importance
Sampling; thereafter, the Importance Sampling is
conducted with its center at the selected design points.
Reliability analyses show that under the combined
action of creep, shrinkage and corrosion, reliability
indices of the PSC box-girder undergo a three-phase
reduction, and fall below the target reliability level ear-
lier than the expected service life. In phase I (about 10
to 12 years after loading), a decrease in the reliability of Figure 2. Reliability index profile.
tendon yielding is observed, see Fig.2 (a). Meanwhile,
a faster increase in deflection occurs due to creep and phase III (about 50 to 60 years after loading), which is
shrinkage, which results in a significant decrease in the mainly controlled by corrosion.
reliability index, as shown in Fig.2 (b). In phase II, the The proposed methodologies in this paper can be
evolution of concrete creep and shrinkage becomes used for a better understanding of the deterioration
slower, and a mild decrease in the reliability index process of PSC bridges, and development of optimal
is observed. A significant decrease is found again in reliability-based maintenance strategies.
389
Condition improvement of deteriorating bridges by using high

performance ceramic materials
H. Nakajima, T. Yamagami & S. Tsukamoto

Hanshin Expressway Engineering Company Limited, Osaka, Japan
H. Sugiyama
Hanshin Expressway Company Limited, Osaka, Japan
T. Ura & H. Koide

Eagle Vision Company Limited, Tokyo, Japan
ABSTRACT real world conditions in the field.The experiments con-

sisted of tests on the material properties, tests on its
As time has elapsed since their completion, aging is ability as composite structures with steel (composite
advancing in the structures of the Hanshin Express- or hybrid structures) and tests on its workability in site
way. Damages to structural parts during this period of when applied to structures (steel structures).
time, increased due to heavy traffic loads. In particu- The experimental results, proved that ceramic mor-
lar, fatigue damages and corrosion in steel structures, tar can be applied as reinforcement materials for mastic
as well as damages in the pavements are problems that asphalt and undrained pavement, without problem
have to be urgently solved. Due to the public nature of from the adherence strength point of view.
roads, assurance of the road safety is required from the Based on these results, actual application in the
road manager and rational methods of management, as field, such as the ones introduced below, are being
well as maintenance are required. planned to reinforce structures of expressway viaducts.
As repair sites are located in severe environmen-
(1) Due to its high bonding and rapid hardening prop-
tal conditions (water, rust, salt), new methods and
erties, application to reinforcement of asphalt
materials that can be applied under these conditions
pothole.
are necessary. To cope with such severe environmen-
(2) Due to its high accuracy and anti-corrosion prop-
tal conditions, a new ceramic material was used for
erties, application to corrosion reinforcement of
this research as a new repairing and reinforcement
steel structures.
material.
(3) Due to its high strength, application to reinforce-
The following characteristics are generally attributed
ment of orthotropic plates, by making composite
to the new ceramic material:
or hybrid structures with steel members.
(1) rapid hardening
However, in applying ceramic materials to actual
(2) high strength
structures, the following topics have to be considered.
(3) no hardening shrinkage
(4) neutral (1) Verify corrosion resistance and weathering resis-
(5) excellent covalent bond tance and confirm durability.
(6) accuracy (2) Verify behavior under repetitive load and investi-
(7) high fire and heat resistance gate fatigue problems.
(8) high thermal insulation (3) Investigate a method to reduce the hardening ther-
(9) chemical resistance mal effects on the adherence strength in steel
(10) workability under low temperature, including structure.
below zero (4) Consider compatibility with construction quality
(11) harmlessness control by increasing the workable time during
(12) environment friendly and hardening
(13) abrasion resistance. (5) Establish standard construction conditions and
quality stability
Among these characteristics, excellent bonding
(6) Investigate the effects of introduction of large
adherence, high strength and early high strength, accu-
scale equipment in its construction.
racy are properties that makes the ceramic material
suitable as repair and reinforcement material. These topics will be eventually cleared, step-by-
The present paper reports the results of a series of step, in future studies.
experiments performed to verify its applicability to
390
Rehabilitation of Haynes Avenue Bridge
A.P. Ranasinghe & V. Ganji

Michael Baker Jr., Inc. USA
ABSTRACT The repair work over the Northeast Corridor had to

be done within a four hour window, between 12 AM
Haynes Avenue Bridge in Newark, New Jersey, USA and 5AM, which included de-energizing and energiz-
was built in 1931 and has a length of 1164 feet and a ing tracks. Shielding was provided when full depth
two-lane roadway width of 42.8 feet. It is a 9 span con- deck repairs were carried out. For the removal of con-
crete encased steel through-girder and floorbeam, and crete encasement or the injection of cracks, the tracks,
4 span reinforced concrete deck girder with floorbeam ties, ballast, and catenary wires were protected from
structure. The bridge crosses over the busy Amtrak and falling debris. A protective shield was prepared over
NJ Transit Electrified Northeast Corridor Lines, Con- the Port Authoritys Monorail Lines.
rail Rail Lines, Port Authoritys Monorail Lines, and a The existing bridge deck was overlaid with asphalt
Service Road. and a Ground Penetration Radar (GPR) system was
The condition of the bridge deck was considered to used to measure the depth of the overlay. GPR was
be poor and the superstructure fair. In August 2006, also used to determine the deterioration of the deck.
the NJDOT elected to rehabilitate the bridge without The existing asphalt layer was removed to carry out the
replacing the deck. The rehabilitation scheme includes deck repairs and was relayed with an asphalt layer. The
the removal of loose concrete from the concrete GPR results were compared with results from other
encased steel girders, repair of spalls and cracks on the conventional techniques such as the chain drag test
deck, parapets, piers, abutments and stairs, repair of and the actual deterioration as found once the entire
steel girders, replacement of the existing rocker bear- asphalt layer was removed.
ings with seismic isolation bearings, replacement of The paper discusses how the rehabilitation of this
deck joints and down spouts, repair and painting of bridge over one of the busiest transportation corridors
bridge railing, and the removal of debris and vegetation in the country was done and how GPR was used very
from the bridge. effectively in the deck rehabilitation work.
A seismic retrofit analysis was carried out and it
was found that the existing rocker bearings and anchor
bolts were inadequate for a seismic event. Therefore, REFERENCES
the existing rocker bearings were replaced with seis- FHWA Publication HRT-06-032, 2006. Seismic Retrofitting
mic isolation bearings. The isolation bearings consist Manual for Highway Structures: Part 1 Bridges.
of steel reinforced elastomeric layers with a lead core. Lichtenstein Consulting Engineers, Inc., 2006. Structural
They were specially designed to reduce the seismic Evaluation, Re-Evaluation Bridge Survey Report (Cycle
displacements. No. 9).
391
Influence of corrosion on prestressed concrete beams: An experimental

survey
Z. Rinaldi & S. Imperatore

Rome University Tor Vergata, Rome, Italy
C. Valente
University G. dAnnunzio of Chieti-Pescara, Italy
L. Pardi
Autostrade S.p.A., Rome, Italy
ABSTRACT Table 1. Test program and corrosion level.
Steel corrosion is one of the most common causes of Corrosion fc

deterioration of reinforced concrete and prestressed Series Beam [%] [MPa]
structures. The phenomenon is particularly felt in
1 7 0 34.0
the bridge structures, mainly if exposed to aggres-
1 8 20 34.0
sive environments and subjected to thaw salts. Unlike 1 9 20 34.0
the normal reinforced concrete, where the corrosion 2 1 20 41.5
damage is highlighted by the formation of cracks in 2 2 0 41.5
the concrete, and by rust spots, the failure for stress- 2 3 14 41.5
corrosion occurs suddenly, as the formation of cracks 3 4 0 47.4
in the metal does not require an intensive corrosive 3 5 20 47.4
attack, causing a strong material loss. 3 6 7 47.4
The study of these aspects and the evaluation of the
influence of the corrosion on the constitutive behaviour
of tendons and the consequent structural implications
are studied by means an experimental campaign car-
ried out at the Laboratory of the University of Rome
Tor Vergata. Three series composed by three rect-
angular prestressed beams each (Fig. 1) have been
subjected to four point bending tests. Besides the ref-
erence sound beams, the tendons of the elements have
been artificially corroded, in order to obtain three dif-
ferent corrosion levels, equal to about 7%, 15% and
20% mass loss (Tab. 1). The obtained results (Fig. 2)
highlight the influence of the corrosion on the global
strength of the element, in terms of ultimate force
reduction, and, mainly, on the variation of the failure
Figure 2. All series, load displacement curves at midspan.
mode of the corroded beam. Contrarily to the sound

beams, whose collapse is governed by the concrete
crisis, the failure of elements corroded with mild and
high corrosion levels (14%20%) is due to the strand
local rupture; the beam subjected to low corrosion
level shows a sharp ductility reduction and collapses
abruptly for the contemporaneous crisis of concrete
Figure 1. Beam geometry. and strand.
392
Development of a compound inspection method to detect fatigue damages

on orthotropic steel deck
H. Sugiyama & A. Tabata

H. Nakajima, T. Yamagami & S. Tsukamoto
M. Uchima
PASCO Corporation, Tokyo, Japan
M. Okuno
Nihon Densokuki Co. Ltd., Fukuoka, Japan
Y. Kohno
Japan Industrial Testing, Corp., Osaka Japan
Steel road bridges having orthotropic decks, due to Because they require contact with the steel
its lightness and short construction time, have been orthotropic deck, proper equipment and scaffolding
widely applied in the coastal area of urban highways. In works are necessary.
recent years, with the increase in the traffic amount and In high viaducts or water crossing, they require
the constant occurrence of overloaded vehicles, fatigue large-scale overall scaffolding works.
damages on orthotropic decks have been reported. Because of the above mentioned requirements, the
Fatigue damages on orthotropic decks can be clas- efficiency or the whole inspection is not good.
sified into 4 groups. Comparing the inspection costs, the costs of the
facilities are high, having budget restraints.
Cracks occurring in along the weld between the
Although the cycle of the inspections are not yet
deck and U-shaped rib
defined, periodic inspections are necessary.
Cracks occurring in the weld between the U-shaped
rib and transverse beams. Thus, the authors have developed a rational and
Cracks occurring in the butt weld between U- economic method, which combined infrared inspec-
shaped rib and transverse beams tion, eddy-current inspection and phased array
Cracks occurring in the weld between the deck plate ultrasonic inspection.
and the vertical stiffeners Through the study, it was found that a rational
In particular, type cracks originated in the root of and economic inspection can be achieved when the
the weld between the U-shaped rib and the deck can 3 inspection methods are applied orderly by starting
be classified into 3 types, as shown in the figure below. from a general and wide-range method, and deepening
and narrowing to a more directed method, that is,
Infrared inspection through thermal images
obtained by an infrared camera mounted on a
vehicle moving in high speed. (Screening)
A simplified method of evaluation applying eddy
current inspection to the points that were selected
during screening.
Detailed investigation through phased array at
Among crack types to , cracks of type -3 points in which cracks perforating the deck were
develop from the weld root in the direction of the found by the simplified evaluation method.
deck upper surface. These cracks not only cannot be
visually detected from the lower surface of the deck, In case shortening of inspection intervals becomes
but also, cannot be found by visual inspection from necessary, the present compound method, that made
the upper surface of the deck without the removal screening possible, can be applied without great
the pavement. Thus, ultrasound inspection carried out increase in the inspection costs.
from the bottom surface of the deck plate was applied. In order to improve the detection limits and preci-
However, ultrasound inspections have the following sion of the present method, the authors shall continue
disadvantages. to collect inspection data for further studies.
393
Application of Bayesian Network for concrete bridge deck

condition rating
A. Tarighat
Shahid Rajaee Teacher Training University, Tehran, Iran
A. Miyamoto
Yamaguchi University, Ube, Yamaguchi, Japan
ABSTRACT In this paper the main goal is to propose a Bayesian

Network which represents links of some cause and
Bridges as one of the important infrastructures are effect relationships between condition rating and
suffering from many defects during their service lives. symptoms or results found during concrete bridge deck
Condition rating (a judgment of a bridge compo- inspection. Therefore it will be also possible to pre-
nent condition in comparison to its original as-built dict the most probable condition rating of the deck in
condition) and damage detection are crucial methods future.
and tools to conduct efficient maintenance and man- It is believed that the obtained network is tolerable
agement of bridge structures. Condition rating index to noisy input data because the probability transition
is a sign of concrete state of the material or bridge of one state to another in output is smooth.
element at a given time.
Although the problems related to define and assign
precise condition rating to bridges are almost obvi- REFERENCES
ous and based on the visual inspection findings, visual
inspection is still the main action used to show bridge Chen W.F. & Duan L. 2000. Bridge Engineering Handbook.
CRC Press.
conditions. As a logical result of importance and prac-
Federal Highway Administration (FHWA). 1995. Recording
ticality of visual inspections some refinements for and coding guide for the structure inventory and appraisal
condition rating methods are required. of the nations bridges. Report FHWA-PD-96-001.
Inspection results show significant variability. This Friedman N. Geiger D. & Goldszmidt M. 1997. Bayesian
variability has direct influence on the assigned condi- network classifiers, Machine Learning. 29: 131163.
tion rating. Condition rating results analysis and data Friedman N. & Koller D. 2003. Being Bayesian About Net-
from some investigations revealed that the condition work Structure: A Bayesian Approach to Structure Discov-
ratings are normally distributed in many visual inspec- ery in Bayesian Networks. Machine Learning. 50: 95125.
tion reports. Since the bridge inspection results and Graybeal B.A., Rolander D.D., Phares B.M., Moore M.E. &
Washer G.A. 2007. Reliability and Accuracy of In-Depth
condition ratings show normal statistical distribution
Inspection of Highway Bridges. Journal of the Transporta-
probability theory can be considered for prediction of tion Research Board. 1749 / 2001: 9399.
condition rating of concrete decks. Hartle R.A., Ryan T.W., Mann J.E., Danovich L.J.,
Bayesian Networks provide a method to repre- Sosko W.B. & Bouscher J.W. 2002. Bridge Inspectors
sent relationships between propositions or variables, Reference Manual. Federal Highway Administration.
even if the relationships involve uncertainty, unpre- Jordan M.I. 1999. Learning in Graphical Models, Cam-
dictability or imprecision. They may be learned from bridge: MIT Press.
data, created by an expert, or developed by a com- Moore M., Phares B., Graybeal B., Rolander D. & Washer
bination of the two. They capture knowledge in a G. 2001. RELIABILITY OF VISUAL INSPECTION FOR
HIGHWAY BRIDGES: Volume I & II. Federal Highway
modular form that can be transported from one sit-
Administration.
uation to another. It is a form people can understand, Morcous G., Lounis Z. & Mirza M.S. 2003. Identification
and which allows a clear visualization of the relation- of environmental categories for Markovian deterioration
ships involved. Therefore Bayesian Networks can be models of bridge decks. J. Bridge Engrg. ASCE. 8(6):
used in modeling uncertainty. The central feature of the 353361.
Bayesian Network approach is that given a scenario, a Neapolitan R.E. 2003. Learning Bayesian Networks. Prentice
Bayesian Network depicts graphically the cause and Hall.
effect relationship between various elements of the Pearl J. 1988. Probabilistic Reasoning in Intelligent Systems.
scenario. San Francisco: Morgan Kaufmann.
394
Vulnerability analysis of a reinforced concrete structure

by visual inspection
M.V. Piras, L. Deias & F. Mistretta

Department of Structural Engineering, University of Cagliari, Italy
ABSTRACT Table 1. Degradation assessment card.
The article outlines a new, reliable and robust approach Type of Linguistic Safety
for estimating the vulnerability of reinforced concrete degradation judgment goal
sructures, developing a survey methodology based Ei Gij Wijk
only on visual inspection.
E1 Small G11 Medium W114
The results of visual inspection are subjective Shear stress Medium G12 Large W126
assessments of the structural safety yet easy to obtain Large G13 Very Large W137
employing unskilled operators. Aim of our work is
E2 Small G21 Slightly Small W213
to obtain an objective assessment of the safety level Longit. stress Medium G22 Medium W224
by manipulating the linguistic judgments using fuzzy Large G23 Large W236
logic (Zadeh 1965), chosen for its ability to quantita-
E3 Small G31 Very Small W311
tively treat the uncertainty of reality and inaccuracy Gravel nests Medium G32 Small W322
of natural language. The goal of our method is to Large G33 Medium W334
manipulate through fuzzy logic the subjective linguis-
tic judgments, expressed by an unskilled inspecting
staff, on the visual signs of degradation, in order to
assess the current safety level of the construction. Dur-
ing visual inspection, assessment cards will be filled in
by the inspecting staff, who will also be making photos
and videos of the construction.
In the card for each structural element we have three
columns. In the first column for each structural ele-
ment are shown the degradation expression that are
more relevant to the structural element itself. There
can be one or more expressions, depending on the
element under consideration. In the second column Figure 1. Weight Wijk membership functions.
the inspection staff will subjectively chose the assess-
ments of gravity attributed to the different types of
degradation. The third column shows the weights that environment. The results are within the values of fail-
each judgment will have on the current safety assess- ure probability that we expected (Zhao & Chen 2001).
ments. To each linguistic variable, and both for the Moreover the robustness of the method is proved when
judgments Gij and weights Wijk , we associate a mem- two different operators give slightly different linguis-
bership function (x) whose domain is [0, 1]. The tic judgments and derives directly from the ability of
triangular membership functions are shown in Fig. 1. fuzzy logic to deal with uncertainty.
It is important to stress out that the card is con-
structed and the relations between judgment and
weight on the safety previously stated according to REFERENCES
the expertise of a technically skilled team, while the Zadeh, L.A. 1965. Fuzzy sets. Information and control 8,
second column is filled in during visual inspection by 338353.
unskilled staff. The reliability of the method is shown Zhao, Z. & Chen, C. 2001. Concrete bridge deterioration
by analyzing an existing reinforced concrete structure, diagnosis using fuzzy inference system. Advances in
a bridge 3 km long, located in an aggressive seawater Engineering software 32, 317325.
395
Remaining lifetime and resistance of corrosion attacked bridge deck
J. Vican & M. Skora

University of Zilina, Zilina, Slovakia
ABSTRACT ACKNOWLEDGEMENT
Bridges are the most significant components of the The paper presents results of the research activi-
transport infrastructure from the viewpoint of safety, ties supported by the Slovak Grant Agency, grant
durability and serviceability and also from economic No. 1/0311/09 and the results of the project
point of view. Therefore, it is necessary to paid implementation: Centre of Excellence in Transport
enhanced attention to their design and exploitation Engineering (ITMS: 26220120027) supported by the
to fulfil required function during the bridge working Research & Development Operational Programme
life. Present standard methods of the reliability theory funded by the ERDF.
are based on verification of the structural reliability
from the viewpoint of limit states caused by the per-
manent, variable and accidental actions. Nevertheless REFERENCES
the actual member resistance can be changed by var-
ious factors as for example degradation of materials Akgul, F. & Frangopol, D. M. 2004. Lifetime performance
analysis of existing steel girder bridge superstructures.
due to aggressive environment, which are taking into
Journal of Structural Engineering v. 130, No. 12.
account indirectly or neither in bridge member design. Bucher, Ch. & Hintze, D. & Ross, D. 2000. Advanced anal-
The investigation of actual reliability and determining ysis of structural reliability using commercial FE-codes.
actual remaining lifetime of structure are the impor- ECCOMAS 2000, Barcelona, Spain, 11. 14. September.
tant characteristics for evaluation of existing bridge Kmiecik, M. & Guedes Soares, C. 2002. Response-surface
structures. approach to the probability distribution of the strength of
The purpose of this paper is to analyze uncertain- compressed plates. Marine Structures. Elsevier.
ties of orthotropic steel plate deck in order to evaluate Qin, S. & Cui, W. 2003. Effect of corrosion models on
its random variable resistance using the finite element the time-dependent reliability of steel plated elements.
Marine Structures. Elsevier.
simulations and to get the time variant stress-state
Rackwitz, R. 2001. Reliability analysis a review and some
influenced by corrosion effects. The effort was focused perspectives. Structural Safety. Elsevier.
on the determination of stress state of compression Reh, S. & Beley, D.J. & Mukherjee, S. & Eng Hui, K.
orthotropic decks creating the upper flanges of bridges 2006. Probabilistic finite element analysis using ANSYS.
with ballast beds. It is consisting of thin-walled sheets Structural Safety, Elsevier.
stiffened by the system of longitudinal and transver- Skora, M. & Vican, J. 2009: Numerical and experimental
sal stiffeners. All of the above mentioned factors are analysis of steel orthotropic plate deck stress state. XVIII.
stochastic in their nature, so the probabilistic approach Russian-Slovak-Polish seminar, Theoretical foundation
to the structural reliability assessment is necessary of civil engineering. Warszawa: Oficyna Wydawnicza
Politechniki Warszawskiej.
to use. There was developed the probabilistic model
ertler, H. & Vican, J. 1995. The real behaviour of steel
in ANSYS software environment using software sup- railway bridges. ilina: ES VDS (in Czech).
port module PDS. Random variables were described Tepl, B. & Novk, D. 1999. Reliability analysis of civil
by means of parameters of probability density distri- engineering structures. Brno: Academic press CERM.
bution according to data determined experimentally. Vican, J. & Kurck, F. 2006. Probabilistic approach to the
The model also comprised the geometrical and mate- ultimate strength of compression chords of railway bridges
rial nonlinearities of the structure. Nevertheless the considering corrosion. 6th International Symposium Steel
actual member resistance can be changed by vari- Bridges. Prague.
ous factors as for example degradation of materials
due to aggressive environment. The time variant loss
of member resistance was computed based on some
corrosion models.
396
The condition assessment and strengthening for the deteriorating

prestressed concrete trussed combination arch bridge
G.W. Yao
Key Laboratory of Bridge-Structure Engineering Ministry of Transport, School of Civil Engineering
and Architecture, Chongqing Jiaotong University, Chongqing, PR China
S. Bai
School of Civil Engineering and Architecture, Chongqing Jiaotong University, Chongqing, PR China
ABSTRACT should be strengthened with cross-section enlarge-

ment to enhance the structural stiffness. And the CFRP
The first prestressed concrete trussed combination arch sheets should be bonded to the bottom of the solid-
bridge was constructed in 1981 in Guizhou Province, web girder and the external prestressed cables should
China. Then the Jiangjiehe Bridge was constructed in be fixed at the midspan to arrest the down flexure and
cantilever assembling by derrick mast with main span increase the load-carrying capacity.
of 330 meters in 1995. This arch bridge is combined At last, some design and construction improvements
through the trussed arch bridge and the T-shaped truss were discussed to develop the prestressed concrete
rigid frame bridge with the advantage of convenient trussed combination arch bridge. this type bridge is
construction, low cost and short duration. But now half worth developing with the latest design method. Most
of them are in great damage and disease after over 20 the existing trussed combination arch bridges can be
years service. kept in service after retrofitting and strengthening.
In this paper, the condition assessment was per-
formed on the deteriorating trussed combination arch
bridges. Their main diseases include down flexure of REFERENCES
the solid-web girder at midspan, Part prestress lost,
concrete cracking in most truss joints between arch ribs Cao, W. 2009. Study on the typical diseases and effect on the
mechanical properties of the trussed combination arch
and precast plates of lower and upper chords (Fig. 1),
bridge. Chongqing: Chongqing Jiaotong University.
and part of precast plates fracture. Most of them are Chen, T.B. 2001. trussed combination arch bridge. Beijing:
assessed at grade B, C or D for the technical condi- Transport Press.
tion. The static and dynamic loading tests show that Huang, X.P. & Wei, Z.S. 2006. Analysis of damage of Yanjin
the carrying-capacity greatly decreases with obvious River Bridge and its strengthening. Highway 2: 1218.
impact effect under living loads. The low construct- Liu, L.S. & Yang, D.P. 2008. Damage mechanism investiga-
ing quality and the design defect endanger them. They tion of trussed combination arch bridge by FEM. Journal
need to be strengthened in structures or degraded in of Wuhan University of Technology 30(4): 110113.
load-carrying capacity. Liu,Y.H. & Huang, F.W. 2005. Estimation of bearing capacity
for old trussed combination arch bridges. Technology of
Then some retrofitting and strengthening methods
Highway and Transport 6: 7073.
were proposed. The cracks need to be repaired to Niu, H. & Jia, L. 2009. Comparative analysis of four kinds
enhance the material durability. The joints need to be of strengthening methods for trussed combination arch
strengthened with bonded steel plates or CFRP sheets bridges. Bridge Construction 1: 7880.
to enhance the structural integrity. The arch rings Pan, C.J. & Chen, T.B. 1989. Design of combined arch bridge
with truss pattern in Guizhou province. Journal of Civil
Engineering 22(4): 4956.
Song, L. & Zhou, Z.X. 2005. The effect to the dynamic
characteristics of long-span trussed combination arch
bridge by damage of the structure. Journal of Chongqing
Jiaotong University 24(4): 15.
Xu, C.C. & Zhong, N. 2009. Analysis of prestressed concrete
trussed combination arch bridge disease. Technology of
Highway and Transport 1: 7072.
Yan, Y. & Xu, H.Y. 2008. Research on the problems in
trussed combination arch bridge and its reinforcement
Figure 1. The truss joints crack greatly between precast scheme. Journal of East China Jiaotong University 25(3):
upper and lower chords, web members and vertical members. 1113.
397
Development of a bridge condition assessment system by using city bus
A. Yabe
KOZO KEIKAKU Engineering Inc. Tokyo, Japan
A. Miyamoto
Yamaguchi University Yamaguchi, Japan
ABSTRACT Main conclusions about foundational investigation

for reasonable method of detecting sever damaged
Because many infrastructures such as bridges will bridge maintenance can be drawn as follows.
entrance their last stage of designed durability in
(1) City bus rear axis acceleration response has liner
Japan, it becomes much more important to judge
relation with bridge acceleration response.
whether replace them or use them continuously over
(2) It is possible to use a bus acceleration monitoring
their designed durability while doing maintenance. An
system for assessment of the bridge condition.
old bridge should be inspected much exactly by expert
(3) Sensitivity analysis results by applied Substruc-
engineers rather than at its initial stage of lifetime.
ture Method were shown that city bus accel-
Because, a last stage bridge be exposed danger that the
eration is useful for evaluation of the bridge
bridge will be damaged severe by usual action. Now,
condition.
the structural health monitoring (SHM) gives us use-
(4) A foundational investigation for reasonable meth-
ful inspection data. And SHM have been researched
ods of detecting bridge with severe damage by
all over the world. Whether will be judged inspec-
bridge maintenance is carried out.
tion by expert engineers or SHM, a cost problem must
be solved. We have been researching the reasonable
method of a bridge inspection. In this paper, the con-
cept of a bridge condition assessment system using REFERENCES
city bus on a regular route.
JEAN-ARMAND CALGARO, ROGER LACROIX, Main-
In inspection by expert engineers or SHM, the cost
tenance et Reparation des ponts, 2002.6, Nippon Struc-
problems for bridge maintenance should be solved tural Inspection and Technology Association., pp. 4851.
first. The above results of experiment and sensitivity JSCE, Standard Specifications for concrete structures-2007,
analysis show possibility of using a city bus accel- Design, 2007, pp. 17181731.
eration monitoring for reasonable assessment of the JSCE, PROCEEDING OF COLLOQUIUM ON BRIDGE
bridge condition. Therefore, it can be considered to VIBRATION,1997.10.30. (in Japanese).
apply a newly monitoring system by using a city Subcommittee on Bridge Vibration Monitoring Committee
bus transportation system. A creative idea concern- of Structural Engineering JSCE, Guidelines for Bridge
ing using city bus transportation system to monitor Vibration Monitoring, 2000.9. (in Japanese).
Fujino Y. A study of train intelligent measurement system
bridges at a certain local area is proposed here. This
using acceleration of train. Sensors and Smart Struc-
system gives following merits. At first, only installing tures Technologies for Civil, Mechanical, and Aerospace
one acceleration sensor on a bus can monitor major Systems. Proceedings of SPIE, 2007, 6529(65291H).
bridges at a certain local area, so that the monitoring Ishii H, Fujino Y, Mizuno Y, Kaito K. A development of train
initial cost problem was reduced. Next, problem such intelligent monitoring system using acceleration of train.
as power supply to sensors will also be solved by using J. of Construction Engineering and Management, Japan
the bus battery. Another monitoring running cost prob- Society of Civil Engineers, Division F, 2008, 64(1): 4461.
lem is sensor maintenance, especially for necessary (in Japanese)
change of sensors sometimes, which could be solved Yabe, A, Araki, H, and Miyamoto, A. 2008. Application
of traffic vibration simulator to damage detection for
by doing maintenance at bus station. Therefore it is no
bridge maintenance, IABSE Conference on Information
necessary to go to a lot of places to install sensors on and Communication Technology for Bridges, Buildings
bridges and to maintenance or change sensors there if and Construction Practice, June 4-6, Helsinki.
using a bridge monitoring system with a bus. These Yabe, A, Yang, K, and Miyamoto, A. 2009. Sensitivity Anal-
merits solve almost all cost problems for sustainable ysis for Deteriorating Bridges using Bus Acceleration
bridge maintenance. Sensors, IABSE Symposium 2009, Sept9-11, Bangkok.
398
SS4: Bridges for high speed railways
Organizer: R. Calcada
Resonant effects on a bowstring railway bridge with orthotropic deck
C.M.C. Albuquerque, R.A.B. Calada & R.M. Delgado

Faculty of Engineering of the University of Porto, Porto, Portugal
ABSTRACT due to global and local modes of vibration of the deck

when crossed by the Thalys high-speed train.
The highly conditioned layout of high-speed railway The influence, on the vertical accelerations of the
tracks has led to an increased number of bridges and deck, of different parameters such as the train speed,
viaducts that use lower deck structures, due to min- the track modeling, the number and types of modes
imum clearance limits. This is the case of bowstring of vibration considered and the adopted modal damp-
bridges, where the deck, hanged by two lateral arches, ing ratios were studied and the obtained results were
often consists of an orthotropic grid of cross girders compared to the standard limitations to this parameter.
and longitudinal stiffeners joined together by a top The performed analysis enabled to draw the follow-
plate or slab. ing main conclusions:
The orthotropic decks are particularly subjected to
1) For bridges where local modes play an important
dynamic amplifications, due to their low mass, small
role on their dynamic behavior, the consideration
span of the cross girders and also due to the direct
of higher modal damping ratios for local modes,
charge by the train axles (Carnerero Ruiz, 2007). In
besides being a more reasonable assumption of
spite of that, standards do not yet cover these spe-
the real behavior of those bridges, can signifi-
cific problems. Until now, most of the studies on the
cantly reduce the estimated values of maximum
dynamic behavior of high-speed railway bridges have
accelerations;
focused on the analysis of the main structural elements,
2) Once again, where local modes play an important
and these studies lead to the most recent standards
role on dynamic behavior, the explicit modeling of
(EN 1991, 2003 and EN 1990, 2005).
the track leads to a better distribution of the loads
In this paper, the dynamic behavior of a bow-
reducing the dynamic response, especially at the
string bridge with an orthotropic deck, located in
each end of the structure;
the Brussels-Kln high-speed railway line, the Prester
3) The consideration of standard damping values
Bridge, is analyzed.
for all modes, and the non-explicit modeling of
Three-dimensional numerical models of the bridge
the track, would lead to excessively conservative
were developed in order to evaluate the resonant effects
results for vertical acceleration.
REFERENCES
Albuquerque, C. 2008. Dynamic behavior of high speed
railway bridges with orthotropic deck; MSc. Thesis (in
Portuguese). Porto: Faculty of Engineering of the Univer-
sity of Porto.
Carnerero Ruiz, A. 2007. Dynamic behavior of high speed
railway bridges with transverse structural elements; PhD.
Thesis (in Spanish). Madrid: Escuela Tcnica Superior de
Ingenieros de Caminos, Caneles y Puertos. Universidad
Politcnica de Madrid.
Comit Europen de Normalisation 2003. EN1991-2
Eurocode 1: Actions on structures Part 2: Traffic loads
on bridges. Brussels.
Comit Europen de Normalisation 2005. EN1990-A2: Basis
of Structural Design Annex A2: Applications for bridges.
Figure 1. Prester Bridge. Brussels.
401
Long-term monitoring of Sesia high speed railway bridge
G. Chellini, F.V. Lippi, L. Nardini & W. Salvatore

Department of Civil Engineering, University of Pisa, Pisa, Italy
ABSTRACT imposed by tracks to vehicles required the realization

of lines characterized by lower values of vertical gra-
Structural monitoring has been widely adopted in dient and curvature radius than classical railway lines.
the last years on small, medium and long-span rail- Many bridge and viaduct solutions were realized in
way bridges for the real-time assessment of structural these new lines by the adoption of classical concrete
reliability and safety, the control of serviceability con- solutions together with new steel-concrete typologies
ditions, the determination of actual load/stress spectra (Chellini, 2009). Maintenance and serviceability of
on main structural elements due to traffic loads. This bridges and viaducts constitute a crucial aspect in the
structural health control approach allows to issue early management of high speed railway lines, due to the
warnings on potentially catastrophic structural dam- economic losses connected with unplanned interrup-
ages, and also to plan in advantage maintenance and tion of traffic. Monitoring systems were installed on
repair operations with consequently economic benefits most bridges and viaducts, to control continuously
(Ko, 2005). the structural health conditions and the serviceability
Among many different techniques, the vibration performance of structures.
measurement approach is one of the most widely In this paper, short-term dynamic analysis results
adopted method for the dynamic evaluation of civil and long-term monitoring system preliminary data
structures. This method is based on the inverse eval- obtained for a new steel-concrete double-box com-
uation of structure dynamical modal properties (i.e. posite high speed railway bridge are illustrated. Pre-
frequencies, shapes and damping ratios), generally by liminary data obtained from the long-term monitoring
the elaboration of structural acceleration data. (Fujino, system have been processed in order to evaluate the
2002). Modal parameters can then be adopted as ref- ability of the system in the estimation of the modal
erence for a monitoring system, as structural damages dynamic parameters and actual railway traffic spectra
or deteriorations significantly change their values. It on the bridge.
is also possible to develop numerical procedures for
the real-time identification of damage entity and posi-
tion along the structure (Fryba, 2001). Monitored
modal parameters can be adopted in the updating REFERENCES
of FE models, obtaining reliable real-time evaluation
Ko, J.M., and Ni, Y.Q. 2005. Technology developments
of forces/stresses in structural elements and, conse-
in structural health monitoring of large-scale bridges.
quently, of global/local safety of the bridge. Strain Engineering Structures, 27, p. 17151725.
measurements have been also widely adopted in the Fujino, Y. 2002. Vibration, control and monitoring of long-
monitoring of civil structures for the real-time con- span bridgesrecent research, developments and practice
trol of strain/stress load spectra in structural elements, in Japan. Journal of Constructional Steel Research, 58, p.
in order to evaluate resistance safety and fatigue life 7197.
behavior of local details (Li, 2002). Strain sensors Fryba, L. and Pirner, M. 2001. Load tests and modal analysis
embedded into structural concrete structural elements of bridges. Engineering Structures, 23, p. 102109.
were adopted to evaluate directly loads, frequencies Karoumi, R. et al. 2005. Monitoring traffic loads and dynamic
effects using an instrumented railway bridge. Engineering
and velocity of vehicles on railway bridges (Karoumi,
Structures, 27, p. 18131819.
2005). Chellini, G. and Nardini, L. and Salvatore, W. 2009.
In the last few years, the construction of new high Dynamical identification and modelling of steel
speed (HS) railway lines across Europe was accom- concrete composite high-speed railway bridges. Struc-
plished. The high value of design speed and the strict ture and Infrastructure Engineering, 2009, DOI: 10.1080/
parameters for vertical and transverse acceleration 15732470903017240.
402
Fatigue assessment of Sesia high speed railway viaduct
G. Chellini, F.V. Lippi & W. Salvatore

Department of Civil Engineering, University of Pisa, Pisa, Italy
ABSTRACT characterized by new typologies of local details,

designed to improve the local and global fatigue
Fatigue and fracture are very important failure modes behaviour avoiding those critical factors that caused
for steel structures, causing about 8090% of total fatigue failures into previous solutions. To this aim,
damage. In particular railway bridges, enduring mil- experimental fatigue tests represent a reliable and suit-
lion of stress cycles during their life, are expected able tool to confirm design assumptions and to evaluate
to be highly vulnerable to fatigue phenomena. The the real improvement in the fatigue response of new
fatigue assessment of new and existing railway bridges steel and steel-concrete composite solutions (Hoorpah
is one of the main issues in current engineering prac- et al. 2004).
tice because of many concurrent events, as the rapid In this paper, fatigue assessments of a new steel-
development of the European railway networks (Euro- concrete double-box composite high speed railway
pean Commission, Energy and Transport DG, 2005) bridge are reported. A suitable procedure, composed
the increase of passenger and freight traffic, the intro- of experimental and numerical analysis, was adopted
duction of new materials, the adoption of new steel and to obtain a reliable evaluation of structure fatigue
steel-concrete composite solutions and local details. behaviour. Long-term monitoring data were ana-
Present management approach of the European lyzed to evaluate the real railway traffic spectra over
Railway Network expects the possible presence on the bridge. Consistent load stress spectra on local
the same line of many train families, with several details were estimated adopting a suitable dynamic
layouts of weight and axel distributions, introducing train-bridge interaction procedure coupled with the
new uncertainties about the most convenient traffic sub-structuring approach. Experimental fatigue tests
load spectra for fatigue assessments (Goicolea et al, permitted to estimate the S-N fatigue life curve of crit-
2004). These new operating conditions require the ical details and constituted the basis for the damage
evaluation of adequacy and effectiveness of fatigue assessments of selected structural joints, adopting the
design rules actually provided by national and Euro- Rain-Flow counting technique and the linear damage
pean codes, as these recommendations result from accumulation law of Palmgren-Miner.
previous studies on old railway lines and could lead
to an underestimation of actual railway traffic effects.
The general increase of operating speeds exposes
REFERENCES
bridge superstructures to resonance phenomena,
especially when train speed goes over 200 km/h European Commission Energy and Transport DG, 2005.
(EN1991-2, 2002). Steel and steel-concrete compos- TEN-T Trans-European Transport Network: priority axes
ite superstructures can experience significant dynamic and projects 2005. Technical report, European Commis-
interaction effects, more than other structural solu- sion. (ec.europa.eu).
tions, due to their light and slender structures, which Goicolea, J., 2004. Dynamic loads in new engineering codes
mass is comparable to the mass of the train. As res- for railway bridges in Europe and Spain. In Bridges for
onance phenomena may influence significantly the High-Speed Railways, Porto, Portugal, June 34.
EN1991-2, 2002. Eurocode 1: Actions on structures
structural safety and the bridge service conditions,
Part 2: Traffic loads on bridges. CEN European Committee
reliable tools for the proper evaluation of structure for Standardization.
dynamic performance are needed. Hoorpah, W., Montens S. and Ramondenc P., 2004. Steel and
Another innovative aspect in the construction of composite bridges for high speed railways The French
railway bridges is the more and more frequent adop- know-how, In Bridges for High-Speed Railways, Porto,
tion of steel and composite steel-concrete solutions Portugal, June 34.
403
New considerations on track-structure interaction in railway bridges
P. Gonzlez Requejo, M. Cuadrado Sanguino & D. Losa Baur

undacin Caminos de Hierro, Madrid, Spain
ABSTRACT The paper will describe the conclusions of the main

research works of the authors as well as the practi-
Due to the longitudinal interaction between track and cal experience gathered. Several issues not addressed
structure, the presence of railway bridges may generate in current standards shall be described, such as: con-
added stress on continuous rails and relative displace- siderations on deformations due to shrinkage and
ments between track and deck or substructure. Thus creep in concrete pre-stressed decks; combination of
an analysis of these phenomena is needed from the actions with a comparison of simplified combination
structure design phase on for ensuring proper system methods versus step by step computation methods;
behavior. special features of interaction with slab track; track-
UIC Leaflet 774-3 and Eurocode EN 1991-2 state structure interaction under seismic conditions and
methodologies for the analysis of track-deck interac- finally the application of tailored methodologies to
tion, as well as actions to be considered and limit values studies on special geometries such as skew bridges
both for stress in rail and displacements. or underpasses.
Although these phenomena may be analyzed the- Finally other relevant aspects shall be presented,
oretically, these reference documents approach inter- such as the relationship between structural types and
action computations by means of numerical models their influence on interaction analysis (eventual expan-
that idealize the behavior of all elements and actions sion devices), the issues associated to the use of
involved to obtain stresses and displacements. deck movement control devices (blockers or dampers)
The authors have more than ten years experience and the alternatives, in several limit cases, to the
in these numerical models and in the application of definitive installation of expansion devices (special
these methodologies to the analysis of viaducts, some fastenings, stress relief and provisional expansion
of which are singular. devices).
404
Comfort evaluation of railway bridge vibration using bridge-train

transfer function
B.G. Jeon & N.S. Kim

S.I. Kim
Korea Railroad Research Institute, Uiwang, South Korea
ABSTRACT
Recent trend of structure is that serviceability is more

emphasized than stability. While re-searches on vibra-
tion serviceability of high speed railway bridge are
already conducted and are being applied to designing
of railway bridges in Europe and Japan. But the vibra-
tion comfort limit of bridge structure by heightening
of speed is not clearly defined in Korea.
Therefore, improvement of quality of high speed
railway bridge vibration serviceability is required to
improve domestic railway transportation competitive-
ness and to ad-vance to overseas market.
In general, methods of calculating car vibration
through bridge train dynamic interaction analysis can Figure 1. Proposed maximum displacements compared
be classified into three. The first is to execute mov-ing with Korea Railway Bridge Specifications(High-speed rail-
constant force analysis, and calculate car vibra-tion way, 2004).
with the use of vibration transfer function. The second
is to use car model expressed with SDOF(single degree
of free-dom) system. And the last me-thod is to execute deflection. Therefore, the vertical vibration acceler-
dynamic interaction analysis by using 3 dimensional ation of car itself that passes through bridge was used
moving car model system. In the case of 3 dimen- as survey indicator. And by modeling 1 passenger
sional system, various effects like inertial force and car of KHST(Korea High-Speed Train) with moving
car rotation etc may be consi-dered. To evaluate the constant force and SDOF system, 3 dimensional sys-
vibration serviceability of car that runs along bridge, tem, dynamic interaction analysis was done, and the
accurate method may be modeling both car and bridge responses were compared. From the results, the vali-
and executing dynamic interaction analysis. However, dity of transfer function was verified, and with the use
dynamic interaction analysis that uses such delicate of it, comfort deflection limit of railway bridge was
model uses various elements for accurate modeling of developed.
actual bridge and car. Therefore, it is difficult to iden-
tify which element affects dynamic behavior. Also, the
ground to specify the acceleration signal inside car REFERENCES
obtained with the use of precisely implemented car
model is insufficient. Therefore, inferring the vibra- Jeon, B.G. and Kim, N.S., 2007, Evaluation of Comfort Limit
tion within car with the use of moving constant force on Bridge Vibration, transaction of the ko-rean society for
analysis and car transfer function is estimated simple noise and vibration engineering, 17(10): 923935.
Bruggers, W.B., 2002, Comfort Related Design of Railway
and appropriate method.
Bridge, Delft; Delft University of Technology.
In this paper, vibration serviceability is defined as Kim, S.I., 2000, Bridge-Train Interaction Analysis of High-
the performance to obtain the comfort of railway us- Speed Railway Bridges, Seoul; Seoul University.
er for car vibration occurring in the accompaniment Jeon, B.G., Kim, N.S. and Kim, S.I., 2009, Vibration Ser-
of structure deflection. And in designing newly in- viceability Evaluation of Railway Bridges Con-sidering
stalled structure, it was estimated as the performance Bridge-train Transfer function, spring conference of the
to prevent excessive vibration to car by structure korean society for railway, pp. 357366.
405
Bridges E-2 and E-3 in the new railway to the Northwest of Spain
C. Jurado
Polytechnic University of Madrid, Spain; Ingecal Ingenieros, S. L. Madrid, Spain
ABSTRACT
During the years 2001 to 2007 it has been constructed

in Spain the new railway line of high speed (L. H. S.)
that connects Madrid with Valladolid, with a length
of 179.6 kilometres. The line Madrid - Valladolid has
been put in service to the public on December 23rd
2007, and the time estimated to cover the distance is
50 to 55 minutes, which makes an average speed of
215 kilometres per hour, but the railway platform is
prepared to reach 350 kilometres per hour.
The cost of the works has ascended to 264.8 million
Euros.
The proximities to Madrid of this new L.H.S., it
implies the interference with essential services for the
capital such as: roads, railways, big pipes of water
supply, etc., and the new necessary structures must Figure 2. F.E.M. of bridge E-2 with 1203 nodes, 1098 frame
be constructed without interrupting the corresponding elements and 345 shell elements.
services.
The first bridge near to Madrid which was necessary
to constructed was the structure called in the project
E-2.
The project of the bridge E-2 was realized by means
of a three-dimensional model of finite elements with
the program SAP2000N that includes all the elements
of the structure.
The solution consisted in the demolition of the cur-
rent structure of arch and the construction of a new
bridge over the existing railway, fulfilling the require-
ments of vertical clearance of 7.00 m and with a span
of 22.64 m needed.
Figure 3. The new bridge E-2 and to the bridge E-3A/3B in

construction.
The bridge E-3 was divided in two Bridges called E-

3 and E-3B. The bridge E-3A has 66 m of length with
a span of 22,70 m. and a vertical clearance of 7 m.
Figure 1. Plan from Madrid to Valladolid.
406
New bridge in pergola for the new railway of high speed trains to the
Northwest of Spain
C. Jurado
Polytechnic University of Madrid, Spain; Ingecal Ingenieros, S. L. Madrid, Spain
ABSTRACT
During the years 2008 to 2010 it will be constructed in

Spain the new railway line of high velocity trains, that
will connect Valladolid with the northwest of Spain.
This part of the Spanish Infrastructure Plan will cost
200 millions of Euros with a length about 183 kilo-
meters. The speed will be about 220 kilometres per
hour, but the railway platform is prepared to reach 350
kilometres per hour.
The connection between Palencia and Leon com-
prises 122.20 km and in the middle of it there is the
section of Ro Cea-Bercianos del Real Camino with a
length of 10.5 kilometers.
The present paper exposes the project and construc-
tion of one of the most singular viaduct in the section
Figure 2. Three Dimensional Finite Element Model of the
of Ro Cea Bercianos del Real Camino, from Palen- Pergola from the NE.
cia to Len, which crosses over the existing motorway
M-31. The solution of the bridge has been made as a
viaduct in pergola with a length of 153 m and with
a width of 14.5 m.
The project of the viaduct was made by the author of
this paper, with a three-dimensional model of finite ele-
ments using the program SAP2000N, that it includes
all the parts of the structure, such as: foundations
made with piles (4,500 m of piles), mat foundations
tying the piles, pillars, and the superstructure com-
prising a prestressed concrete slab, and the complete
two abutments.
Figure 3. Aerial view of the A-231 motorway and the

viaduct in Pergola in construction.
This model permits to obtain all the responses of the

Pergola such as moments, shears, axial forces, etc. in
all the elements. In order to construct the prestressed
upper slab it is necessary to cut one of the senses (two
lanes) of the highway and to deviate the traffic tem-
porarily for the other sense (two lanes), in order to
scaffold one of the parts of the viaduct without any
risk for the traffic, and to prestress the slab of the deck.
The viaduct is now in construction.
Figure 1. New railway line Madrid-Valladolid-Len.
407
Fatigue damage assessment of steel bridge members using paint cracking
Y. Okumura & M. Sakano

Kansai University, 3-3-35, Yamate, Suita, Osaka, Japan
Y. Horie & H. Kobayasi

Hanshin Expressway Company, 4-1-3, Kyutarou, Cyuou-ku, Osaka, Japan
ABSTRACT
Since steel bridge members are usually covered with

paint coatings, fatigue damage and cracking are devel-
oped underneath them. The paint coatings cover up
cracks. But fatigue cracks of steel members may be
able to be detected using paint coatings. The inspec-
tions of bridge are usually conducted by visual inspec-
tion. Therefore, if some information whether cracks
occur underneath paint coatings were seen from a part
of the paint crack, it could be useful. The purpose of
this study is to detect damage of steel members using
cracks of paint coatings. In the past study, monotony
and alternating load experiments of specimens with
paint coating cut out from removed highway bridges
were conducted. And it was evaluated about damages
of steel members and paint coatings in the plastic area. Figure 2. Crack Length in Steel Members and in Paint
In this study, high cycle fatigue test is conducted using Coating.
fillet welded joint specimens with paint coating cut
out from removed highway bridges after Hyougoken-
Nanbu Earthquake, in order to evaluate the damage of (2) The paint crack develops after the steel crack hap-
steel members using damage of paint coatings as well pens and grows up after that by almost equal
as fatigue strength of actual bridge members. length.
The principal results obtained through this study are (3) The paint crack generation life is almost equal to
as follows; the breaking life of the welded joint fitting in a low
stress long-lived area.
(1) The fatigue strength of vertical stiffener attached
portion satisfies Class D recommended by JSSC.
REFERENCES
Sakano, M., Horie, Y., Kobayashi, H.,Kawachi, S. and Mis-
umi, Y., Basic Study of Fatigue Damage Assessment of
Steel Members Utilizing Damage of Paint Coating, Jour-
nal of Constructional Steel, Vol. 4, pp. 407414, (1996)
(in Japanese).
Sakano, M.,Yonemoto, E., Kano, K., and Mikami, I., Fatigue
Strength of Welded Vertical Stiffner Joints in Steel High-
way Bridges, Proc. Of JSCE, No.612/I-46, pp. 3343,
(1989) (in Japanese).
Sakano, M., Mikami, I., Yonemoto, E., and Nishimura, T.,
VariableAmplitude Loading FatigueTests on WeldedVer-
tical Stiffener Joints in Steel Railway Bridges, Journal of
Structural Engineering, Vol.41A, pp. 845854, (1995) (in
Figure 1. Form and Dimensions of Specimen. Japanese).
408
Movable Scaffolding Systems for 72 m span in one casting stage
A.A. Pvoas
AP Bridge Construction Systems Portugal, Ltd
ABSTRACT
The evolution of bridge construction equipments con-

tinues and we are now able to produce movable
scaffolding systems able to cast in situ spans of 72 m
or even larger, with girder weights up to 300 KN/m, in
one stage only.
In this article we will present one of these equip-
ments and a 70 m span bridge made with it, the Engano Figure 1. MSS 148 m long with 63 m between supports.
Viaduct, n\ear Santiago de Compostela, Spain.
Some high speed train bridges constructed with cantilever of the previous span could hold the rear
these new equipments will also be referred. reaction 9 m ahead of the rear pier.
After building some heavy bridges for the Spanish But with an interacting design of both the bridge
High-Speed Train Line with 55 m span between piers, and the MSS, we concluded that we could use 1/4th of
but with 54 m distance between supports for the MSS the span joints, with the MSS rear support 16 m ahead
we went forward for another challenge the mythic of the rear pier, i.e. with 54 m between supports.
frontier of the 70 meters concrete spans. This solution being globally cheaper than the pre-
The MSS capacity is measured by 3 factors: vious one, also allowed us to conclude that by using
1/4th of the span construction joints and bringing the
The admissible live load
rear supports to the edge of the cantilever, the existing
The distance between concreting supports of the
machines could also perform 80 m spans, with 63 m
MSS concreting span
between the MSS supports, by placing the rear support
The distance between launching supports of the
17 m ahead of the rear pier, over a 20 m cantilever.
MSS flying span.
With this project and some others with very heavy
In previous jobs done in 2006-2009 we have used formworks we have learnt how to open smoothly and
our most advanced models prepared to hold 300 KN/m quickly those formworks, launch the whole machine
girders plus the necessary external and internal form- and close the formwork in very short periods, and we
work self weight, for concreting and considering its know now how to get to 90 m spans and larger with
self weight and the external formwork self weight Moving Scaffolding Systems, casting the full span in
during launching. one stage.
Our models used in Vila Pouca de Aguiar Viaduct This construction method already being the cheap-
(Portugal) and Toxa and Martixe Viaducts (HST) in est for Viaducts with multiple spans, allows new
Spain were capable of supporting 280 KN/m slabs plus challenges for the bridge designers and also for the
60 KN/m of the formworks, with the supports of the construction engineers providing them with tools that
MSS 54 m apart, and flying over 63 m. can replace the form travelers, cutting down the
For Viaducto do Engano, the MSS was previ- construction costs and delays.
ously designed 148 m long, for 63 m or less between We are ready for those new challenges and willing
concreting supports, with 1/5th span cantilever joints. to assist and cooperate with bridge design engineers to
This design configuration of the MSS allows it to allow them to start using this system in their projects
reach 72 m span between concrete piers if the concrete and new sites.
409
SS5: Industrial smart material applications
for civil infrastructure (ISMA)
Organizer: T.B. Messervey
Streicker Bridge: Initial evaluation of life-cycle cost benefits of various

structural health monitoring approaches
B. Gliic & S. Adriaenssens

Princeton University, Princeton, NJ, USA
ABSTRACT approximately US$ 185,000 with some 150 man-

hours for the installation. At the time of writing the
The Streicker Bridge is a newly built pedestrian bridge paper, the cost of material used to build the bridge
at Princeton University campus. The bridge represents is approximately US$ 3,100,000 (the bridge is nearly
the entrance gate to the campus, monument of struc- completed), thus the estimated costs of the equipment
tural art and is an important landmark. Thus, beyond its for monitoring based on global structural monitoring
functional purpose the bridge also has strong symbolic approach and integrity monitoring approach represent
and aesthetic impacts. The bridge is equipped with var- approximately 4.4% and 6.0% of the cost of construc-
ious monitoring systems with an overall aim to trans- tion material. According to National Safety Council a
form this structure into a multipurpose research and comprehensive cost of each pedestrian or cyclist fatal-
teaching laboratory. An important long-term research ity is estimated to US$ 3,840,000 (includes medical
topic studied in this bridge-laboratory is the evaluation expenses, victim work loss, public services and lost
of life-cycle cost benefits of various structural health quality of life), while injury costs on average $52,000.
monitoring approaches. Four scenarios will be stud- Daily Road User Costs (DRUC) for Washington Road
ied over time: global structural monitoring, integrity are estimated (Herbsman et al. 1995) to be US$ 4660
monitoring, local material monitoring, and various per each day the road is closed due to malfunction of
combinations of these three scenarios. At present, Streicker Bridge. The late-repair costs at the present
since the bridge is new and the project is in its initial stage could not be evaluated, but they will be studied
phase, only preliminary cost estimations are available in the course of the project.
and presented in this paper. The savings related to pedestrian/cyclist fatality or
Currently implemented monitoring systems are injury, DRUC and late-repair are considered as quan-
based on global structural monitoring approach and tifiable benefits of monitoring. However, the loss of
integrity monitoring approach (Glisic & Inaudi 2007). human life is invaluable, as well as the loss of the
Global structural monitoring approach is based on the bridge considered as a monument of structural art and
use of discrete long-gage average strain sensors, and heritage for future generations, and their pre-servation
it features high resolution and accuracy, but its spa- represents unquantifiable benefit.
tial resolution depends on the density of the sensors The initial evaluation of costs presented in this paper
which is frequently kept low in order to contain the will serve as an input for life-cycle cost benefits of
costs. Integrity monitoring is based on the use of dis- SHM based on global structural and integrity monitor-
tributed deformation sensors with an excellent spatial ing approaches using deformation sensors. The other
resolution, but the strain resolution and accuracy are approaches and other types of the sensors will also be
about order of magnitude lower. included in the research.
The sensor networks for both systems were
designed with respect to research and education aims
of the project. However, they helped evaluate the cost REFERENCES
parameters if similar approaches were used for the
Glisic, B. & Inaudi, D. 2007. Fiber Optic Methods for
bridge safety and maintenance purposes. The cost of Structural Health Monitoring. Chichester: Wiley.
equipment for global structural approach will amount Herbsman Z., Chen W. & Epstein W. 1995. Time is money:
to approximately US$ 136,000 and installation will Innovative Contracting Methods in Highwal Construction.
consume approximately 200 man-hours. The equip- Journal of Construction Engineering and Management,
ment for integrity monitoring will be more expensive, ASCE, pp. 273281.
413
Optimal design of bridge SHM systems based on risk

and opportunity analysis
D. Inaudi
SMARTEC SA, Manno, Switzerland
ABSTRACT Table 1. 7-step SHM Design methodology.
Structural Health Monitoring (SHM) aims to provide Step Properties

more reliable and up-to-date information on the real
conditions of a structure, observe its evolution and 1 Identify structures needing monitoring
2 Risk/Uncertainty/Opportunity analysis
detect the appearance of new degradations. The bene-
3 Identify responses
fits of having a Structural Health Monitor-ing system 4 Design SHM system and select appropriate sensors
installed on a bridge or any significant structure are 5 Installation and Calibration
many and depend on the specific application. Here are 6 Data Acquisition and Management
the more common ones: 7 Data Assessment and analysis
Monitoring reduces uncertainty
Monitoring discovers hidden structural re-serves
Monitoring discovers deficiencies in time and
increases safety combine different measurement technologies. Once
Monitoring insures long-term quality the sensors are installed and verified, data collection
Monitoring allows structural management can start. If these logical steps are followed and the
Monitoring increases knowledge monitoring data is correctly acquired and managed,
When designing a Structural Health Monitoring sys- data analysis and interpretation will be greatly simpli-
tem, one should always focus on the specific require- fied. This process guarantees that each sensor placed
ments of the structure under exam. To achieve an in the structure serves at least one specific purpose and
optimal design it is however beneficial to follow a leads to a lean and cost-effective system.
well-defined and proven procedure. This paper presents a generalized 7-step method-
The first step in the design process consists in iden- ology for designing on optimized Bridge SHM mon-
tifying the risks and opportunities associated with the itoring system and a practical example from a field
bridge under examination. Examples of risks include application: the new I35W Bridge in Minneapolis.
the probable degradation mechanism due to ageing
(e.g. corrosion or fatigue) or external actions (e.g. seis-
REFERENCES
mic, impact or overload). Examples of Opportunities
include the existence of reserve capacities due to better Del Grosso, A. and Inaudi, D. (2004) European perspective
material properties, synergic effects and over-design. on monitoring-based maintenance, IABMAS 2004, Inter-
Risks and opportunities are present in both new and national Association for Bridge Maintenance and Safety,
existing structures. The purpose of an SHM system is Kyoto, Japan (on confer-ence CD).
to identify and quantify them, so that the consequences Glisic, B. and Inaudi, D. (2003) Components of structural
of risks can be avoided (e.g. a collapse) and the benefits monitoring process and selection of monitoring system,
of opportunities can be exploited (e.g. safely extending PT 6th International Symposium on Field Measure-
ments in GeoMechanics (FMGM 2003), Oslo, Norway,
the lifetime of a bridge).
755761.
Next, the expected responses to the expected degra- Glisic, B. and Inaudi, D. (2007), Fibre Optic Methods for
dations and the effects of the possible opportunities Structural Health Monitoring, Wiley.
are established and an appropriate Structural Health Inaudi D. et al (2009), Structural Health Monitor-ing Sys-
Monitoring Systems is designed to detect such con- tem for the new I-35W St Anthony Falls Bridge, 4th
ditions. Only at this stage, the appropriate sensors International Conference on Structural Health Monitoring
are selected. When selecting the best sensors for the on Intelligent Infrastructure (SHMII-4) 22-24 July 2009,
specific task, it is often necessary and beneficial to Zurich, Switzerland.
414
Smart textiles and their application in bridge engineering
T.B. Messervey & D. Zangani

Innovation & Research Division, DAppolonia S.p.A., Genova, Italy
J.L. Withiam
DAppolonia, Monroeville, PA, USA
ABSTRACT
The materials of tomorrow will be multifunctional

serving multiple roles and crossing multiple disci-
plines. Sensor-embedded load bearing elements will
be able to transmit data, report in-service load condi-
tions, and enable the assessment of health as well as
perform their traditional role.Already used extensively
in construction and in composites, textiles make an
excellent candidate for a smart material. Their manu-
facturing process is well suited to the direct integration
of fiber optic sensors through weaving and warp-
Figure 1. Textile-matrix-stone composite (courtesy of the
knitting. As such, large-area sensor-embedded textiles Karlsruhe Institute of Technology and Selcom Multiaxial
can be manufactured offsite, are mass producible, Technologies).
and are inherently cost effective. For the engineer,
they can provide both reinforcement and monitor- superstructures are replaced. In such cases, their load
ing capabilities across an array of applications that history and performance over time can be critical in
span both new and the retrofit of existing engineered deciding whether or not they can be reused or must be
structures. In doing so, in-service data can be utilized reconstructed.
to validate design assumptions, to control construc-
tion operations, to assist with life-cycle management
actions, and as a mechanism to provide alert during REFERENCES
extreme events. This paper highlights recent progress
in the development and testing of sensor-embedded Frangopol, D.M. & Messervey, T.B. 2009. Life-cycle cost and
textiles within the European research project Poly- performance prediction: Role of structural health monitor-
functional Technical Textiles Against Natural Hazard ing. Chapter 16 in Frontier Technologies for Infrastructure
(www.polytect.net) with special emphasis on the con- Engineering, S-S, Chen and A.H-S.Ang, eds., Structures
and Infrastructures Book Series, Vol. 4, D. M. Frangopol,
sideration of smart textiles in bridge engineering.
Book Series Editor, CRC Press/Balkema, Boca Raton,
With respect to bridges superstructures, smart com- London, New York, Leiden.
posites and smart textiles open the possibility to Gladstone, R., Anderson, P., Fishman, K.L., & Withiam, J.L.,
reinforce existing bridges economically, to construct 2006. Durability of Galvanized Soil Reinforcement: More
new bridges with high performance light weight mate- Than 30Years of Experience with Mechanically Stabilized
rials, and to manage maintenance intensive bridge Earth. Transportation Research Board No. 1975: 4959.
superstructures more optimally. With respect to bridge Liehr, S., Lenke, P., Wendt, M., Krebber, K., Seeger, M.,
substructures, smart textiles provide a tool to address Thiele, E., Gebreselassie, B., & Munich, C., 2009. Polymer
challenges associated with weak soils, to increase Optical Fiber Sensors for Distributed Strain Measurement
and Application in Structural Health Monitoring. Sensors
safety and to reduce risk at worksites and to revisit
Journal, IEEE, 9(11), pp. 13301338.
the underlying assumptions present in existing codes Wu, J., Lee, K., Helwany, S., & Ketchart, K., 2006. Design
and guidelines to ensure consistent levels of reliability and Construction Guidelines for Geosynthetic-Reinforced
across the foundation, abutments, and superstructure. Soil Bridge Abutments with a Flexible Facing. NCHRP
Also particular to the substructure and foundation ele- Report 556, Transportation Research Board, Washington,
ments is the possibility of their reuse when bridge D.C.
415
Conical couplers for non-pigtailed, free-space optical coupling to fiber optic

sensors for bridge monitoring
L. Qiu & K. Goossen

University of Delaware, Department of Electrical and Computer Engineering, Newark, DE, USA
D. Heider
University of Delaware, Center for Composite Materials, Newark, DE, USA
D.J. OBrien & E.D. Wetzel

Army Research Laboratory, Aberdeen, MD, USA
ABSTRACT
Traditional structural monitoring utilizes electro-

mechanical sensor system. However, in recent years,
innovative fiber optical sensors (FOSs) have attracted
intense interest. Specifically, fiber Bragg grating
(FBG) sensors have shown great utility for safety man-
agement of bridges. They can be either embedded
into bridges or simply mounted onto the surface to
measure various parameters. One major drawback of
the embedding application, however, can be the need
for extending a fiber pigtail off the structure for
mechanical coupling with the optical source and detec-
tor. Therefore, we have developed techniques of free
space optical coupling to FBG sensors, including the
machining of 45-degree mirror in the fiber. In this tech-
nique, we have developed spliced multimode coupler/
single mode FBG sensor fibers and embedded them
into composite panels, and used free space optical cou-
pling to interrogate the sensor and measure mechanical
strain. However, it is impractical to machine the mir-
ror after the fiber is embedded into the bridges. It is
also difficult to maintain the orientation of the 45-
degree mirror if it is polished before being embedded. Figure 1. Free space coupling to FBG: (a) with 45-degree
Therefore, in this paper, we have developed rotation- mirror; (b) with conical coupler.
ally symmetric concave conical optical couplers, that
couple light independent of the rotational state of the
fiber with respect to the outside of the structure and the conical coupler. Therefore, we conclude that this
the optical source. We theoretically studied the con- approach could potentially solve the problem of fiber
ical couplers on both plastic optical fiber (POF) and pigtailing in structure monitoring with FOSs.
multimode optical fiber (MMF), using computer sim-
ulations. In our preliminary experimental attempt, we
were able to machine the conical coupler into a POF REFERENCES
with a core diameter of about 3 mm. We also have Qiu, L., Teitelbaum, M. E., Goossen, K. W., Heider, D.,
been able to couple free space light into FBG with OBrien, D. J., & Wetzel, E. D. 2009. Normal free space
the conical coupler on MMF. We believe that with fur- coupling to fiber Bragg grating sensors. Paper presented
ther modification to the current system, we will be at the SPIE Smart Structures/NDE 2009, San Diego, CA,
able to also couple the light out of the FBG through USA.
416
SS6: Advances in structural robustness:
dependability framework
Organizer: F. Bontempi
Dependability of complex bridge structural systems
F. Bontempi & S. Arangio

School of Engineering, University of Rome La Sapienza, Rome, Italy
ABSTRACT
Modern structural design in Civil Engineering has

changed and has to change furthermore to cope with
the actual challenging superstructures and large infras-
tructures which are now of interest for the Society.
Invariably, Society asks for more demanding safety
and performance levels, while ever-increasing interest
is focused on durability of constructions (life cycle)
and accidental resistant capacity, widening then both
the time horizon and the event possibility.
All these requirements are, in some sense, in con-
trast with simplified or even over simplified design
formulation and assessment, unfortunately sometimes
too much emphasized, by traditional standards and
codes. These approaches are in fact devoted to gen-
erally simple and habitual design situations and can-
not be extended beyond this field of application, as
sometimes regrettably proposed.
This paper will summarize the basic ideas that
form, in the Authors experience, a modern approach
to structural design. One can consider:
Figure 1. Main aspects of dependability.
1. the emphasis on the systemic nature of structures,
that leads the concept of structural systems with its Starossek, U., Wolff, M. Design of collapse-resistant struc-
indispensable hierarchical organization; tures, JCSS and IABSE Workshop on robustness of
2. the importance on the ordered formulation of the struc-tures, Building Research Establishment, Garston,
Waterford, UK, November 2005.
design problem, taking methodically into account
Bontempi, F. (2006). Basis of Design and Expected Per-
both the qualitative and the quantitative aspects of formances for the Messina Strait Bridge. Proceedings of
the problem, leading to the so-called Performance- BRIDGE 2006, Hong Kong [CD-ROM].
based Design; Bontempi, F., Gkoumas K., Arangio S. (2008). Systemic
3. the necessity to have a clear vision of the complex- approach for the maintenance of complex structural sys-
ity of the design problem; tems, Structure and Infrastructure Engineering, 4, 7794,
4. the need to formulate in a very general way the DOI: 10.1080/15732470601155235.
requirements for structural quality introducing the Aviienis, I., Laprie, J. C., Randell, B. (2004), Dependability
composite concept of dependability. and its threats: a taxonomy, 18th IFIP World Com-
puter Congress, Toulouse (France), Kluwer Academic
In the paper, some reference is made to the proposed Publishers.
design for a long-span suspension bridge. Handling Exceptions in Structural Engineering 2008, Rome,
Finally, it is believed that these ideas can lead not DOI:10.3267/HE2008
only to a proper development of future constructions Frangopol D.M., Curley, J.P. (1987). Effects of Damage and
Redundancy on Structural Reliability, ASCE National
but also (and perhaps more important) to a way to
Convention, Atlantic City.
educate engineer students to develop a wider point of
view of the design problem.
REFERENCES
NASA, 2007. NASA Systems Engineering Handbook, Revi-
sion 1, NASA/SP-2007-6105.
419
Multilevel structural analysis for robustness assessment

of a steel truss bridge
F. Brando
Structural Engineer, New York, United States
R.B. Testa
Columbia University in the City of New York, New York, United States
F. Bontempi
Sapienza University of Rome, Rome, Italy
ABSTRACT The analyses have been performed using a 3D finite

element model of the case study bridge. After bridge
In recent decades, robustness and collapse-resistance subjected to seasonal and daily temperature changes as
have become significant aspects to consider in the well as construction loads, one built up section mem-
design of structures. While engineers always aim to ber is analyzed for local failure in compression. For the
avoid them, failures do occur and may indicate a lack local model, the numerical results of a buckling anal-
of robustness in a structural system. It becomes impor- ysis obtained by detailed FEM analysis are compared
tant to understand both the approach to the analysis of with results obtained from simple formulas.
potential local failures and to determine their effect on The multilevel analysis is illustrated for the truss
the overall structure structure of the I-35W Bridge for one particular assess-
The statically determinate trusses in actual bridges ment: the response of one bottom chord segment of
have some redundancy mainly due to continuity at con- the main truss. Such a procedure can be applied to
nections. In the current open debate about robustness investigate not only potential sources of the collapse of
and collapse-resistant design the effect of such redun- that bridge but also in assessing robustness of similar
dancy becomes important. In the present paper it is structures.
taken into account by a multilevel analysis going from
the global to the local level. This approach is illus-
trated by application to the truss portion of the recently REFERENCES
collapsed I-35W Minneapolis Bridge which has redun-
dant reactions but is a functionally non redundant National Transportation Safety Board. 2008. Collapse of
I-35W Highway Bridge, Minneapolis, Minnesota. High-
structure.
way Accident Report NTSB/HAR-08/03.Washington,
The concepts of redundancy, progressive collapse DC, November 14.
and robustness are closely intertwined. In order to URS Corporation, 2006. Fatigue Evaluation and Redun-
assess robustness in a complex structure it is neces- dancy Analysis, Bridge No. 9340 I-35W over Mississippi
sary to understand not only the global interaction of River, Draft Report URS Corporation, Minneapolis, MN,
the structure with the surroundings but also the behav- July.
ior at various local levels. Approaching the problem Wiss, Janney, Elstner Associates Inc., 2008. I-35 Bridge
using a multilevel perspective, the global system has over the Mississippi River: Collapse Investigation, Final
been broken down in order to perform a detailed analy- Report, November, WJE No. 2007.3702.
sis and evaluate the system performance at global and
local levels.
420
A framework for robustness assessment in the context of corroded

RC structures
E.S. Cavaco & L.A.C. Neves

Universidade Nova de Lisboa, Civil Engineering Department, Monte de Caparica, Portugal
J.R. Casas
Technical University of Catalunya, Civil Engineering Department, Barcelona Spain
A.H. Huespe
International Center for Computational Methods in Engineering (CIMEC), Santa Fe, Argentina
ABSTRACT
Structural robustness seems to be an emergent concept

related to the structural response to damage. At the
present time, robustness is not well defined and much
controversy still remains around this subject. Even if
interest on robustness has seen growing as a conse-
quence of catastrophic consequences due to extreme
events, the fact is that the concept can also be very
useful when used on more probable exposure scenarios
such as deterioration. This paper intends to be a contri-
bution to the definition of structural robustness, espe-
cially in the analysis of reinforced concrete structures Figure 2. Normalized peak load carrying capacity as a
subjected to corrosion. To achieve this, a deterministic function of the corrosion level Xp .
measure for robustness is proposed given by the area proposed by Oliver et al. (2008). Reinforcement was
below the curve defined by the normalized structural modeled using a slipping fiber model upgraded with
performance f subjected to a normalized damage d. the M-pull model proposed by Bhargava et al. (2007)
To illustrate the proposed concept, an example of a in order to predict bond strenght deterioration. For
corroded reinforced concrete foot bridge was analyzed corrosion levels from 0% to 100% load carrying capa-
using nonlinear analysis numerical methods based on city was predicted using the referred methodology (see
the continuum strong discontinuities approach and Figure 2).
isotropic damage models for concrete. The method- Finally structural robustness of the corroded struc-
ology consisted on a two step analysis proposed by ture was assessed by computing the area below the
Snchez et al. (2008). On the first step a cross section normalized load carrying capacity in Figure 2.
analysis was performed in order to capture deterio-
ration and cracking of concrete due to expansion of
corrosion products (see Figure 1). REFERENCES
On the second step, a 2D longitudinal model of Bhargava, K., Ghosh,A., Mori,Y., and Ramanujam, S. (2007).
the corroded structure was built based on the results Corrosion-induced bond strength degradation in rein-
obtained from the cross section analysis. Concrete forced concrete analytical and empirical models. Nuclear
was modeled using a composite material approach Engineering and Design, 237(11): 11401157.
Oliver, J. ,Linero, D., Huespe, A., and Manzoli, O. (2008).
Two-dimensional modeling of material failure in rein-
forced concrete by means of a continuum strong disconti-
nuity approach. Computer Methods in Applied Mechanics
and Engineering, 197(5): 332348.
Snchez, P., Huespe, A., Oliver, J., and Toro, S. (2008).
Numerical modelling of the load carrying capacity degra-
dation in concrete beams due to reinforcement corrosion.
In 8th World Congress on Computational Mechanics
(WCCM8), 5th European Congress on Computational
Figure 1. Isodisplacement lines showing cracks on Methods in Applied Sciences and Engineering (ECCO-
concrete. MAS 2008).
421
Buckling of steel gusset plates
C. Crosti
University of Rome La Sapienza, Rome, Italy
D. Duthinh
National Institute of Standards and Technology, Gaithersburg, MD, USA
ABSTRACT edge loading, the unfactored Load Factor Rating

(LFR) or Load Resistance Factor Rating (LRFR)
As a result of the structural failure of the I-35W bridge strength is acceptable up to an initial deformation of
in Minneapolis in 2007, FHWA issued guidelines for 1t = 13.3 mm = 310 mm/23.3 = L/23.3, although the
the load rating of gusset plates and recommended plate would have to undergo an additional out-of-
that the capacity of these plates on non- load-path- plane deformation of the same order before this
redundant steel truss bridges be verified. The intent of strength is reached. Initial imperfections in excess
the guidelines is to be safe and simple, i.e., amenable of 1t would be unsafe. For the more realistic Case
to hand calculations. According to the guidelines, the C and edge loading, the LFR or LRFR value is
compressive strength of a gusset plate is obtained from quite conservative, even for an initial imperfection of
column strength curves in the inelastic range, which 2t = 26.6 mm = 310 mm/11.65.
assume an initial out-of-straightness of L/1500. The The finite-element model was not refined enough to
purpose of this paper is to examine the buckling behav- determine the load carried by individual bolts, so vari-
ior of steel gusset plates in greater detail, accounting ous load distributions were assumed. The assumption
for parameters that were not explicitly included in the that the compression is carried by the first row of bolts
guidelines, such as initial deformations of the gusset (closest to the plate edge) produces results that are very
plate, stiffness of the framing members, load distribu- similar to the previous case of edge loading, except for
tion from the framing members to the plate and load the almost perfect plates, which show a small strength
eccentricity. To this end, a finite-element model of a increase. As initial deformations increase, the effect
gusset plate was developed and verified against exper- of a small change in column length decreases. A tri-
imental measurements. Results show that the FHWA angular (or linear) load distribution was also studied,
guidelines for load rating are conservative and safe with the first row of bolts (near the edge of the plate)
for larger initial out-of-plane deformations, up to one carrying the highest load and the last row of bolts (near
plate thickness. This is true for in-plane compressive the center of the plate) carrying no load. Here the plate
loads with no moment and no eccentricity. buckling strength is considerably greater than for edge
Experimental results were obtained from tests of loading, and the LFR or LRFR value is adequate even
gusset plates performed at the University of Alberta. for an initial deformation of 2t.
The test parameters included thickness and size of the The LFR or LRFR value cannot be achieved even
gusset plate, as well as angle, moments and restraint for an almost perfect plate loaded by a wide flange (I)
conditions of the framing members. Finite-element section framing onto one side of the plate. The load is
results show, as expected, that the stiffness of the applied with an eccentricity equal to half of the depth
framing member plays an important role in the buck- of the I-section, at a location corresponding to the first
ling strength of the gusset plate. The experimental row of bolts. Initial deformations of 1t and 2t produce
value falls between Case B, where the flange thick- negligible effects compared to the load eccentricity.
ness of the framing member is added to that of the When load eccentricity and initial imperfections act
gusset plate, and Case C, where both the flange and in opposite directions, load-induced deformations can
the web of the framing member add to the stiffness of reverse direction and allow the plate to achieve much
the gusset plate. greater strength than for a nearly straight plate.
As initial imperfections increase in magni-
tude, the plate strength decreases. For case B and
422
Cable-loss analyses and collapse behavior of cable-stayed bridges
M. Wolff
Grassl Engineering Consultans, Hamburg, Germany
U. Starossek
Structural Analysis and Steel Structures Institute, Hamburg University of Technology, Hamburg, Germany
ABSTRACT
The general aim in designing structures, where the

consequences of a collapse are high, must be col-
lapse resistance. This means that no structural damage
should develop that is disproportionate to the trigger-
ing event. Generally, structures can be made collapse
resistant by ensuring a high level of safety against local
failure or by designing for the failure of elements and
thus increasing the robustness. Increasing the robust-
ness of cable-stayed bridges is achieved by means of
designing for the loss of cables.
This paper investigates the loss of any one cable
by nonlinear dynamic analysis of a three-dimensional
model of a cable-stayed bridge. Dynamic amplifica-
tion factors for quasi-static analyses are determined. Figure 1. Instability failure of bridge deck after the loss of
The aim of these analyses is to give advice on how to three short cables.
determine this dynamic amplification factor, to prove
if the use of a uniform amplification factor to calcu- Additionally to having appropriate analysis tools
late the maximum responses to cable loss is valid and for creating robust structures, it is important to know
if reductions of this factor are generally possible. which structural properties are important to increase
The results show that a unique dynamic amplifica- structural robustness. These properties are identified
tion factor cannot be specified. Instead, the dynamic by investigating the collapse behavior of a cable-stayed
amplification factor depends on the location of the bridge after the loss of cables. The results show that
ruptured cable and on the type and location of the the normal forces in the bridge girder is the collapse
state variable being examined. Using a factor smaller promoting attribute of a cable-stayed bridge. For this
than 2.0 is only possible for the bending moments reason, self-anchored cable-stayed react less robust to
in the bridge girder. Here, an explicit calculation of the loss of short cables where the normal force in
the DAF can be beneficial. A dynamic amplification the bridge deck is highest. For the investigated cable-
factor of 2.0 is necessary for the safe design of the stayed bridge, two adjacent short cables in one cable
cables. Regarding the bending moments in the pylons, plane plus one cable in the second cable plane can fail
large dynamic amplification factors result because of without disproportionate collapse. In case of the fail-
large dynamic responses resulting from a complex ure of three adjacent cables, the bridge collapses due to
excitation. Dynamic time-history analyses are there- instability failure of the bridge deck (Figure 1). Other
fore recommended, at least for the cable loss cases collapse scenarios are presented, too.
which yield the highest responses. These are gener- The robustness of the bridge can be increased by
ally the longer cables but not necessarily the longest preventing instability. This is possible by increasing
cable. the stiffness of the bridge girder or by reducing the
The dynamic amplification factor can be deter- unsupported length, which means closer cable spac-
mined in linear dynamic analyses. Different live load ing. The former is recommended here due to a higher
positions do not have to be considered. failure probability of closely spaced cables.
423
SS8: Nondeterministic schemes for structural
safety & reliability of bridges
Organizer: S. Arangio
Degradation history simulation a tool for assessment of structural lifetime
M.A. Ahrens, P. Mark & Ch. Fust

Institute of concrete structures, Ruhr-University Bochum, Germany
ABSTRACT
For several years the number of reinforced concrete

bridges, which show a critical state of preservation
due to degradation processes, is significantly increas-
ing. In these cases structural experts are challenged to
decide whether or not a structure can definitely resist
an ongoing unrestricted usage, might be upgraded by
repairs, or must be sumptuously replaced by a new one.
Today, the use of the finite element method is well
proven in application of deterministic structural anal-
ysis. However, reinforced concrete exhibits a strong
nonlinear behavior and thus demands an application
of a complex nonlinear material model.
Thereby, reinforced concrete contains many sources
of uncertainties essentially affecting its mechanical
properties. For nonlinear finite element simulations of Figure 1. 3D time-deflection curves of 50 nonlinear simu-
structural lifetime in general, especially in the case of lations with respect to uncertain corrosion rate and geometry
prediction of (residual) lifetime, these scattering prop- as well as fitted distribution of deflections each 10 years.
erties should be considered carefully. Beneath material
properties, relevant damage driving forces introduce A finite element model of the bridge structure
even more uncertainty to the model. Although well is simulated, until failure due to time-dependent
founded methods of nonlinear and probabilistic struc- degradation mechanisms occurs. Sensitivity of life-
tural mechanics are separately at disposal, their com- time, with respect to relevant uncertain parameters,
bination still causes great difficulties with respect to is studied (Fig. 1). The distribution of simulated life-
efficiency. In this context, Latin Hypercube Sampling time response is established by hypotheses testing.
(LHS) enables to decrease notably the number of Concerning structural lifetime, uncertain but irrele-
simulations required, still ensuring an adequate rep- vant parameters are identified. Their exclusion from
resentation of the underlying probability distributions stochastic data base enables to reduce the variability
of relevant input parameters. Accordingly, the spa- of response. Further, the framework of RSM enables
tial scatter is mapped onto the finite element mesh by to establish a regression model, which detaches expen-
random fields. sive nonlinear simulation from prediction. Confidence
The present work transfers the above mentioned intervals of lifetime prediction capability of the pro-
theoretical framework to a complex engineering struc- posed model as well as tests of model adequacy are
ture a reinforced and pre-stressed concrete arch performed.
bridge located in Germany to demonstrate its appli-
cability in practice. The bridge, built in 1951 two years
before the first standard on pre-stressed structures has REFERENCES
been published in Germany, has been deconstructed Petryna, Y.S., Breitenbcher, R & Ahrens, M.A. 2009. Appli-
in 2006 owing to an expertise, which identified loss of cation of Lifetime-Oriented Analysis and Design: Arched
pre-stress due to fatigue corrosion of tendons respon- Reinforced Concrete Bridge. In F. Stangenberg, et al.
sible for a critically reduced bearing capacity (Petryna (eds.), Lifetime-Oriented Structural Design Concepts:
et al. 2009). 616634. Heidelberg: Springer
427
Non deterministic approaches in the current structural codes for assessing

the safety and reliability of bridges
S. Arangio
M. Ghosn
City College of New York, NY, USA
ABSTRACT In the first part of the paper, the advantages and

drawbacks related to both the measures are discussed
The main goal of bridge agencies is to ensure the safety and it is shown that, under conditions of uncertainty,
of the traveling public and the health of the economic bridge management systems should be reliability
cycle by maintaining a safe bridge network system and based. In fact, it is through the probabilistic approach
extending its useful life at minimal life cycle cost. To that the uncertainties can be reflected properly in
achieve this goal, bridge agencies select an appropriate bridge management and strategies that integrate cap-
bridge management approach that may include bridge ital and maintenance activities at the lowest expected
instrumentation for health monitoring purposes. The life-cycle cost can be determined.
implementation of such monitoring systems requires The second part of the paper gives an overview of
the availability of damage assessment models to eval- existing codes and guidelines in Europe and the USA.
uate the degree of structural degradation that would In the USA, a unified condition assessment procedure
trigger a warning mechanism when a bridge reaches has existed for a long time. In Europe, there are cur-
a critical state. Health monitoring systems include rently different condition assessment methods in each
instrumentations for the non destructive evaluation of country. Harmonization of these methods is needed in
materials, load testing, corrosion detection, and ana- the near future. Bridge authorities must adopt proven
lytical models for using the collected information for methods to assess the safety and reliability of existing
load rating. To this end, appropriate criteria are needed bridges in order to make appropriate provisions for
to identify threshold values that separate the damage more refined maintenance methods.
state into critical and noncritical stages. In the last part, different methods for the evalu-
Different approaches exist for assessing the perfor- ation of reliability indices of existing structure are
mance of bridges. The most commonly used method discussed. It is also pointed out that redundancy and
of bridge evaluation is the so called condition rating robustness of the structure should be considered.
method. All agencies perform such condition ratings
on a regular basis. Condition ratings can be either
numerical ranging between 1 for very poor condi- REFERENCES
tion to 10 for excellent as an example or descriptive
by classifying bridges as poor, acceptable, good, etc. American Association of State Highway and Transportation
Officials (AASHTO), Standard Specification for Highway
But, using the results of the inspection in a compre-
bridges, Washington DC, 2002.
hensive and objective way has been a big challenge. CAN/CSA-S6-00: Canadian Highway Bridge Design Code
In fact, the same bridge, assessed by two different and Commentary, Canadian Society for Civil Engineering
engineers can be rated with different grades. In the (CSCE), 2000.
past three decades, a new measure for the assessment European Committee for Standardization (CEN), UNI EN
of existing structures has been developed within the 1990 Eurocode: Basis of Design, CEN, Brussels, EN
probabilistic framework based on the reliability index. 1990, 2002.
From a probabilistic point of view accepting existing HMSO, Design Manual for Roads and Bridge (DMRB), UK,
structures as sufficiently safe, becomes the result of a 2001.
ISO 2394: General Principles on reliability for structures,
decision-making process guided by some optimality
1998.
criteria. JCSS (Joint Committee on Structural Safety), Probabilistic
The condition rating and the reliability-based per- Assessment of Existing Structures, RILEM Publications,
formance assessment techniques serve the same 2001.
purpose but have different theoretical bases and com-
pletely different formulations.
428
Basic aspects for the uncertainty in the design and analysis of bridges
F. Bontempi & L. Giuliani

ABSTRACT
The aim of this paper is to remark basic ideas about

uncertainty related to structural problems connected
with the design of bridges and to devise suitable
strategies that can be pragmatically applied in real
problems.
The main ideas about risk, evolutive vs. innovative
design (Fig.1), high probability low consequences
(HPLC) vs. low probability high consequences
(LPHC) (Fig.2) have been firstly presented, while two
well-known case histories have been then pertinently
reported. Figure 2. HPLC vs. LPHC situations and corresponding
Both the ideas and the examples appear simple analysis strategies.
or even trivial. Nevertheless, it seems that there is
not sufficient consideration for the features connected
(Handling Exceptions 2008).
high consequences scenarios are not easily amenable
One possible reason is that the strategies neces-
of analytic treatment.
sary to handle innovative design or low probability
Fortunately, these situations can be handled with
engineering knowledge and can be in the future even-
tually solved by Artificial Intelligence methodologies.
Finally, for the operative point of view, the deep
consideration for robust structural systems appears
indispensable (Starossek, 2009).
It is expected that this material can be useful for
students and practitioners, more than for researchers
and scholars.
REFERENCES
Handling Exceptions in Structural Engineering 2008, Rome,
DOI:10.3267/HE2008.
Figure 1. Evolutive/Innovative designs in the Starossek, U., (2009) Progressive collapse of structures,
(performance parameter) space. Thomas Telford.
429
Block shear failure of steel gusset plates
C. Crosti
University of Rome La Sapienza, Rome, Italy
D. Duthinh
National Institute of Standard and Technology, Gaithersburg, MD, USA
ABSTRACT since Atn < 0.58 Avn , failure is by block shear. The ulti-
mate load is 2088 kN at a displacement of 14 mm, and
Following the catastrophic failure of the I-35W bridge thus the LRFR value provides a factor of safety of 2.14.
in Minnesota in 2007, the Federal Highway Adminis- Plate 4 has tension length Lt of 0.150 m and shear
tration issued guidelines for the load rating of bolted length Lv of 0.273 m. In this case, Atn > 0.58 Avn and
and riveted gusset plates in truss bridges (FHWA, failure is by tension. The ultimate load is 2190 kN at a
2009). This paper develops finite-element models displacement of 7.8 mm, and the LRFR value provides
capable of predicting the behavior of gusset plates in a factor of safety of 2.38.
tension, resulting in possible failure by block shear, Thus, Plates 2 and 3, which failed by block shear,
for comparison with the guidelines. Block shear fail- even though they had rather different hole arrange-
ure is a limit state that combines tension failure on one ment, behaved very similarly. Plate 4, which failed by
plane and shear failure on a perpendicular plane. The tension, had a slightly higher factor of safety but less
analysis accounts for the nonlinearity of the material ductility than Plates 2 and 3. In all cases, the FHWA
and large displacements. For validation, Plate 1 simu- values are safe and adequate.
lates a physical test by the University of Alberta (UA). Compared to Plate 2, Plate 5 has a third row of
The material is bilinear elasto-plastic, with Youngs bolts. The ultimate load is 3468 kN at a displacement
modulus of 215 GPa, yield strength of 410 MPa and of 47.3 mm, and the LRFR value provides a factor
tangent modulus of 2.15 GPa. The analysis uses true of safety of 2.73. Results show the beneficial effect
stress and true strain and the tension load is applied of adding internal bolts, which increase the strength
by three point loads on the upper half of each bolt and ductility of the gusset plate. A simplified analy-
hole. Results from the present two STRAND meshes sis that only accounts for the perimeter bolts would
agree well with the ABAQUS model used by UA. All underestimate the strength and ductility of the gusset
three finite-element results slightly underestimate the plate.
test results, especially at the onset of yielding. The Finally we analyze one of the example plates from
Load Resistance Factor Rating (LRFR) value falls on the guidance document. Its ultimate load is 10750 kN
the limit of the elastic range, whereas the unfactored at a displacement of 43.2 mm, and thus the LRFR value
FHWA value produces a small amount of yielding. provides a factor of safety of 2.28.
Plate 2 has a tension length Lt of 0.139 m, a shear In conclusion, nonlinear finite-element analysis val-
length Lv of 0.358 m and 16 bolt holes arranged in two idated by experimental data confirms the safety and
rows. Since net tension area Atn < 0.58 Avn net shear validity of the FHWA load rating formulas for the
area, failure is by block shear. Assuming as failure block shear strength of riveted and bolted gusset plates.
criterion a maximum strain of 100%, results show an For a variety of geometries, the Load and Resistance
ultimate load of 2720 kN at a displacement of 14 mm, Factor Rating (LRFR) value produces factors of safety
and thus the LRFR value provides a factor of safety of between 2.1 and 2.7. This study also provides guid-
ultimate load/LRFR value of 2.15. ance on the mesh density required around the holes,
The bolts in Plate 3 are distributed over a narrower the application of bolt loads, and the approximation
width. This gusset plate has a tension length Lt of involved in modeling the perimeter holes only.
0.0462 m and a shear length Lv of 0.358 m. Again,
430
Generation of modified earthquake time-histories using Hilbert-Huang

transform
S.-H. Ni, W.-C. Xie & M.D. Pandey

Department of Civil and Environmental Engineering, University of Waterloo, Waterloo, Ontario, Canada
ABSTRACT the target design spectrum based on an actual earth-

quake record, is proposed. The actual earthquake
Seismic Response History Analysis (SRHA) is a major record is decomposed to obtain several components of
analysis method for seismic qualification. Represen- amplitude and frequency by Hilbert-Huang Transform
tative earthquake time-histories are required for this (Huang et al. 1998). The compatible time-history is
analysis method. then constructed using optimization technique to min-
There are three types of earthquake time-histories: imize the difference between response spectrum of the
actual earthquake records, modified earthquake time- modified time-history and the target design spectrum.
histories based on real earthquake records, and arti- The results obtained show that the proposed modifica-
ficial earthquake time-histories. In eastern North tion method is simple and credible in generating the
America (ENA) area, there are very few actual earth- required compatible time-histories.
quake records available.
According to existing earthquake-resistant codes
(e.g. ASCE 2005, CSA 2006), modified time histories REFERENCES
compatible with the target design spectra are required.
It is thus practically important to be able to gener- ASCE 2005. Seismic Design Criteria for Structures, Systems,
and Components in Nuclear Facilities. ASCE/SEI Stan-
ate modified time-histories based on the few available
dard 43-05. Structural Engineering Institute, American
earthquake records. Society of Civil Engineers.
Existing modification method (ASCE 2005) is to CSA 2006. Canadian Highway Bridge Design Code
scale the Fourier amplitudes of the actual record such CAN/CSA-S6-06. Canadian Standard Association.
that the resulting response spectrum is compatible with Huang, N.E., Shen, Z., Long, S.R., Wu, M.C., Shih, H.H.,
the target design spectrum. Zheng, Q., Yen, N.-C., Tung, C.C. & Liu, H.H. 1998. The
However, it is difficult and not practical to use this empirical mode decomposition and the Hilbert spectrum
method to generate compatible time-histories because for nonlinear and nonstationary time series analysis. Pro-
of the deficiencies of the Fourier transform and the ceedings of the Royal Society London A: Mathematical,
Physical and Engineering Sciences 454: 903995.
complexity of the procedure.
In this paper, a new modification method, which
can generate a modified time-history compatible with
431
Influence of model parameter uncertainties on the seismic vulnerability

analysis of continuous steel-concrete composite bridges exhibiting
dual-load paths
E. Tubaldi
Dipartimento di Architettura Costruzione e Strutture, Universit Politecnica delle Marche, Ancona
M. Barbato
Department of Civil & Environmental Engineering, Louisiana State University, Baton Rouge, Louisiana, USA
A. DallAsta
ProCAM, Dipartimento di Progettazione e Costruzione dellAmbiente, University of Camerino, Viale della
Rimembranza, Ascoli Piceno (AP), Italy
ABSTRACT
The performance of multi-span steel-concrete com-

posite bridges in recent seismic events has shown
that these structures are very sensitive to earthquake
loading. Following the Performance-Based Earth-
quake Engineering (PBEE) framework, the vulnera- Figure 1. Bridge longitudinal profile.
bility assessment of these systems must take rigor-
ously into account all pertinent sources of uncertainty,
including uncertainties in the loading and in the
structural/mechanical/geometrical properties (model
parameter uncertainty). The model parameter uncer-
tainty affects not only the structural capacity, but also
the Engineering Demand Parameters (EDPs). How-
ever, most of the procedures for fragility calculation
focus on the variability of EDPs due to input ground
motion uncertainty (record-to-record variability) but
neglect model parameter uncertainty effects, or incor- Figure 2. System fragility curve accounting for randomness
porate these effects only in a simplified way. This only (R) and randomness and uncertainty (RU).
paper aims at studying the effects of model parameter
accounting for both randomness and model parameter
uncertainty on the seismic response and on the seis-
uncertainty (denoted by RU).
mic vulnerability of steel-concrete composite bridges
It is observed that, for the considered benchmark
with abutment transverse restraints. In this paper,
structure, (a) the effects of seismic ground motion
the Extended Incremental Dynamic Analysis (EIDA)
uncertainty are predominant compared to model
method (Vamvatsikos & Fragiadakis 2009) is used to
parameter uncertainty, (b) model parameter uncer-
account for all sources of aleatoric uncertainty. EIDA
tainty has a negligible effect on the system fragility,
is applied to a benchmark SCC bridge system with
and (c) the correlation between different failure modes
dual load path (Figure 1) whose seismic response has
is significant.
already been investigated in Tubaldi et al. (2009). By
means of EIDA, the first- and second-order statistics REFERENCES
of the EDPs of interest are estimated, the sensitivity
of the structural response to both model parameter Tubaldi, E., Barbato, M. and DallAsta, A., 2009. Paramet-
uncertainty (i.e., uncertainty, U) and ground motion ric study of continuous steel-concrete composite bridges
uncertainty (i.e., randomness, R) is assessed, and exhibiting dual load path, Technical Report, Department
of Civil and Environmental Engineering, Louisiana State
the seismic vulnerability accounting for all pertinent University and A&M College, Baton Rouge, LA.
sources of uncertainty is evaluated. Vamvatsikos, D. and Fragiadakis, M., 2009. Incremental
In Figure 2, the system fragility curve accounting dynamic analysis for estimating seismic performance
for seismic ground motion uncertainty only (denoted sensitivity and uncertainty, Earthquake Engineering and
by R) is compared with the system fragility curve Structural Dynamics, In Press.
432
Reliability evaluation of complex bridges under multiple limit states
J.S. Zhu
Tianjin University, Tianjin, P.R. China
D.M. Frangopol
ABSTRACT recently (May 2009), It is located in Tianjin, China.

The results of the reliability evaluation can be used
Efficient maintenance, repair and rehabilitation of to establish a cost-effective health monitoring for this
existing bridges require the development of a method- bridge.
ology that allows for an accurate assessment and
prediction of bridge performance. Reliability concepts
and methods have been used to quantify the structural REFERENCES
performance when uncertainties are involved. How-
ever, there are very few successful real-life examples Das P. C. 1998. Application of reliability analysis in bridge
management. Engineering Structures 20(11): 957959.
on the combined use of reliability theory and struc-
Estes A.C. & Frangopol D.M. 2001. Bridge lifetime system
tural health monitoring (SHM) in bridge management. reliability under multiple limit states. Journal of Bridge
Therefore, it is a great challenge to develop an effective Engineering, ASCE 6(6): 523528.
reliability analysis method which can be used together Frangopol D. M. 2009. Life-cycle performance, manage-
with SHM in bridge management systems. ment, and optimization of structural systems under uncer-
In this paper, a radial basis function (RBF) is pro- tainty: Accomplishments and challenges, Safety, Reliabil-
posed to simulate the response of complex bridges, ity and Risk of Structures, Infrastructures and Engineering
and the uniform design method is used to construct Systems, H. Furuta, D.M. Frangopol, and M. Shinozuka,
the training datasets for the RBF network (Zhu & eds. 3860, 2010, CRC Press,Taylor and Francis Group,
London.
Xiao 2007, Zhu et al. 2007). The finite element soft-
Frangopol D.M. & Liu M. 2007. Maintenance and manage-
ware ANSYS is used to obtain the structural response ment of civil infrastructure based on condition, safety,
of complex bridges. The RBF neural network-based optimization, and life-cycle cost, Structure and Infrastruc-
meta-model is combined with Monte Carlo Simula- ture Engineering, Taylor & Francis 3(1): 2941.
tion (MCS) in order to improve the computational Liu M. & Frangopol D.M. 2004. Optimal bridge maintenance
efficiency of reliability analysis of complex structures. planning based on probabilistic performance prediction.
The main objective of this paper is to present a Engineering Structures 26(7): 9911002.
RBF-MCS method for reliability evaluation of com- Zhu J.S. & Xiao R.C.2007. Research on neural network
plex bridges under multiple limit states. A long span model to structural simulation of large-span PC cable-
stayed bridges. China Railway Science, 28(1): 5864. (in
steel truss arch bridge, the Guotai bridge (see Figure 1)
Chinese).
has been analyzed in order to illustrate the efficiency Zhu J.S., Xiao R.C. & He L.Z. 2007. Reliability assessment of
of the proposed methodology for multiple limit state large-span cable-stayed bridges based on artificial intelli-
reliability evaluation of complex bridges. The Guo- gence. China Civil Engineering Journal 40(5): 4147.(in
tai bridge with the main span of 146 m opened very Chinese).
Figure 1. Guotai Bridge in Tianjin.
433
SS10: Life cycle bridge engineering in Korea
Organizers: H.-N. Cho & J.-S. Kong
Study on bridge asset management in Korea based on infrastructure asset

management methodology
K.-H. Park, J.-W. Sun, & Y.-K. Hwang

C. Park
Department of Civil Engineering, Kangwon National University, Samcheok, Korea
M.-J. Lee
Department of Civil Engineering, Choongnam National University, Daejeon, Korea
ABSTRACT Table 1. Definition of bridge LOSs.
Recently, researchers in the field of bridge mainte- Customer value LOS (Level of Service)
nance and management have tried to enhance the
satisfaction of both users and agencies of social infras- Sustainability Provide sustainable services without
adverse function to the environment
tructures by applying the asset management concept.
Approachability* Allow users to approach bridges
Asset management of bridges is clearly a national mat- without restrictions
ter that must be set up in preparation for a possible Validity of costs** Efficiently manage bridges
sharp increase in maintenance costs in the near future. Quality* Validate bridge performance of
Nevertheless, the asset management has still remained basic functions
incomplete, given the lack of relevant government sup- Health & safety* Let users safely use bridges
ports or policy in Korea. Therefore, it is urgent to carry Reliability and Provide predictable and continuous
out in-depth studies to build a bridge management sys- responsiveness** services
tem at the asset management level by establishing a Minimize impact by promptly
counteracting emergencies
support system for making decisions such as the for-
Customer service** Kindly respond to service demand
mulation of bridge maintenance strategies and the log- Minimize civil complaint
ical budget allocation. In Korea, diverse efforts have
recently been performed to build an asset management
system for the entire infrastructure facilities. functions of the suggested processes are defined con-
This study is to develop more efficient asset man- sidering the domestic circumstances in Korea. This
agement framework for bridge management. The basic study also defines a diversity of LOS (level of service)
concept of a bridge asset management is defined and of bridges with respect to the accomplishment of the
the core technologies to establish systems are pro- local communities and customer values, and clarifies
posed by analyzing the current status and functions performance measures for quantitative assessment of
of the existing bridge and asset management systems the LOS. By dividing asset management of bridges
both in and out of Korea. In addition, the bridge into the establishment of strategies and that of plans,
asset management process is suggested and major various decision-making subjects are outlined.
REFERENCES
FHWA (Federal Highway Administration) 1999. Asset man-
agement Primer, Office of Asset Management, US DOT.
INGENIUM and IPWEA 2006. International Infrastructure
Management Manual -Version 3.0, ISBN 0-473-10685-X.
Kobayashi Kiyoshi 2008. Efficient Budgeting for Sustainable
Management, Bridges and Foundations, No.8, 109111.
Korea Institute of Construction Technology (KICT) 2006.
Development of life-cycle cost analysis method and sys-
tem for the life-cycle cost optimum design and the
life-time management of steel bridges, Research Report,
Ministry of Land, Transport and Maritime Affairs
(MLTM) and Korea Institute of Construction and Trans-
portation Technology Evaluation and Planning (KICT-
Figure 1. Bridge asset management process. TEP) (in Korean).
437
Consideration of safety in the revision of the bridge management

system in Korea
H. Park
IAM corporation, Korea University, Seoul, Korea
J.-S. Kong & J.-H. Kim

Korea University, Seoul, Korea
K.-H. Park
Korea Institute of Construction Technology, Ilsan, Korea
S.-Y. Oh
Ministry of Land, Transport and Maritime Affair, Gwacheon, Korea
ABSTRACT deterioration but standardized safety deterioration

models were not available in general.
The bridge population in Korea was already large and So under these circumstances, the standardization
ageing. At that time, little consideration was given to of performance degradation and BMS with consider-
assess the bridge performance with safety. This situa- ation of safety is immediately necessary to develop.
tion may lead to loss of maintenance cost at a national In this study, the current status of bridges in Korea
wide due to no considerable strategy based on safety has been investigated through arrangement of safety
of the bridge. Therefore it is growing awareness that variation with/without maintenance activity. Also we
requirement of long-term bridge performance evalu- have developed practicable performance degradation
ation for bridge management in the future, including model classified by some typical type of bridges and
application of LCC in Korea. However most of Korea members to derive optimum maintenance scenario
bridge management system has not considered per- considering safety. The model has been developed
formance degradation with safety to evaluate bridge based on analytical/computational framework for sta-
condition. tistical regression. And the revision of Bridge Manage-
Such various condition state evaluation methods as ment System have considered safety model to manage
eye inspection, load carrying capacity evaluation, peri- bridges efficiently
odic inspection, diagnostic workup and movements
monitoring analysis etc. have been developed for effi-
cient maintenance of social infrastructures, however, REFERENCES
these methods are mostly remained in the initial stage
of research and not enough to use for the engineers in Korea Institute of Construction Technology (2005), Life
the field because the methods are too complicated or Cycle Cost Analysis Method and System Development
for Optimum Design and Economic Maintenance of Steel
simplified to be applied.
Bridges.
These systems, collectively named Bridge Man- KISTEC (2000), Plan for Elongation of Service Life of Road
agement Systems, have been developed with varying Bridges.
degrees of sophistication. Early system were mainly KISTEC (2000), Criteria of Relative Evaluation of Facilities
databases which stored information collected during (Bridge).
inspections together with inventory data such as the An, Yeong-Gi (2003), Research of Deterioration Prediction
location of the bridge, road category and the con- Model for Estimation of Life Cycle Cost.
struction date of each structures, then, over time, more Lee, Seung-Jun (2004), Credibility Analysis of Structure by
functions have been produced and systems have been Response Surface Method.
Achintya Haldar (2000), Probability, Reliability, and Statis-
developed which enable the user to undertake activities
tical Methods in Engineering Design.
such as inspection planning, deterioration prediction, Frangopol, D.M., Lin, K.Y. (1997), Life-Cycle Cost Design
structural assessment and the economic evaluation of of Deterioration Structures.
repairs George E. Ramey (1997), Bridge Deterioration Rates and
To derive the optimum maintenance scenario, it Durability/Longevity Performance, Practice Periodical on
is needed to analyze safety of bridge caused by Structural Design and Construction.
438
Development of life-cycle analysis based new bridge management

system in Korea
J.-W. Sun, K.-H. Park, S.-Y. Lee, & J.-J. Song

J.-S. Kong
Civil, Environment and Architecture Engineering, Korea University, Seoul, Korea
H.-Yong
ITS & Road Environment Division, Office of Transport Policy, Ministry of Land, Transportation and Maritime
Affairs, Korea
ABSTRACT
Different to other social infrastructures, bridges can

be collapsed and this may cause death toll and severe
social and economical damage. The occasional col-
lapse of bridge makes aware of the importance of
accurate safety inspection and the establishment of
budget for performing the appropriate repair and rein-
forcement as well as the acknowledgement of error
in design and maintenance. Therefore, it is neces-
sary to establish the rational and scientific decision
making method which can decide appropriate main-
tenance budget based on bridge life-cycle condition
and performance. The function of bridge management
can be briefly divided into three categories, perfor-
mance management, asset management and decision
making including life-cycle analysis through the status
analysis of each countrys research development.
Recently, Ministry of Land Transport and Mari-
time Affairs (MLTM) of Korea has re-searched on
the existing bridge management system (BMS) and
its improvement. The current BMS for national road
managed by MLTM was developed in late 1990s.
However, its computational operating system was only
changed but its performance and function have limited
in simple function such as inventory and inspec-
tion data storage. For developing new BMS with the Figure 1. Analysis procedure to establish the optimal main-
analysis of research trend, the research has been devel- tenance strategies.
oping classified into the performance evaluation, the
maintenance cost analysis and the decision making REFERENCES
technique for optimal maintenance strategy and bud-
Korea Institute of Construction Technology (KICT) 2006.
get plan. The contents of research for establishing the Development of life-cycle cost analysis method and system
advanced BMS combining asset management concept for cost-effective optimum design and economical main-
are summarized as follows; first, it may need life-cycle tenance of steel bridges, Research Report, MLTM and
performance evaluation and estimation of the effect of KICTTEP (in Korean).
maintenance interventions, and asset evaluation and Ministry of Land, Transport and Maritime Affairs (MLTM)
its calculation technique related to maintenance cost. and Korea Infrastructure Safety and Technology
Second, decision making technology based on these Corporation (KISTEC) 2009. Detailed Guidelines of
various performances and cost related information is Safety Inspection and Precise Safety Diagnosis for Bridges
(in Korean).
necessary for the life-cycle management of bridge.
Transportation Research Board (TRB) 2008. International
Finally, a proper computational system and practical Bridge and Structure Management, 10th International
education for applying a new method and system Conference on Bridge and Structure Management.
should be developed.
439
SS11: ARCHES: Assessment & rehabilitation of
Central European highway structures
Organizers: T. Wierzbicki & J.R. Casas
Smart cathodic protection systems
R.B. Polder
TNO Built Environment and Geosciences, Structures and Safety, Delft, The Netherlands
Delft University of Technology, Civil Engineering and Geosciences, Delft, The Netherlands
J. Leggedoor & G. Schuten

Care4concrete.nl, Gasselternijveen, The Netherlands
A. Sajna & A. Kranjc

ZAG, Ljubljana, Slovenia
ABSTRACT CP is based on conservative assumptions. Lightweight

anode materials and numerical modelling in the design
Cathodic protection (CP) is intended to deliver cor- phase may provide more cost effective systems, which
rosion protection in concrete structures exposed to can be termed smart CP systems. A CP trial on
aggressive environments, e.g. in deicing salt and a bridge substructure in Slovenia was carried out
marine climates. Although preventative application is in European FP6 research project ARCHES. In the
possible (called Cathodic Prevention), most cases con- trial, various types of lightweight anodes were applied
cern structures that already have developed corrosion including conductive coating and titanium strip anodes
and some level of damage to the concrete cover (crack- in mortar dykes on the concrete surface. Numerical
ing, spalling). CP involves polarizing the reinforcing Finite Element modelling was applied to the trial CP
steel by a low amount of direct current, originating systems. In the modeling a geometrical analogue is
from a conductor on the surface or in the cross sec- built in two dimensions, as three-dimensional finite
tion of the structure connected to a low voltage power element modeling is beyond present computer capa-
source. Spalled areas need to be repaired using normal bilities. Input parameters are either obtained from the
concrete repair methods. A large difference between structure (steel potentials, concrete resistivity) or from
CP and the conventional repair approach is that in the literature (electrochemical parameters). In our calcu-
latter case, all chloride contaminated concrete has to be lations the input parameters had to be fine-tuned to a
removed and new has to be reapplied. With CP, physi- certain extent in order to obtain realistic current levels.
cally sound but chloride contaminated concrete can be With the adjusted input, local polarisations were cal-
left in place. European experience with many hundreds culated in good to reasonable agreement with values
of CP systems shows that they perform well, provided measured in the trial. Consequently, it is possible to
that a minimum of (routine) maintenance is carried predict the performance of a CP system in the design
out. An inventory in the Netherlands of one hundred phase, which allows for economical optimisation. For
well-documented CP systems shows that working lives this case, life cycle costs were calculated for conven-
are at least more than 13 years and probably more than tional repair (assuming 25 years life of the repairs
25 years. Individual cases that needed maintenance are and also taking into account that re-repairs would be
discussed, based on the need to replace electrical con- needed half way the 25 year life), CP with a conductive
nections, reference electrodes, power units and anode coating (including re-application of the coating once in
materials. All such cases have occurred to a limited a 25 year period) and titanium mesh with a shotcrete
extent. It is felt that the quality and life of items need- overlay (that will last at least 25 years). CP proved
ing replacement in the past have now been improved to be more economical than conventional repair, tak-
or have become considerably cheaper, allowing some ing into account that the life of conventional repairs is
redundancy to be installed. Maintenance is generally limited, as suggested by a European study, and conse-
carried out by the contractor who installed the system, quently there is a significant probability that re-repairs
for a relatively low annual fee. This involves routine are needed. For the trial site, it was found that up to 10%
checks of power units and testing depolarisation of lower life cycle costs over a period of 25 years were
reinforcement at least twice a year. Regular design of possible with CP compared to conventional repair.
443
Assessment and monitoring of existing bridges to avoid unnecessary

strengthening or replacement
J.R. Casas
Civil Engineering Department. Universitat Politcnica de Catalunya. Barcelona, Spain
ABSTRACT An accurate estimation of traffic load in the bridge

can be carried out by obtaining a good estimate of
The increasing volume of European transport urgently the static load of vehicles and of the dynamic ampli-
requires an effective road and rail system in New fication factor (DAF). Thus, the paper is divided into
Member States (NMS) of the European Union (EU) 3 main parts also related to 3 of the main tasks of
and Central and Eastern European countries (CEEC). the project: 1) Bridge traffic load monitoring aimed
To bring this transport infrastructure up to modern to quantify the increased traffic loading on bridges
European standards will require an immense invest- in NMS after EU enlargement. It combines avail-
ment and therefore difficult to achieve in the medium able information and newly collected traffic load data
term. New motorways will be required with many new (weigh-in-motion (WIM) measurements) to propose
bridges. Numerous existing bridges will need to be simplified assessment traffic load models at two lev-
assessed, and a large portion of them improved or els: by taking into account the measured WIM data
replaced. on a bridge of concern (a site-specific model) or by
The European project ARCHES (Assessment and applying more generalized, but measured, WIM data
Rehabilitation of Central European Highway Struc- at the network level (road-specific). The main results
tures) had as main objective to develop ways to raise are available in ARCHES-D08 (2009). 2) The task on
the standard of the highway structures of NMS and Load carrying capacity based on load testing results
CEEC to the level necessary for their full economic optimized bridge capacity assessment by using load
integration into the EU and for the future develop- tests or computational analysis to find reserves in load
ment of the Union. Therefore, it becomes important carrying capacity. Results of this task are reported
to develop more appropriate tools and procedures to in ARCHES-D16 (2009). 3) The task on Reducing
avoid unnecessary interventions in bridges. Rehabil- dynamic loading of bridges dealt with the assess-
itation and replacement should be as far as possible ment of dynamic impact on bridges. It is shown in
avoided by developing better safety assessment meth- ARCHESD10 (2009) how heavy trucks and multi-
ods that may result on an extended life-time of the presence of vehicles can lead to an important reduction
structure. This has been analyzed within Workpack- of this dynamic interaction.
age 2 (WP2) of ARCHES. An improved assessment
needs both: first, an accurate assessment of the actual
bridge capacity and, second, a good assessment of the
actual loading. The last one is mainly referred to the REFERENCES
actual traffic live load. Unfortunately, traffic loading
conditions in NMS and CEEC highway structures are ARCHES-D08 2009. Recommendations on the use of results
mostly unknown. The reason for such a situation is of monitoring on bridge safety assessment and main-
a lack of consistent systems that collect traffic data tenance. Deliverable D08. ARCHES Project. VI EU
in a way appropriate for bridge design and assess- Framework Program. Brussels (available on line at http://
ment. The load carrying capacity of many highway arches.fehrl.org)
structures is not known either, especially for very old ARCHES-D10 2009. Recommendations on dynamic
allowance. Deliverable D10. ARCHES Project. VI EU
bridges where the design and construction documents
Framework Program. Brussels (available on line at http://
are not available. With that many unknowns, both on arches.fehrl. org)
the loading and on the resistance sides, it becomes ARCHES-D16 2009. Recommendations on the use of soft,
really difficult to obtain a reliable estimate of the actual diagnostic and proof load testing. Deliverable D16.
bridge safety and serviceability and to propose optimal ARCHES Project. VI EU Framework Program. Brussels
rehabilitation measures. (available on line at http://arches.fehrl.org)
444
Composite UHPFRC-concrete construction for rehabilitation most

recent advances and applications
A. ajna & J.S. uput

Slovenian National Building and Civil Engineering Institute, ZAG, Ljubljana, Slovenia
E. Denari & E. Brhwiler

Ecole Polytechnique Fdrale de Lausanne, MCS-IIC-ENAC, Lausanne, Switzerland
G. Habert & P. Rossi

L. Recic
Salonit Anhovo, Deskle, Slovenia
T. Wierzbicki
IBDIM, Warsaw, Poland
ABSTRACT
The increased volume of European transport urgently

requires an effective road and rail system in Cen-
tral European and Eastern countries (CEEC) with a
major investment in building new and assessing and
rehabilitating old structures.
Ultra-High Performance Fibre Reinforced Con-
cretes (UHPFRC) are characterized by a unique com-
bination of extremely low permeability, high strength
and deformation capacity (tensile strain hardening),
Extensive R&D works performed during EU projects
SAMARIS and ARCHES have demonstrated that the
UHPFRC technology is fast, efficient and price com-
petitive for the rehabilitation or reinforcement of Figure 1. Global Warming potential for the Log cezsoki
reinforced concrete structures. rehabilitation, considering the life cycle. Traditional rehabil-
An innovative concept of Ultra High Performance itation system with standard concrete = 100%, Habert et al.
matrix with a high dosage of mineral addition has been (2009).
developed that makes the application of UHPFRC
technology feasible with a wide range of locally avail- demonstrated at an industrial scale the ability of the
able cements and superplasticisers. This concept was newly designed UHPFRC mixes to reply to the dif-
validated with both Slovenian and Polish raw materi- ficult challenges of the site, without any increase of
als. In the next step the rheology of those mixes was rehabilitation costs, but to the great satisfaction of the
adapted to enable them to accommodate 5% slopes owner, user and contractor.
of the substrates at fresh state. This new material was Further, the newly designed E-UHPFRC recipes
successfully applied in July 2009, for the rehabilitation have a dramatically reduced cement content which
of the deck and footpaths of the Log Cezsoki bridge makes them more economical and particularly attrac-
(Slovenia) with challenging slopes of 5%, opening the tive from an environmental point of view.
way to a wider dissemination of this concept, in most
demanding conditions of application, see ARCHES
deliverables D06 and D14, Denari et al. (2009a and b). REFERENCES
The intervention was fast (1 month instead of 3 Brhwiler E., Denari E., .2008, Rehabilitation of concrete
month with traditional technique) and by means of a structures using Ultra-High Performance Fibre Rein-
newly developed surfacing technique it was possible to forced Concrete, in Proceedings UHPC-200, March 05
achieve uniform textured UHPFRC footpath surfaces 07, 2008, Kassel, Germany. Kassel University Press,
on which barefoot walking is possible.This application pp. 895902.
445
Denari et al. 2009a, Recommendations for the tailoring of Habert, G., ajna, A., Rossi, P., Denari, E., 2009, Environ-
UHPFRC recipes for rehabilitation, deliverable ARCHES mental evaluation of a bridge rehabilitation system with
D06, http://arches.fehrl.org/. a new local material based UHPFRC, Journal paper in
Denari et al. 2009b, Recommendations for the use of preparation.
UHPFRC in composite structural members rehabili-
tation Log ezsoki bridge, deliverable ARCHES D14,
http://arches.fehrl.org/.
446
Corrosion resistant steels as reinforcement in concrete
I. Stipanovic Oslakovic & D. Bjegovic

Institut IGH d.d. (Civil Engineering Institute of Croatia), Zagreb, Croatia
M. Serdar
Faculty of Civil Engineering University of Zagreb, Croatia
A. Legat, A. ajna, T. Kosec, V. Kuhar & N. Gartner

Slovenian National Building and Civil Engineering Institute, ZAG, Ljubljana, Slovenia
ABSTRACT Results that were achieved and evaluated during

the ARCHES program included a laboratory study of
Maintenance and repair costs of concrete structures rebars in pore solution, rebar in concrete examination
constitute a major part of the current spending in and two field tests of columns and a large concrete
infrastructure worldwide and often overhead the costs reinforced structure.
of structures themselves. The enormous costs and The use of corrosion resistant steel has not been
safety issues associated with corrosion of steel in widely adopted because of perceived concerns relat-
concrete have resulted in the development of a wide ing to initial cost penalties. Overall conclusion from
range of new technologies and materials to increase the project is that the selective replacement of carbon
the durability of reinforced concrete structures and steel with corrosion resistant steel and with tailored
their repairs. One of these methods is the application selection of stainless steel grade the increase of initial
of stainless steels, which are resistant to corrosion in investment can be less than 10%, which is negligible
concrete environment even when the concrete is highly comparing to the reduction in maintenance and life
contaminated with chlorides, as substitution of a part cycle costs of the structure.
or all reinforcement in concrete. (Nrnberger, 1996).
The aim of research within working task 3.1 Val-
idation and application of low-alloyed steel of the ACKNOWLEDGMENTS
specific targeted research project ARCHES was to val-
idate the application of alloyed steels in the concrete This research was performed within projects The
and finally to prepare Recommendations for the use of Development of New Materials and Concrete
corrosion resistant steels. In this paper the overview Structure Protection Systems (082-0822161-2159),
of experimental results are presented. Experimental funded by Croatian Ministry of education, science and
program was designed with the aim of evaluating cor- sport and ARCHES (031272), funded by the European
rosion behavior of alloy steel in non-corrosive and Community under the Sixth Framework Programme.
corrosive environment.
Corrosion behavior of seven different steel types
was studied: two types of steel with lower content REFERENCES
of alloying elements: TOP12 (1.4003) and 204Cu
(1.4597), two grades of duplex steel: UGIGRIP 4362 Nrberger, U. 1996, Stainless Steel in Concrete, State of the
Art Report, Publication No. 18, Institute of Materials,
(1.4362) and SAE/UNS S3 2205 (1.4462), two grades
London; UK.
of stainless steel: AISI 304 (1.4301) and AISI 304L Schiegg, Y., Vote, C.-H., Peter, H., Hasler, S., Urlau, U.
(1.4306) and B500B for comparison. 2004, Initiation and Corrosion Propagation of Stainless
The research program was divided into: Steel Reinforcements in Concrete Structures, Proceed-
ings Eurocorr04, Nice, France.
laboratory testing of steel specimens in pore solu-
Shreir, L. L., Jarman, R. A., Burstein, G. T. 1994, Cor-
tions with different pH values and chloride con- rosion. Vol. 1, Metal/Environment Reactions and Alloys,
tents; Butterworth-Heinemann, 3rd Edition, Oxford, UK.
laboratory testing of steel, embedded into small ajna, A., Legat, A., Bjegovic, D., Kosec, T., Stipanovic
cylindrical and prismatic concrete specimens and Oslakovic, I., Serdar, M., Kuhar, V., Gartner, N., Pardi,
medium size concrete specimens; L., Augustynski, L. Recommendations for the use of
on site exposure and testing of steel embedded into corrosion resistant reinforcement, D11 Report, Project
larger concrete slabs and columns. ARCHES, 2009, in process of publishing.
447
SS13: Current advancements in bridge technology
Organizer: A.H. Malik
Planning and design of 1-pylon suspension bridge, Dandeung Bridge
H.S. Choi, J.G. Yoon & J.H. Kim

DAELIM Industrial co. ltd., Seoul, Korea
ABSTRACT
The Dandeung suspension bridge, the 1-pylon sus-

pension bridge connecting Sinsido and Munyeodo in
Korea, is due to be constructed in the second section
of connecting road works of the Gogunsan islands in
Jeollanamdo.This project is started in November, 2009
and the construction period is just 48 months. The con-
struction of this bridge has been planned to make traffic Figure 1. Aerial view of Dandeung bridge.
for inhabitants of the islands convenient, to promote
local development, and to improve the international
Table 1. Structural dimensions and design condition.
marine tourism circumstance. This bridge is designed
to have one pylon and asymmetric main cable config- Component Design
uration. The main span length of Dandeung bridge is
400 m which will be the worlds longest one among Span composition Cable : 425 + 280 = 705 m
suspension bridges with 1-pylon. Girder : 400 m
In the basic planning of this project, Dandeung Girder Type Edge box girder
bridge was designed as a self-anchored suspension Slope
bridge with 2 pylons, 220 m long main span and two Longitudinal 2%
side spans. However, the planning had to be changed to Transverse 2%
a different bridge type to be able to avoid marine works Design Traffic Speed 60 km/h
Width 20.0 m
because the site for this project has the deep depth and
Live Load DB-24, DL-24
rapid current of sea. And the basic planning was tra- Design Wind Speed 34.4 m/s
ditional, and the originality for tourism was lacked. Seismic Load Seismic Grade 1
Consequently, in the basic design, the 1-pylon sus-
pension bridge with the only one main span of 400 m
long that is the worlds longest one was suggested and
planned to avoid the poor marine condition, to stand inclined arrangement, the installing space of each main
for the regional character, and to build a landmark cable is 20 m at anchorage 1, 3 m at the top of pylon,
structure for tourism. and 11m at anchorage 2. The key point of a stiffening
A pylon of Dandeung bridge is designed as a A- girder design is to minimize the weight of girders and
shaped rahmen concrete structure with the height of secure the aerodynamic stability at the same time. As
105 m and a D-shaped additional column symbol- a result, in this bridge, two edge box section is planned
ized a sail of boat. Such a A-shaped and D-shaped as the final design.
pylon can secure the structural safety and stability
of non-symmetric suspension bridge with 1 pylon
under loading and maximize the harmony with the REFERENCES
regional marine landscape. Main cables are planned to Hyunsok, C. & Jaehong, K. 2009. The planning and design
be installed on two planes because two planes arrange- of the Gwangyang suspension bridge in Korea, IABSE
ment can improve the aerodynamic stability of cables symposium on sustainable infrastructure, Bangkok, 811
and bridge and help travelers to turn back toward oppo- September 2009.
site side easily when accidents take place on the bridge. Seungpil, J. 2006. Guideline for the design of cable supported
Also, as main cables are designed to be installed with steel bridge. Seoul: Korean Society of Civil Engineers.
451
Recent development of wireless bridge monitoring system
C.C. Fu
Department of Civil & Environmental Engineering, University of Maryland, College Park, MD, USA
ABSTRACT data acquisition system (DAQ), (3) power supply, (4)

data transmission, and (5) data processing and data
The purpose of this study is to find state-of-the-art storage.
technology by the industry and research institutes Four case studies were presented. Case study 1 was
which use intelligent wireless monitoring systems wireless structural monitoring by Ivoncon on a newly
instead of conventional on-site bridge testing. The replaced fiber reinforced plastics (FRP) bridge deck.
system could include (1) a practical plug-and-play Case study 2 was using Microstrain on the pilot study
battery-operated wireless data acquisition unit, and of Bridge MD140 over MD27. Case Study 3 was
(2) state-of-the-art wireless data transmission and using ATI on the final study of Bridge MD140 over
communication technologies. The second objective MD27. Case study 4 was conducted for a case with
of this project is to investigate further and integrate a portable, proto-type integrated remote (web based)
the off-the-shelf wireless remote monitoring system bridge monitoring system which were field-tested on a
with the latest technology for the projects use in the typical bridge sites. The integrated system was used in
State of Maryland. the Tydings Bridge field testing. Strain gauges were
Based on the survey documented in this report, it installed at four locations on the first (right) lane of
is found that wireless communication, together with the Tydings North Bound Bridge while the traffic was
its applications and underlying technologies, as shown diverted away. Measurement was made and transmit-
in Figure 1, is among todays most active areas of ted through the Verizon cellular networks to be viewed
technology development. Figure 1 shows the require- on a remote site. 12-minute raw data was extracted
ment of hardware and supplies in (1) sensors, (2) from gauge number 3 and is shown in Figure 2.
Figure 1. Remote Wireless Bridge Monitoring System.
Figure 2. 12-minute raw data was extracted from Tyding Bridge gauge number 3.
452
Technical advancements in bridge technology
A.H. Malik
Delaware Department of Transportation, Delaware, USA
ABSTRACT carrying capacity and longer lifespan. The Akashi

Kaikyo Bridge, Hangzhou Bay Bridge, and Sutong
There have been very encouraging technical advance- Bridge are the examples of record breaking cable
ments in bridge technology, which are reflected in the supported bridges, which utilized the latest advance-
construction of remarkable bridge structures through- ments in design, construction, and fabrication using
out the world. This paper will briefly cover some of the high performance materials.
the aspects developed at this time. The use of high ENHANCED CONTRACTING and PARTNERING
performance materials and more sophisticated design Innovative contracting methods which include
and construction specifications to meet the enhanced Design Build, Lane Rental, A+B and Design
live load demands. The major contributor towards this Build have saved time and project cost for the DOTs.
advancement in bridge technology, as in many other DESIGN and ANALYSIS SOFTWARE
fields, is the computer-aided software programs and Bridge design and analysis software has been a
the related innovations. great asset towards the innovative approaches in
the design and global analysis of complex and

ACCELERATED BRIDGE CONSTRUCTION
advanced bridge structures. Model generating tech-
The Accelerated Bridge Construction (ABC)
niques provide specific menu to cover all loadings,
piloted by FHWA aims at minimizing traffic
structural configurations, complex geometry and
delays during construction, minimizing environ-
exact support conditions.
mental impacts and to achieve 100-year service life
BRIDGE DESIGN and ANALYSIS SOFTWARE
for new bridges.
Bridge design and analysis software has been a
HIGH PERFORMANCE MATERIALS
great asset towards the innovative approaches in
The introduction of high performance steel (HPS)
the design and global analysis of complex and
and high performance concrete (HPC) in the bridge
advanced bridge structures. Model generating tech-
technology has established the confidence of the
niques provide specific menu to cover all loadings,
designers, the bridge owners, and the steel indus-
structural configurations, complex geometry and
try. The high strength of HPS and HPC cuts down
exact support conditions.
relatively the number of girders, as well as the
weight of the superstructure and overall cost of the The design and analysis methodologies, construc-
project tion, fabrication, and erection equipments, construc-
RECORD BREAKING LONGER SPAN BRIDGES tion materials and inspection techniques have been
The advancements in bridge technology are tak- developed to meet the ever increasing challenges
ing place in all parts of the world which is of heavy loads, environmental changes and extreme
demonstrated by record breaking spans, heavy load events.
453
Can nanotechnology address todays civil engineering challenges?
D.M. Mulenga & P.C. Robery

Halcrow Group Ltd, Birmingham, United Kingdom
ABSTRACT
A Nanotechnology has become very much the buzz-

word of the 21st century. While tangible benefits
are being realised in industries ranging from med-
ical to consumer products, the conservative con-
struction sector has been slow to embrace the
new technology. While products are available that
claim nanotechnology-based improvements in prod-
ucts ranging from concrete admixtures to repair mor-
tars, are these developments true breakthroughs in
themselves or is it that we can now see structures
and interactions under the nanoscope for the first Figure 1. Lotus Leaf Effect Developments based on Lotus
time? Leaf effect are being applied in road signs, buildings etc.
Identifying step-change developments in concrete
materials and systems due to nanotechnology is a chal-
lenge for practising engineers. Developments in civil addressing civil engineering challenges therefore still
engineering are likely to be small, iterative improve- remains a subject of further research and debate.
ments, such as surface treatments that make concrete This paper explores some of the recent develop-
self cleaning. As more breakthroughs are made, so ments in nanotechnology relevant to construction and
the sector may benefit from intelligent, cost effective, studies the potential benefits on the life and perfor-
safer, durable and environmentally friendly materials mance and management of civil engineering structures
and structures. such as bridges. It highlights areas of actual practi-
Nanotechnology provides access to the world of the cal application of nanotechnology in construction and
smallest things a world that is unfamiliar to the civil looks at products that may have great potential and
engineer, used to dealing in massive constructions; benefit to address current and futures civil engineering
the diameter of a human hair being around 50,000 challenges.
nanometres (50 micrometres). The challenges facing the industry at the start of
Nanoparticles occur naturally or are man-made by- the 21st century are also highlighted and areas of
products (such as vehicle exhaust particulates), but application for nanotechnology discussed.
increasingly nanoparticles are being deliberately engi-
neered. However, the term nano is appearing in all
manner of products and systems. The addition of a REFERENCES
small percentage of nanoparticles is being claimed
to impart wonder properties to a diverse range of prod- Crichton, M., 2002. Prey, HarperCollins Publishers London.
ucts from face creams and car polish to concrete and FHWA Advanced Research Forum, 2005. Summary Report,
Berkeley California, USA (2627 Oct).
repair mortars. However to the suspicious engineer,
Bartos, P.J.M. and Trtik, P., 2002. Nanotechnology RILEM
it has become obvious that in some quarters nano- TC-197-NCM: Nanotechnology in construction materials.
technology is being loosely ascribed to particles Mulenga D.M. and Robery, P.C., 2008. Nanotechnology in
several hundred nanometres in diameter close to the Repair, Maintenance and Asset Management of Bridges
more familiar micrometre size. The impact of Nano- and Structures 12th International Conference + Exhibi-
technology and its potential benefit to the industry in tion, 10th12th June 2008, Edinburgh, Scotland.
454
Impact of longitudinal tendons on long-term deflections of long-span

concrete cantilever bridges
Z. Pan
Department of Civil & Environmental Engineering, University of Maryland, College Park, MD, USA;
Department of Civil Engineering, Southeast University, Nanjing, China
C.C. Fu
Department of Civil & Environmental Engineering, University of Maryland, College Park, MD, USA
Z. L
Department of Civil Engineering, Southeast University, Nanjing, China
ABSTRACT
Due to the advantages of higher stiffness, fewer

expansion joints, saving on large-tonnage bearings,
better comfort for driving, stronger anti-wind and
anti-seismic performance, and more convenient main-
tenance during the period of service, the prestressed
concrete continuous rigid-frame box girder bridge is
one of the preferred design plans within the 100 m
to 350 m span range in China. However, for over ten
years, excessive deflections at the mid-span have com-
monly appeared in long-span prestressed concrete box
girder bridges during their service years, especially
for the segmental, cast-in-place concrete cantilever
bridges. Lessons from some existing bridges, such as
the Koror-Babeldaob Bridge, the Huangshi Yangtze Figure 1. Shear force provided by cantilever tendons.
River Bridge and the auxiliary shipping channel of
Humen Bridge, the design of longitudinal prestress-
Table 1. Upward deformation at mid-span under the three
ing tendons to avoid the excessive deflections seems types of layout of external tendons.
to be extremely important.
This paper is initiated with the objective to improve Type of tendons Upper Bottom Broken-line
the excessive deflections problems by designing lon-
gitudinal tendons. In the beginning, the development Deformation (mm) 11 6 11
of design of tendons for long-span concrete cantilever
bridges in China is discussed. Nowadays, the bent-
down tendons bring back the designers attention,
In addition, taking account of the advantages of
mainly due to the webs bent-down tendons which can
maintaining easily, replaceability, re-tensioning and so
provide a pre-shear force and are very effective to limit
on, the method of external tendons accompanied with
the principal tensile stress. Therefore, the elimination
the internal tendons as the continuity tendons will have
of webs bent-down tendons is found one of the main
a bright future. In the design of the external tendons,
causes for the inclined cracks which are harmful to the
we should find the optimal layout to effectively mit-
deflections. The comparison between the webs bent-
igate the deflections. The efficiency of three types of
down tendons and the straight layout of tendons is
external tendons (upper tendons, bottom tendons and
shown in Figure 1.
broken-line tendons) for reducing the deflections was
Moreover, two approaches used in the continuous
analyzed and compared, shown in Table 1.
rigid frame of Sutong Bridge to mitigate the deflections
are introduced, which are the pre-setting internal ten-
dons in the bottom slabs and the preparatory external REFERENCES
tendons tensioned during the bridges service. Fur-
thermore, the impact of them on the deflections was Bazant, Zdenek P. & Li Guanghua 2008. Explanation of
analyzed. excessive long-time deflections of collapsed record-span
455
box girder bridge in Palau. 8th International Conference long-span continuous rigid-frame bridges. Journal of
on Creep and Shrinkage of Concrete, preliminary report: China & Foreign Highway 25(1): 5658. (in Chinese).
120. Ise-Shima, Japan. L Zhitao & Pan Zuanfeng 2010. Several problems of design-
Zhan Jianhui & Chen Hui 2005. Analysis of causes of ing long-span prestressed concrete box girder bridges.
excessive deflections and cracking of box girder in China Civil Engineering Journal 43(1): 17. (in Chinese).
456
Assessment of effective flexure rigidity and composite capacity

of corrugated steel-concrete composite deck with I-beam welded
J.J. Shim, S.N. Hong, J.M. Park, J.W. Choi & S.K. Park
Sungkyunkwan Univ., Suwon, Korea(south)
K.B. Han
Expressway & Transportation Research Institute, Hwaseong, Korea(south)
ABSTRACT section, and material connecting methods, and tested

to compare and analyze the actual flexural rigidity
The steel-concrete composite structure is applied and the effective flexural rigidity calculated with the
extensively to the construction of bridges and build- formula suggested by ACI.
ings recently. This study suggested a new type of Role of groundwater in upland streamflow genera-
steel-concrete composite deck plate to replace the tion is far more complicated than previously consid-
cast-in-place reinforced concrete deck of the past. ered and has important implications for upland water
The perforated plate type shear connector was use- quality.
ful for providing horizontal shear resistance between 15 specimens were produced by reflecting 4 vari-
corrugated plate and concrete. The corrugated steel- ables (use of shear connector, section shape, span
concrete composite deck reinforced with I-beam length, and change of connecting method) and flexural
among corrugated steel-concrete composite decks test was performed to examine the effect of adhesion
enhanced the workability and reduced the weight com- and material connecting method on effective flexural
pared with existing RC decks by burying I-beam rigidity. Moment of inertia was obtained based on the
inside the corrugated steel-concrete deck. Currently load-deflection relationship obtained through experi-
the calculation of the effective flexural rigidity of gen- ments, and it was compared with formula suggested
eral RC structures applies the methods recommended by Standard Specifications for Highway Bridges and
by the Standard Specifications for Highway Bridges ACI (ACI 318R-02, 2002) to examine changes.
and ACI. And In Eurocode 4, the bending resistance of a com-
For the effective flexural rigidity of structures, posite beam with a partial shear connection can also be
Branson suggested formula (1) by approaching it conservatively specifiyed by an interpolation method.
through elastic analysis of stress and strain in 1965 In order to investigation the degree of shear connection
(Branson D.E., 1965), and Porter suggested the aver- of composite deck used Eurocode 4
age of moment of inertia and the moment of inertia on
non-cracking section as the effective moment of iner-
tia.(Poter M.L. et al., 1976) Also, Lamport suggested REFERENCES
his effective moment of inertia that provides better
fitting values by adjusting constants and exponents ACI Committee 318. 2002. Building Code Requirements for
appropriately using statistical processing of value structural Concrete(ACI 318R-02). American Concrete
suggested by Branson.(Lamport W.B., 1990) Institute, Farmington Hills, Michigan.
This study applied the method of calculating effec- Branson, D. E. Instantaneous and Time-Dependent Deflec-
tive flexural rigidity recommended by the Standard tions of Simple and Continuous Reinforced Concrete
Beams. HPR Rep. No. 7, Part 1, Alabama Highway
Specifications for Highway Bridges and ACI to the
Department. Bureau of Public Roads, Montgomery, Ala.
steel-concrete composite deck reinforced with I-beam Eurocode 4. 1994. Design of composite steel and concrete
to evaluate its applicability and made comparisons structure. Part 1 : General rules for building.
with reinforced concrete deck. Total 15 specimens Kim, T. H. 2006. Horizontal shear resistance of steel-concrete
were produced by considering 4 variables; the exis- composite bridge decks with various shear span lengths.
tence of stud, changes between spans, the shape of the Suwon: Sungkyunkwan University.
457
SS14: Using technology to manage, preserve, &
renew landmark signature bridges
Organizers: D.S. Lowdermilk & F.L. Moon
Load capacity estimation for the Burlington Bristol Bridge

ABSTRACT instrumented and monitored with a series of both high

speed and vibrating wire strain gages to capture the
The Burlington-Bristol Bridge spans 3000 ft across variation of dead load strain due to temperature, radi-
the Delaware River and has a main lift span com- ation, and other seasonal effects. The second phase
posed of a 540 ft through-truss. In 2007, this bridge included detailed ambient vibration monitoring study
was load rated using both analytical methods and tra- and a subsequent physics-based structural identifi-
ditional truck load tests, and the results were found cation to estimate total mass, mass distribution and
to be inconclusive. Following the load tests, there boundary/continuity conditions. The third and final
were several models that matched the live load strains phase included the use of a portable x-ray diffraction
equally well and predicted significantly different rating technology to estimate the intrinsic stress at various
factors. The key uncertainty was related to the distri- locations around the critical upper chord member.
bution of dead load actions, which represented over This paper and will present the details of this study
95% of the demand on the critical main span mem- and discuss the proposal of how uncertainty can be
bers and was completely unobservable during tradition reduced using this technology while the presentation
truck load tests. In an attempt to estimate dead load will discuss the completion of these techniques on the
demand, a three phased approach was developed and Burlington Bristol Bridge and the resulting effect on
carried out. First, the critical top chord member was the load capacity estimation.
461
Design and implementation of load cell bearings to measure dead

and live load effects in an aged long span bridge

Drexel University, Philadelphia, Pennsylvania, United States
ABSTRACT To validate the selected design, a prototype was devel-

oped and tested in the laboratory under various axial
In the spring of 2010, the expansion bearings of the forces, unidirectional and bidirectional moments, and
Burlington-Bristol Bridge were replaced due to a poor shear forces. Once validated, 14 smart bearings
condition rating (that caused the bridge to be clas- were fabricated and installed on the Burlington-Bristol
sified as structurally deficient). The replacement of Bridge. To verify their proper operation, a series of
these bearings represented a unique opportunity to load tests were carried out following installation. This
configure the new bearings to monitor the dead load paper and presentation will detail the development
and live load actions as well as their variation with and validation activities for the smart bearings as
environmental conditions. Towards that end, a series well as the results from the first few months of long-
of trial designs were developed with various bearing term monitoring. In addition, non-technical challenges
types, load cell configurations, etc. These candidate associated with developing appropriate design and
designs were then evaluated through a series of finite performance specifications for the bearings will be
element analyses and a single design was selected. discussed.
462
Structural health monitoring of signature bridges:

An engineers perspective
D. Lowdermilk, L. Egan & J. Prader

Pennoni Associates Inc., One Drexel Plaza, Philadelphia, PA, USA
ABSTRACT
The text in this paper is for visual purpose only. No

rights can be taken from this.
The aim of this paper is to detail the circumstances
under which an infrastructure owner facing diffi-
cult decisions related to bridge safety and repair for
a movable bridge on a major highway link made
the judgment that a best practices application of struc-
tural identification (St-Id) was warranted. Although
the integrated use of experimental and simulation
technology cost the owner more than the traditional
engineering inspection and load rating approaches
(which were completed in parallel), from the owners
perspective the potential benefits associated with such
an approach clearly outweighed the guaranteed costs.
Figure 1. Oceanic Bridge Bascule Span.
The decision hinged on the reduced uncertainty that
St-Id offered, which the owner viewed as critical to
making rational decisions related to posting, rehabili-
are costly considering the life span of this bridge at 5
tation, and ultimately the replacement of the bridge in
to 7 years.
question.
The work on this project has shown that utiliz-
The Oceanic Bridge spanning the Navesink River
ing Structural Health Identification can, by reducing
in Monmouth County New Jersey was constructed
uncertainty give a more clear picture of current behav-
in 1939 and rehabilitated in 1970. The structure con-
ior of the bridge in question. Through the instrumen-
sists of 57 spans with a 33 meter double leaf bascule
tation and modeling intrinsic in the identification, we
span over the navigable channel (Prader et al. 2009).
can get a better understanding of load paths and criti-
The current bridge is under consideration for replace-
cal members, and can, thereby develop a rehabilitation
ment due to the significant deterioration of the bridge
and preservation plan that more closely matches the
superstructure and piers. Currently, the bridge (due to
needs of the bridge. In the case of the Oceanic Bridge,
concerns with the bascule span) is posted for 3 tons,
we were analyzing a structure that must stay in ser-
which severely disrupts the travel of school buses,
vice for 5 to 7 more years. By performing targeted
emergency vehicles, and commerce in the surround-
retrofits, we can be more certain that we are placing
ing area. Due to the length of the project delivery
this bridge, and the County in a position of strength to
process for the proposed bridge, the existing bridge
nurse this geriatric structure through the remainder of
would ideally need to serve another 5 to 7 years at a
its life cycle.
higher posting with only absolutely necessary rehabil-
itation of the structure. The County is also planning
to perform retrofit construction on corroded bridge
members and a full grid deck system replacement for REFERENCE
which the construction cost is estimated at over $3M.
Another possibility the County is considering includes Prader, Egan and Lowdermilk, 2009. Recommendations
full replacement of the bascule span which carries a Report for Oceanic Bridge. Monmouth County, New
Jersey, USA.
construction cost of over $8M. Both of these options
463
Health monitoring of the Tacony-Palmyra bridge bascule span
J. Weidner
ABSTRACT
The Tacony-Palmyra Bridge (Figure 1) is a steel struc-

ture which spans the Delaware River between New
Jersey and Pennsylvania. The bridge was constructed
in 1929 by Modjeski, Masters and Chase as a replace-
ment for the inadequate Tacony-Palmyra Ferry which
operated at the same location. The design of the bridge
includes viaduct approach spans, six continuous truss
spans, a 550 arch span and a 260 double leaf rolling
bascule span. On average the bridge, which is owned
by the Burlington County Bridge Commission, sees
approximately 50,000 vehicles a day. The bridge is
tolled leaving New Jersey, with two lanes of traffic in Figure 1. Tacony-Palmyra Bridge.
that direction and one lane leaving Pennsylvania.
The Tacony-Palmyra Bridge is one of several major 4. Characterization of the types of live load that
assets under the care of the Burlington County Bridge the bridge is subject to by way of correlation of
Commission. They are responsible for the Burlington measured responses with images
Bristol Bridge, a vertical lift bridge, which also crosses 5. Tracking of transient forces like the axial force
the Delaware River and the Riverside-Delanco Bridge, at the leaf tips resulting from the bridge operator
a swing bridge on Rancocas Creek. In addition to one preload the system during closings
of each of the three most popular types of movable 6. Characterization of long-term effects due to envi-
bridges, the Commission also owns several smaller ronmental changes
fixed bridges throughout the county. All of the Com- 7. Immediate feedback in relation to safety issues like
missions assets are maintained with toll revenue on icing of the roadway surface
the Burlington Bristol and Tacony-Palmyra Bridges.
Despite the potential for delays to bridge openings, These deliverables will aid the Commission in both
these structures still experience a substantial traffic the day to day operation of the structure, as well as
load, partially because of lower toll rates, and par- in the development of a long-term plan for indefinite
tially because of the ties to the community which the preservation of this crucial, historic structure.
Commission works to maintain.
This paper aims to discuss the methodology
employed to develop the full instrumentation plan for REFERENCES
the TPB Bascule span monitoring. This includes dis-
Modjeski, et al (1931) Tacony-Palmyra Bridge over the
cussing the relationships between the owner, engineer Delaware River: Final Report. Final Engineers Report
and researcher, the goals of the monitoring project, the Submission to the Tacony-Palmyra Bridge Company,
functionality of the movable span of the bridge and the November 1931.
final instrumentation design. Koglin, T. 2003 Movable Bridge Engineering, Hoboken, NJ:
The proposed system will provide the following: John Wiley & Sons Inc.
1. Continuous check on the balance of the bascule
span
2. Immediate feedback to the bridge operator on the
quality of every bascule span opening
3. A better understanding of the load carrying mech-
anisms of the span; in particular the distribution of
load between the dead and live load trusses
464
SS15: Modeling of bridge seismic response
Organizer: M. Fischinger
Application and evaluation of a performance-based methodology

for the seismic assessment of multi-span simply supported deck
bridges
D. Cardone, G. Perrone & S. Sofia

DiSGG University of Basilicata, Potenza, Italy
L. Pardi
Autostrade per lItalia S.p.A., Rome, Italy
ABSTRACT In this paper, the proposed procedure has been

applied to a set of nine multi-span simply supported
Nonlinear Static Methods (NSM) based on pushover deck bridges of the Italian A16 Napoli-Canosa high-
analysis represent a good alternative to Nonlinear way, characterized by different types of piers (i.e.
Time-History Analysis (NTHA) and they are fre- single shaft, simple frame and single wall) and differ-
quently used as a practical engineering tool for the ent types of bearing devices (i.e. neoprene pads, steel
seismic assessment of existing structures. This paper hinges, sliding/roller bearings, pendulum systems).
focuses on the verification of a new performance- The IACSM predictions have been compared to
based adaptive methodology for the seismic assess- the results of NTHA, carried out using a set of 7
ment of bridges. accelerograms, compatible with the response spectrum
The proposed method, called IACSM, is based on provided by Eurocode 8 for soil type B, scaled to the
an inverse (I), adaptive (A) application of the Capacity PGA values provided by IACSM for each DS.
Spectrum Method (CSM). The capacity curve of the The comparison has been made in terms of maxi-
bridge is derived from a Displacement-based Adap- mum deformed shape of the deck and maximum pier
tive Pushover (DAP) analysis of the structure, carried displacements.Three indexes have been also computed
out separately in the transverse and longitudinal direc- to measure the precision of IACSM in capturing the
tion of the bridge. The bridge is modeled according to exact maximum deformed shape of the deck, maxi-
the principles of the so-called Structural Components mum pier displacement profiles and seismic demand
Modeling (SCM) approach, in which, the structure is to the critical element (pier, bearing device, abutment
divided in a number of independent rigid diaphragms, or joint) of the bridge. The target value of the three
modeling the bridge decks, mutually connected by indexes is always 1.
means of a series of nonlinear springs, modeling Results show that IASCM is able to predict with
bearing devices, piers, joint and abutments. good accuracy the actual response of the bridge
A number of Performance Levels (PLs), for which derived from NTHA, even when the bridge response is
vulnerability and seismic risk shall be evaluated, are strongly nonlinear. Moreover the critical members pre-
defined. Each PL is associated to a number of Dam- dicted by IACSM coincide with those resulting from
age States (DSs) of the critical members of the bridge NTHA. In approximately 75% of the totality of DSs
(piers, abutments, bearing devices, joints), identified considered, the three indexes result between 0.9 and
by a number of points on the DAP curve of the bridge. 1.1, both in the longitudinal and transverse direction.
The IACSM provides the earthquake intensity level Results clearly indicate the good accuracy of IACSM
(PGA) corresponding to the attainment of the selected in predicting the PGA values associated to slight-to-
DSs, using high-damping elastic response spectra as severe damage states, for bridges representative of the
Demand Curves. Italian bridge inventory.
The seismic vulnerability of the bridge is described The results of this study prove that the proposed
by means of fragility curves associated to the PGA val- methodology can be confidently used by the network
ues previously derived. Finally, the seismic risk is eval- managers for screening and prioritization of a wide
uated as convolution integral of the product between bridge inventory, to take responsibly decisions con-
the seismic vulnerability of the bridge (expressed by cerning possible repair or seismic retrofit measures.
the fragility curve) and the seismic hazard of the bridge
site (expressed by a proper hazard curve).
467
Probabilistic seismic response and retrofit assessment of aging bridges
J. Ghosh & J.E. Padgett

Department of Civil and Environmental Engineering, Rice University, Houston, Texas
ABSTRACT Table 1. Improvements in median values of system fragility

after the bridge is retrofitted with CFRP jackets at t = 25
Extreme events such as earthquakes continue to years.
threaten the security of bridge infrastructure. How-
ever, current approaches to seismic retrofit address Years 25 50 75
% Increase in median values 15.4 21.2 28.0
the enhanced seismic protection of bridges without
explicit consideration of other threats such as aging
and deterioration of the structure. Presently, more
than 26% of the nations bridges are estimated to
be structurally deficient and it is evident that aging
bridges are more vulnerable to natural hazards, such as
earthquakes. Hence, it is important to identify retrofit
strategies that simultaneously target both aging, such
as corrosion protection, as well as reduction of seismic
vulnerability of aging reinforced concrete members.
This paper presents a probabilistic performance
evaluation of aging bridges under seismic loading
using bridge fragility curves, which are statements of
the conditional probability of bridge failure. Along
the service life of the bridge, mechanisms such as
the loss of cross sectional area of the reinforcement
bars and degradation of steel bridge bearings due to
corrosion leads to significantly deteriorated seismic
performance. In the present study, fragility curves
are developed for typical degraded highway bridges
Figure 1. System level fragility curves at different points in
by incorporating probabilistic time dependent corro- time for the retrofitted and unretrofitted bridge.
sion models for key bridge components, specifically
reinforced concrete columns. Other effects of cor- REFERENCES
rosion deterioration like concrete spalling are not
considered in the present model due to their lack of Teng, J. G., Chen, J. F., Smith, S. T., and Lam, L. (2002).
influence on bridge fragility. Performance of carbon FRP-strengthened RC structures. John Wiley and Sons.
fiber-reinforced polymer (CFRP) column wraps as a Nielson, B. G., and DesRoches, R. (2007a). Analytical Seis-
potential retrofit measure for both mitigating corro- mic Fragility Curves for Typical Bridges in the Central and
sion deterioration and improving seismic performance Southeastern United States. Earthquake Spectra, 23(3),
is presented and analytically assessed for its influence 615633.
on bridge fragility. The effectiveness of conducting
this retrofit at different points in time along the service
life of the bridge is evaluated through comparison of
the fragility curves for the deteriorated and retrofitted
bridge.
468
Seismic response of repaired bridges by pseudodynamic tests
D. Lavorato & C. Nuti

DIS, University of Roma Tre, Rome, Italy
ABSTRACT Albanesi, T. & Lavorato, D. & Nuti C. & Santini, S. 2008.

Pseudo-dynamic tests on repaired and retrofitted bridge.
Eight bridge circular pier specimens (1:6 model of Proceedings of The 14thWorld Conference on Earthquake
the most stressed piers) severely damaged after previ- Engineering, Beijing, China, 1217 October 2008.
Albanesi, T. & Lavorato, D. & Nuti C. & Santini, S.
ous pseudodynamic and cyclic tests until failure, were
2009a. Experimental program for pseudodynamic tests
repaired and/or upgraded by means of epoxy adhesive, on repaired and retrofitted bridge piers. European Jour-
stainless steel rebars, self compacting concrete and C- nal of Environmental and Civil Engineering, vol. 13
FRP wrapping and tested by pseudodynamic, under No. 6/2009: 671683. Paris: Lavoisier.
the seismic action used for first tests on undamaged D.P.C. (Italian Civil Protection), ReLUIS. 2009. Guidelines
specimen. The undamaged pier specimens were repre- for Retrofitting and Repairing of Structural Elements,
sentative of tall and squat circular r.c. piers designed Partitions, non Structural Walls, Draft august (in italian).
according to Eurocode 8 or Italian Code before 1986. De Sortis,A. & Nuti, C. 1996. Seismic response by pseudody-
The damaged piers have been retrofitted by remov- namic tests of RC bridges designed to EC8. Proceedings
of 11th World Conference on Earthquake Engineering,
ing the damaged cover at the base substituting longitu-
Acapulco, Mexico.
dinal rebars with inox bars making the new cover with De Sortis, A. & Monti, G. & Nuti, C. 1994. Problemi di
SCC concrete and wrapping the pier with C-FRP to scala nella sperimentazione pseudodinamica di pile da
increase shear strength and ductility. ponte in c.a. Proceedings of Workshop on Danneggia-
In the pseudodynamic test, the repaired and/or mento, Prove Cicliche e Pseudodynamic, Dip. Analisi e
retrofitted piers specimens (one column per bridge) Prog. Str., Universit Federico II, Napoli, Italy, 23 june
are physically tested in the lab (Department of Struc- 1994.
tures, DIS, University of Roma Tre) while the rest of Eurocode 8. 1998. Design of structures for earthquake
the bridge is numerically simulated. After pseudody- resistance, Part 2: Bridges. (Draft march 2005).
Eurocode 8. 1998. Design of structures of earthquake resis-
namic tests on the bridge the single piers are brought
tance, Part 3: Assessment and retrofitting of buildings
to failure by cyclic tests. Damage states after the previ- (Draft november 2004).
ous tests are assessed and description of the repairing Hosotani, M. & Kawashima, K. 1999. A Stress-Strain Model
and retrofitting of the piers as well as the description of for Concrete Cylinders Confined by both Carbon Fiber
the In-House pseudodynamic apparatus and dedicated Sheets and Tie Reinforcement. Journal of Concrete Engi-
software implemented using LabVIEW and MATLAB neering, JSCE, 620/V43: 2542.
are given and finally a comparison between the seis- Italian National Seismic Code. D.M.LL.PP. 24.01.1986 .Min-
mic behavior of the original and the retrofitted bridge istry of Infrastructures. 1986. (in italian).
piers are shown in terms of seismic response. Tests Italian National Seismic Code. D.M. 14.01.2008. Ministry of
Infrastructures. 2008. (in italian).
results are compared to nonlinear analyses using the
McKenna, F.& Fenves, G.L.& Filippou, F.C. 2002. OpenSees:
computer program OpenSees. Results are discussed. Open System for Earthquake Engieering Simulation.
Future development will be the inclusion in pseudo- Pacific Earthquake Engineering Research Center. USA.
dynamic test of the soil-structure interaction and deck (http://opensees.berkeley.edu).
torsional stiffness, while inox rebars of new type will Shing, P. B. & Vannan, M. T.& Cater, E. 1991. Implicit
be considered. time integration for pseudodynamic tests. Earthquake
Engineering & Structural Dynamics 20, Issue 6: 551576.
REFERENCES
Albanesi, T. & Lavorato, D. & Nuti, C. 2006. Prove sper-
imentali monotone e cicliche su barre di acciaio inox.
Proceedings of National Conference Sperimentazione su
materiali e strutture: 357366. Venezia. Italy.
469
Response of curved steel bridges to seismic loading
D.G. Linzell & J. Seo

The Pennsylvania State University, University Park, PA USA
V.P. Nadakuditi
Stress Offshore, Inc., Houston, TX, USA
INRODUCTION
Given the lack of research related to curved, steel,

bridge seismic response and the continued increase
in curved bridge design and construction, there con-
tinues to be a need to investigate seismic behavior. Two
projects that studied curved, steel, plate girder bridge
seismic response are summarized. The first examined
the influence of certain parameters on forces and reac-
tions generated in representative curved bridges. The
second studied seismic response of a group of curved
bridges to generate fragility curves.
Findings for the first study indicated that, for
simply-supported curved plate girder bridges, vary-
ing girder radius of curvature had a slightly more
pronounced effect on reactions reported at the bridge
bearings for the interior (lowest radius) girder (Fig-
ure 1), especially under vertical excitation, than vary- Figure 1. Effect of radius of curvature on maximum vertical
ing cross frame spacing. Variations in cross frame reactions under vertical (red) and horizontal (blue) excitation.
spacing had marginal effect on maximum loads sus-
tained in a select cross frame member during a seismic
event. Vertical seismic excitation was shown to have
more influence than horizontal excitations for the cases
that are presented.
The second study detailed methods used to gen-
erate seismic fragility curves for a select group of
horizontally curved steel bridges using RSMs. Pro-
cedures used to: (1) select and screen important input
and output parameters, generate finite element mod-
els and synthetic ground motions used to create the
RSMs and; (2) apply the RSMs to efficiently generate
select fragility curves for a group of 99 curved steel
bridges were presented. One representative fragility
curve, related to the likelihood of bearing damage from
tangential deformations as a function of increasing Figure 2. Representative Seismic Fragility Curve, Tangen-
PGA, was shown (Figure 2). tial Bearing Displacement.
470
The design of r.c. bridge deck subjected to horizontal actions

by strut-and-tie models
M. Mezzina, F. Palmisano & D. Raffaele

Technical University of Bari, Bari, Italy
ABSTRACT designer but also it can be extremely time consuming.

This is the reason why many procedures (e.g., Load
Strut-and-Tie Model (STM), conceived by the French Path Method, optimization criteria), that aim at find-
builder Franois Hennebique as a simple representa- ing the most accurate solution with the minimum
tion of a reinforced concrete element subjected to shear effort, have been proposed in the last decades.
and bending, has been later developed by Ritter and In this paper, taking as example the application of
Mrsch. In the twentieth century several studies have the Strut-and-Tie Method to a typical r.c. bridge deck
been carried out about STM and the results constitute subjected to horizontal transversal action, the need of a
the fundamentals of some prescriptions in many inter- step-by-step procedure that goes from the comparison
national codes. Schlaich et al. (1987) proposed a global with pre-solved similar cases to the use of the Load
approach to the structural design by means of STM. Path Method (Schlaich et al. 1987, Palmisano et al.
The Strut-and-Tie method implies that the structure 2002) and optimization criteria (e.g. Bi-directional
is designed according to the lower bound theorem of Evolutionary Structural Optimization method, Yang
plasticity (Schlaich et al. 1987). Since concrete per- et al. 1999) has been highlighted.
mits only limited plastic deformations, the STM has Moreover, for the case under study, the influence on
to be chosen in a way that the deformation limit (i.e. the model of both the geometry of the bridge slab and
the capacity of rotation) is not exceeded at any point the distribution of longitudinal and transversal deck
within the structure before the assumed state of stress beams has been investigated.
is reached in the rest of the structure. This ductility
requirement is fulfilled by adapting each element of
the model to both the direction and size of the internal REFERENCES
forces as they would come from the theory of elasticity
(Schlaich et al. 1987). Palmisano, F., Vitone, A., Vitone, C. 2002. Form & Struc-
ture. The Rome Auditorium: load path method (LPM).
It is often not necessary a deep knowledge of the
DArchitettura 18: 168173.
Strut-and-Tie method to find truss models that best fit Schlaich, J., Schafer, K., Jennewein M. 1987. Toward A Con-
the regions under study. This is also due to the fact sistent Design of Structural Concrete. PCI Journal 32 (3):
that often it is possible to adapt well known pre-solved 74150.
examples to the analysed case. Yang, X.Y., Xie, Y.M., Steven, G.P., Querin, O.M. 1999. Bidi-
In non-standard cases the development of the opti- rectional evolutionary method for stiffness optimization.
mum truss model can require not only an expert AIAA Journal 37 (11): 14831488.
471
Hybrid seismic isolation design of Sakarya-II Viaduct in the proximity

of the North Anatolian Fault
C. Ozkaya, N. Celebi & F. Tulumtas

Yuksel Proje Uluslararasi Co. Ankara, Turkey
M. Dicleli
Engineering Sciences Department, Middle East Technical University, Ankara, Turkey
ABSTRACT
In this paper, the seismic isolation (SI) design of the

Sakarya-II viaduct located within 300 meters of the
North Anatolian fault in Turkey is studied. Sakarya-II
is a 385-m long, 9 span, slab-on-steel-girder viaduct.
A view of the bridge is presented in Figure 1.
Strong near-field characteristics such as directiv- Figure 1. View of the bridge.
ity and liquefaction are expected due to soft soil
conditions under the scenario earthquake having a
magnitude of Mw = 7.5.
Preliminary analyses indicated that using a con-
ventional SI-system results in very large isolator dis-
placements and forces due to strong directivity effects
expected at the bridge site. Consequently, a hybrid
SI-system is chosen for the bridge to minimize the
magnitude of the isolator displacements and forces.
The hybrid SI-system consists of four natural and two
lead-rubber bearings supporting the six steel-girders
at each substructure.
Lead cores in the lead-rubber bearings are designed
to have an overall characteristic strength equal to 10%
of the superstructure weight. This large characteristic Figure 2. Lead rubber bearing after dynamic prototype test.
strength is intentionally chosen to limit the isolator dis-
placements under near-field effects. Furthermore, the
supplementary four natural rubber bearings over each and test velocities of the bearings. Lead rubber bear-
substructure provided additional stiffness to achieve a ings have 1100 mm and 1400 mm diameter. Lateral
small post elastic period to deflect the earthquake input design displacement of the bearings is 850 mm.
energy transferred through the underlying soft soil Test velocity is calculated as 2.2 m/sec. Static pro-
and to limit the isolator displacements under near-field totype and dynamic prototype tests were performed.
effects. Static tests were performed at bearing manufac-
Nonlinear time history analyses of the viaduct with turer Alga Spa, Milano laboratory. Dynamic proto-
hybrid SI system have revealed that the hybrid SI- type tests were performed at EUCentre laboratory,
system reduced the sensitivity of the bridge response Pavia, Italy. Very high demands in dynamic proto-
to the characteristics of the near-fault earthquake and type tests resulted in peculiar problems in the test
liquefaction. Superstructure with steel girders resulted equipment. Tested bearings were not damaged dur-
in lower earthquake forces due to its lower weight ing dynamic prototype tests although a very stringent
compared to a concrete superstructure. test procedure was applied. View of a lead rubber
Prototype testing of lead rubber bearings utilized bearing after dynamic prototype test is presented in
in the viaduct is a challenging task due to dimensions Figure 2.
472
Seismic abutment-deck interaction of a four-span R/C bridge model
M. Saiidi
Department of Civil and Environmental Engineering, University of Nevada, Reno, Nevada, USA
A. Ebrahimpour
Department of Civil and Environmental Engineering, Idaho State University, Pocatello, Idaho, USA
N.B. Johnson
Nevada Department of Transportation, Carson City, Nevada, USA
ABSTRACT
A quarter-scale, four-span, 33.5 m (110 ft) long asym-

metric conventional reinforced concrete bridge model
was tested using the shake table system at the Univer-
sity of Nevada, Reno. The bridge received biaxial hor-
izontal motions at the bases of the three two-column
bents from three separate shake tables. In addition,
the bridge was simultaneously subjected to longitudi-
nal motions from actuators attached to the abutment
seats.The ground motions, based on a 1994 Northridge
earthquake record, were applied in seven test runs
with increasing amplitudes. This paper focuses on the
Figure 1. OpenSees model of the four-span bridge: (a) over-
nonlinear computer modeling of the bridge includ- all model, and (b) North abutment.
ing the interaction between the bridge deck and the
abutments, the in-plane rotation of the bridge, and the response history analysis of the bridge that included an
corresponding transverse bent displacements. abutment spring representing the force-displacement
A three-dimensional finite element model of the relationship of a typical seat-type abutment.
bridge was developed in OpenSees (references are The abutment-deck interaction was modeled by a
given in the main paper). A view of the computer set of two contact elements at each abutment. Further-
model is shown in Figure 1(a). Appropriate masses more, realistic abutment gap openings were included
were assigned to the bridge deck nodes as shown by the that were based on the start and the end of each test
solid circles in Figure 1(a). The elevated nodes above run. Several sensitivity analyses were conducted. The
the bridge deck correspond to the masses of the addi- results of the friction sensitivity analysis gave high
tional blocks that were placed on the bridge. The bridge variability in the calculated transverse residual dis-
columns were modeled with the OpenSees nonlinear- placements. In some cases, a small perturbation in
BeamColumn element with a fiber-discretized cross one of the contact element friction coefficients pro-
section composed of Concrete01 and Hysteretic uni- duced substantially different results. However, using
axial materials for concrete and steel, respectively. the average of the residual displacement square root
Column bond-slip moment-rotation behavior was rep- of the sum of the squares (SRSS) within each of the
resented by zeroLength elements with Hysteretic uni- seven cases considered, it appears that imposing fric-
axial material placed at the column ends. The super- tion at the NE or the SW or a combination of the NE
structure was assumed to stay within the linear-elastic and SW produced transverse bent displacements that
range during the seven experimental tests; thus, elas- were closer to the experimental results. More abutment
ticBeamColumn elements were used for all the bridge contact with the NE and the SW corners are consistent
deck. The biaxial earthquake motions were imposed with the observed deformed shape of the bridge deck.
to the column bases as shown by the double-headed Given the variability of the predicted responses,
arrows in Figure 1. The longitudinal actuator motions obtaining a set of optimal friction coefficients is dif-
were imposed on both the North and the South abut- ficult. Through a large number of simulations, it may
ments. These motions were obtained from a pre-test be possible to obtain such an optimum set.
473
Influence of shear key modeling on the performance of bridges

under simulated seismic loads
P.F. Silva & T.L.T. Nguyen

The George Washington University
ABSTRACT that can represent the spatial distribution of the shear

keys along the abutment length. Results have shown
Exterior shear keys are commonly used at the abut- that using multiple elements the bridge superstruc-
ments of small to medium span bridges to provide ture develops rotational restrains at the abutments,
transverse support for the bridge superstructure under which are not properly replicated when using a sin-
small earthquake. When subjected to large magnitude gle element. This restraining of the superstructure at
earthquakes, these shear keys have been assumed in the abutments can be shown analytically to decrease
design and analysis of bridges to perform as structural substantially the displacement ductility demand on the
fuses. Previous experimental research results clearly supporting columns. Furthermore and in conjunction
show that shear keys at the abutments do not perform with the modelling of the shear key gaps, this paper
as structural fuse. In fact and according to current presents results from a series of finite element analysis
reinforcing details, shear keys possess significant post that were conducted using the open source finite ele-
damage residual capacity. Analytical results show that ment program OpenSees. This paper provides detailed
the residual capacity of shear key can have a consider- results from multiple modelling cases that consider
able impact on the seismic response of medium span the shear keys hysteretic response, spatial distribution
bridges. modelling and gapping at the abutments. These analy-
Another modelling aspect of these shear keys, is the ses are presented and discussed in further detail in the
use of a single finite element or a series of elements paper.
474
A case study of analysis techniques for precast segmental bridges subjected

to vertical seismic excitation
M.J. Veletzos
Merrimack College, North Andover, MA, USA
J.I. Restrepo
University of California at San Diego, La Jolla, CA, USA
ABSTRACT segmental bridges for the California Department of

Transportation.
Precast segmental construction of bridges can accel- The results indicated that variations in the damp-
erate construction and minimize the cost of medium ing value of different modes due to Rayleigh damping
span bridges in environmentally sensitive and dif- caused the difference between elastic time history and
ficult to access locations. Consequently, the use of modal analyses. The Rayleigh damping for vertical
precast segmental construction is on the rise world- analyses should be based on the dominant vertical
wide. Numerous analyses techniques exist to estimate modes, not the dominant longitudinal mode and 10Hz.
seismic demands on precast segmental bridges. These It is recommended that Rayleigh damping be defined
techniques range from a simple linear elastic response at frequencies where (i) the cumulative vertical modal
spectra analysis to a complex nonlinear time history mass exceeds 20% and (ii) the cumulative vertical
analysis using truss elements to model the segment modal mass exceeds 80%.
joint behavior. Each technique has benefits and draw- The results from the linear elastic time history were
backs and every bridge may have a different set of consistent with the results from the nonlinear inelastic
performance requirements, thus it can be difficult to model. Both models indicate that the superstructure
determine the most appropriate modeling technique to segments will exceed the decompression limit state
accurately and efficiently assess the seismic demands. at select segments, primarily the midspan segments
This paper presents a case study of analysis and under negative bending and only due to the 2500 year
modeling techniques to determine the most appropri- hazard. The nonlinear elastic model predicted seg-
ate modeling method to estimate the vertical seismic ment joint rotations that were up to five times larger
demands on precast segmental bridges. Three sepa- than the rotations predicted by the nonlinear inelastic
rate two-dimensional analytical models were devel- model. This is because the NEM used a bi-linear curve
oped for a 91 m (300 foot) span bridge. The first that assumed that the segment joint was already fully
model simulated the superstructure using only lin- cracked.
ear elastic members. The second model simulated the Based on the analyses performed and the results pre-
superstructure using simple non-linear elastic lumped sented in this paper it is recommended that a modal
plasticity members. The third model utilized numer- analysis using a linear elastic model should be used
ous axial only gap-hook elements to explicitly model for the functional evaluation of Ordinary bridges.
the behavior of concrete and post-tensioning tendons For the safety evaluation of Ordinary bridges, it is
at critical superstructure segment joints. These mod- recommended that designers satisfy the no collapse
els were subjected to five different earthquake hazard criteria by checking the capacity of all vertical collapse
levels. The results were compared to determine the mechanisms relative to the vertical design spectrum.
most appropriate modeling method for Ordinary and For both the functional and safety evaluation of
Important segmental bridges and were used in the Important bridges, a time history analysis using a
development of seismic design guidelines for precast nonlinear elastic model is recommended.
475
Compressive stress-strain model for high-strength concrete confined

with spirals
J.C. West & R.A. Hindi

Bradley University, Peoria, Illinois, USA
ABSTRACT different trend. While the equation for a single-spiral

column using high-strength concrete has already been
The design of high-strength concrete columns under determined, the cross-spiral column does not follow
seismic loading is not only based on the maximum the same equation, and this paper offers an equation
strength of the column but also the ductility of the that predicts the cross-spiral column strength. In either
column. A combination of the axial stress and strain the case of high-strength concrete or normal-strength
of the column can be represented as the total amount concrete, the base equations, such as the core effective-
of energy that the column can absorb before failure. ness and the confinement pressure, need to be modified
An increase in the strength of the concrete core or to encompass the cross-spiral configuration in order to
increase in maximum strain before failure will increase obtain the final equation.
the amount of energy that the specimen can absorb. Several conclusions can be obtained from the anal-
For the design of circular column, the method to ysis of the 21 high-strength concrete columns along
improve the amount of energy the column can absorb with the 10 normal-strength columns.
is based on the spacing of the transverse reinforce-
Columns using normal-strength concrete using
ment used. This paper presents a stress-strain model
single and cross-spiral column configuration can
to predict the axial behavior of high strength concrete
be described by the same equation that Mander,
confined with both single spiral and two opposing
Priestly and Park (1988) predicted.
spirals. Data from twenty-one of high-strength con-
Columns using high-strength concrete follow a dif-
crete columns along with ten normal-strength concrete
ferent trend when dealing with single-spiral verses
columns is presented and used to derive the model. To
cross-spiral configuration.
have an understanding of the correlation between the
The confinement pressure can be predicted for
amount of transverse reinforcement and the strength of
cross-spiral columns with the same method Man-
the concrete core, the range from the minimum spac-
der, Priestly and Park (1988) used for hoops or
ing to the maximum spacing as specified by ACI-318
single-spirals in which the average percentage error
was considered. Also, the effect of the amount of lon-
from the test samples was 4.68 for normal-strength
gitudinal reinforcement was considered in deriving the
and 3.28 for high-strength concrete.
model.
The method Mander, Priestly and Park (1988) used
The data from the columns that were constructed by
to predict the core effective can be modified to
Marvel (2008) along with Al-Qattawi (2004) in their
include cross-spiral columns as well.
research was used to derive equations to find the fcc
for a column using cross-spiral configuration.
In order to find the full trend of cross-spiral rein-
forcement, both normal-strength concrete and high- REFERENCES
strength concrete are used to predict the strength
increase caused by the spiral reinforcement. In the case ACI Committee 318, Building Code Requirements for
with normal-strength concrete, the increase is derived Structural Concrete (ACI 318-05) and Commentary
(318R-05), American Concrete Institute, Farmington
from previous equations that are based on the concrete
Hills, Mich., 2005, 430 pp.
being able to fully engage the steel reinforcement at Mander, J.B., M.J.N. Priestly, and R.Park. Theoretical stress-
maximum loading. With high-strength concrete being strain model for confined concrete, Journal of Structural
more brittle, the maximum strength of the concrete is Engineering, ASCE, 114, No. 8, 1988, pp.18041825.
sooner than when the spirals have been fully engaged
which causes the high-strength concrete to follow a
476
SS16: Recent challenging bridge structures
Organizer: I.S. Darwish
Assessing and extending the service life of bridges
C.A. Ligozio, S.T. Wyatt & A.T. Ciolko

CTLGroup, Skokie, IL, USA
ABSTRACT Effective service life analysis integrates inspec-

tion and testing methods to assess existing condition
Traditional bridge management practice is reactive to and identify causes of degradation observed along
the needs of specific bridges. When routine inspection with quantitative analysis of anticipated structural vul-
identifies that the condition of structures has degraded nerabilities and a statistical evaluation of the likely
sufficiently, a more detailed inspection and study is progression of deterioration. Often the analysis will
performed to develop a repair approach to address consider a rehabilitated condition.
the observed deficiencies. This approach is gener- Over the last decade, the authors have developed
ally insufficient to assess the remaining service life and implemented this approach for evaluating service
of a structure and provide the information necessary life of high-value bridge assets. Three case studies
to develop cost effective rehabilitation plans and/or involving the condition evaluation of concrete sub-
evaluate the feasibility of reusing components of an structures will be described and presented, to illustrate
existing structure. the tools and methodologies available for evaluation
Service life-based evaluation techniques histori- of structures, to determine existing conditions, assess
cally adapted by the authors for concrete bridge ele- remaining service life, and to develop appropriate
ments, adopt life cycle cost analyses together with and timely rehabilitation schemes. These case studies
durability modeling. This approach provides an effec- are gathered from significant structures with condi-
tive tool for the development of rehabilitation planning tions typical nationwide, including the northern cli-
and can also reduce the overall life cycle cost of mate, deicer-affected and marine climate-influenced
a structure by preventing the onset of more severe environments.
degradation mechanisms.
479
The role of finite element analysis in bridge assessment and design
T. Cakebread
LUSAS Vice President North America
ABSTRACT lower lateral bracing system that vibrated excessively

under truck crossings and required remedial action
Finite element (FE) analysis is being used more and is mentioned along with mention of solutions for the
more these days for bridge engineering because of Milford-Montague bridge retrofit.
the more economical and accurate assessments and The role that FE plays in creating innovative new
designs its use produces. designs is illustrated by the Gateshead Millennium
This paper describes the basic development of FE Bridge in the UK, and the initial proposed New Missis-
analysis and illustrates the role that it can play in just sippi River crossing by Modjeski & Masters These two
some areas of bridge analysis, assessment and design boundary-pushing structures show the real benefits of
with reference to a number of real-life FE bridge case using FE analysis.
studies carried out by US and UK consultants. When a re-design is undertaken using FE analysis
FE analysis allows for a more rigorous design or there can be occasions when great savings can be made
analysis approach to be adopted which is often sig- and the Estero Parkway Flyover in Fort Myers, Florida,
nificantly more accurate and economical than codified is cited as an example of this.
methods. To illustrate this mention is made of a study For erection engineering two cable stayed foot-
by Catbas & Gokee into the use of AASHTO dis- bridges, the Wichita Riverfront Footbridges and the
tribution factors which produced results found to be Bob Kerrey Pedestrian Footbridge show how fabrica-
typically 2540% over-conservative when compared tion geometry can be checked and cable tensioning can
to FE solutions.A similar load capacity study byAtkins be optimized to speed-up construction.
for a pair of braced steel beams showed that a nonlinear
FE analysis can provide a significantly larger collapse
load factor than that obtained from a design code. REFERENCES
When the diaphragms of a steel box bridge do not
comply with assessment code criteria, FE analysis will Catbas, N & Gokee B. 2009. A Novel Approach to Analyse
Existing Bridges Efficiently. ASCE Structures Congress
allow a detailed analysis to be performed in order to
Hambly, E.C. 1976. Bridge Deck Behaviour. London: E &
help prove the integrity of the design. Raith Bridge and F.N. Spon.
the Midlands Link Viaducts, both in the UK are given Hendy, C.R. 2008. The implications of the change to
as examples. Eurocodes for bridge design. Proceedings of the ICE
Two, US steel truss bridges the US-2 over Cut Bridge Engineering Vol 1. Issue 1.
River, and the M-55 over Pine River, and a typical UK OBrien, E.J. & Keogh, D.L. 1999. Bridge Deck Analysis.
steel through bridge at Hackbridge (which was anal- London: E & F.N. Spon.
ysed for a proposed increase in vehicle axle weight), LUSAS. 19972009. Finite element analysis case studies.
are given as examples of how load capacity can be See http://www.lusas.com/case/bridge/index.html
Murphy, T.P & Irwin M.C. 2006. Retrofitting the Milford-
proved on structures even if the latter was seen to fail
Montague Truss: Challenges and Solutions. 5th National
according to a design code. Seismic Conference.
Bridge assessments requiring retrofit solutions are Zhou, E & Biegalski A.M. 2008. Problem Diagnosis and
another area where FE analysis can assist greatly with Retrofit of Lateral Bracing System of a Truss Bridge.
what-if scenarios. A two span deck truss bridge with a Structures Congress, Canada.
480
Challenges in design, rehabilitation and construction of bridge structures
M. Kasi, H. Abdou & D. Morrill

Alfred Benesch & Company, Chicago, Illinois, United States
ABSTRACT destroyed a Metra rail bridge impacting the daily com-

mute of thousands of riders. Temporary and permanent
Engineering challenges may appear in the form of situ- solutions were needed immediately. Metra had never
ations or constraints that require engineers to respond before faced a service disruption of this magnitude.
in a creative way to them. They offer opportunities They quickly responded by assembling a team to assess
to engineers to demonstrate their technical skills and existing conditions, develop a design-build plan and
creativity. execute the reconstruction of this critical commuter
This paper refers to four case studies in design, line. Planning and design of a temporary structure
rehabilitation and construction where innovative started immediately and proceeded concurrently with
approaches were used to solve engineering chal- the removal of the remains of the double track bridge.
lenges. In each case, engineers analyzed the situations, The team designed and rebuilt the structure in eight
explored various possibilities and selected appropriate days.
solutions. The third case study is the staged deck replace-
The key to addressing any challenge is to under- ment of a tied arch bridge where the solution was
stand the issues. TQE is Alfred Benesch & Com- carefully implemented in stages. The two-way bridge
panys unique approach for understanding and address- was supported by a single arch structure. Part-width
ing these challenges. Challenges to design, con- construction caused an asymmetrical loading condi-
struction and rehabilitation require consideration of tion. The challenge to the designer and the contractor
constructability, integrity of existing structures dur- was to ensure that the behavior using the Finite Ele-
ing rehabilitation, cost and schedule. Constructability ment Model was replicated in the field and that the
includes construction staging, construction method, appropriate measures during the different stages of
field conditions and maintenance of traffic. construction were taken in order to produce the desired
The first case study is the shortening of a truss deck profile. Some of these measures included sur-
which had never been tried before. The design chal- veying the top of the ties before and after each
lenge was to widen the roadway within the limits of stage and making sure that the elevations are within
the truss bridge. This project evolved due to the need to an acceptable tolerance of the elevations shown on
improve an interchange at the north end of the truss. the construction drawings. The project was com-
The interchange was built in the late 1950s and had pleted successfully without changes to the design
numerous design deviations from current standards, procedure.
including several sub-standard safety features. Several The fourth case study details the construction of a
places within the interchange had accident rates that curved bridge where local materials were utilized to
were more than ten times higher than statewide aver- reduce the construction cost. The girders had to be
ages. Thus, the interchange needed improvement, but held with temporary supports to restrict its vertical
the proximity of the truss posed a significant site con- and lateral movements. Available beam sections were
straint. Shortening the truss was proposed as a solution stacked over to provide the needed section modules
to rebuild the end with a wider cross-section. Short- and concrete blocks to stabilize the support system.
ening part of the anchor arm has a major impact to The temporary structures were designed, fabricated,
the cantilever and suspended span. Benesch engineers constructed and the decks poured within three months.
devised a load-transfer system to cut the truss in a All deflections and lateral movements were well within
safe way. the allowed amounts.
The second case study describes the design and
construction of a bridge in eight days. A major fire
481
Innovative design approach to a GFRP pedestrian bridge: Structural aspects,

engineering optimization and maintenance
S. Russo
Iuav University of Venice, Italy
A. Adilardi
Engineer of Pratos Municipality, Italy
ABSTRACT
This paper deals with a pedestrian bridge made with

pultruded GFRP (Glass Fiber Reinforced Polymer)
profiles. Here are highlighted solutions from a struc-
tural point of view and from a maintenance standpoint.
The footbridge is now under construction and it is
entirely designed with GFRP Profiles.
Design criteria of short-span bridges made with
pultruded profiles have been analyzed focusing on
the most relevant challenges, especially those related
to global instability and dynamic performance, and
finally comparing with similar items for steel bridges.
The relatively low modulus of glass fibre FRPs also
Figure 1. Mode 1.
requires that buckling characteristics be assessed a
little more critically than would be the case with a Table 1. Mechanical characteristics of GFRP material.
steel component, which generally have an abundance
of stiffness relative to their strength. Items Properties
The high strength, high fatigue resistance,
Density 1800 kg/m3
lightweight, and corrosion resistance of composites Long. Tensile strength 200500 MPa
are highly desirable characteristics for bridge applica- Modulus elasticity (tensile) 20.00030.000 MPa
tions and the maintenance program in the design of the
bridge demands only a biennial inspection.
At present the construction of such a bridge cer- is linked to the very low weight (for structural employ-
tainly is a fascinating matter in that, despite several ment it become from 1800 to 2000 kg/m3 and 60% of
studies been carried out, the actual application has fibers) but it could be considered as a weakness aspect
proven difficult. One of the reasons is the innovation as to the risk of the resonance phenomenon. Besides,
being related to the use of new materials; another one is a possibility due to the very light and durable struc-
linked to a likewise pioneer construction management. tural material characterized also by very high value
Furthermore from a point of view of public con- of strength (but also with not negligible deformability
struction the use of a non-conventional material level), represents anyhow an extraordinary opportu-
has to deal with a lack of normative and of a well nity in term of innovative design approach.
consolidated price history.
A detailed numerical analysis of the dynamic struc- REFERENCES
tural answer of the All-GFR pedestrian bridge strictly
related to the low weight has been carried out, and the Adilardi A., Frasconi, L. Design of a pedestrian bridge
evolution of construction design of lightweight struc- made with pultruded profiles of fibreglass-reinforced plas-
tures as pedestrian footbridges has been pointed out. tics in Prato Proc., Footbridge 2008, 2nd-4th July 2008,
In the all-GFRP constructions field it is necessary University of Porto, Portugal.
to pay attention particularly to the dynamic perfor- Russo, S. (2007). Strutture in Composito: Sperimentazione,
Teoria e Applicazioni. Edited by Ulrico Hoepli, Milano,
mance expected in function also to the evaluation of Italy.
the fundamental frequency value, both for traditional Bastianini, F., Boscato, G., Russo, S., Sciarretta, F. (2007).
frame structures, made by beam-column connection, Natural frequencies of pultruded profiles with different
and vehicular or pedestrian bridge. In fact, the major cross-section. Proc., Advanced Composites in Construc-
benefit due to the construction made by GFRP material tion, ACIC 07, University of Bath, Bath, UK, 433440.
482
SS17: Life-cycle design of structural systems
Organizers: F. Biondini & D.M. Frangopol
Structural health monitoring role in bridge life cycle analysis
S. Alampalli
New York State Department of Transportation, Albany, NY, USA
M. Ettouney
Weidlinger Associates, Inc., New York, NY, USA
ABSTRACT
The primary objective of asset/bridge management is

ensuring safety of users while maintaining or reduc-
ing costs of ownership. Decision making tools are
often used by bridge owners to assist in reaching these
objectives and one such tool is life cycle analysis.
In particular, the Bridge Life Cycle Cost Analysis
(BLCCA) technique is extensively advocated in the
infrastructure arena. But decision making capabili-
ties can be enhanced by an accurate estimation of
the benefits and life span of the structure. Thus,
Bridge Life Cycle Analysis (BLCA) is a metric that Figure 1. SHM Role in BLCA and Decision Making.
requires consideration on all three components and
includes BLCCA, BLCBA (Bridge Life Cycle Benefit exploring two modes where SHM can provide advan-
Analysis) and BLSA (Bridge Life Span Analysis). tage to the BLCA process. Following these modes can
Since accuracy of BLCA depends mainly on the help decision makers in achieving their safety and cost
quality of information and assumptions that are used, optimizations goals. (Ettouney and Alampalli 2010).
it is clear that SHM can play an essential role in all
of the BLCA components (see Figure 1). One major
issue in this type of analysis is lack of reliable data that REFERENCES
is needed for accurate results. Structural Health Moni-
toring (SHM) is one of the tools that have the potential Ettouney, M., and Alampalli, S. 2010. Infrastructure Health
to assist in BLCA. The paper will discuss this by in Civil Engineering, CRC Press (in preparation).
485
Lifetime seismic performance of concrete bridges
F. Biondini & E. Camnasio

A. Palermo
University of Canterbury, Christchurch, New Zealand
ABSTRACT 1992). Moreover, the material deterioration has differ-

ent influence on the bridge systems analyzed. Bridges
The seismic performance of concrete bridge structures for which the ductility demand tends to be equally dis-
under material degradation induced by the diffusive tributed along the piers are less affected by the effects
attack of aggressive agents over the structural life- of material deterioration. Therefore, in order to mit-
time is investigated. In a traditional seismic ductile igate the effects of material deterioration on bridge
design philosophy, the inelastic response of bridge systems, a suitable configuration of bridge pier pro-
occurs in the bridge substructure with development files has to be achieved. In particular, change of section
of plastic hinge regions at the bottom and/or top of the profile or reinforcement along bridge pier should be
piers (Paulay & Priestley, 1992). As a consequence, implemented in order to equally distribute the ductil-
if aggressive agents act on bridge piers, the struc- ity along bridge piers. Finally, the results show that
tural behavior varies during lifetime and hence the the behavior factor, assumed as constant value in the
seismic performance is generally reduced in terms of seismic design codes (CEN-EN 1998-1, 2004), varies
both ductility and strength (Biondini et al., 2004). The over time depending on the environmental exposure of
interaction between the diffusion process and the cor- the structure. Further investigations are then needed to
responding mechanical damage can be complicated properly identify the structural bridge typologies more
if different geometrical parameters, as distribution of vulnerable to lifetime material degradation effects.
pier heights and deck stiffness, are considered. In
fact, these aspects can considerably affect the overall
seismic performance of bridge systems. REFERENCES
In this paper geometrically regular and irregular
continuous bridge structures with piers having box Biondini, F., Bontempi, F., Frangopol, D.M., Malerba,
P.G., 2004. Cellular Automata Approach to Durability
cross-section are analyzed. The time-variant struc-
Analysis of Concrete Structures in Aggressive Environ-
tural performance of the critical cross-sections of the ments. Journal of Structural Engineering,ASCE, 130(11):
piers is evaluated in terms of bending moment vs. 17241737.
curvature relationships. Time-variant pushover anal- Biondini, F., Palermo, A., Toniolo, G., 2010. Lifetime seis-
yses are then carried out over the structural lifetime mic performance of precast concrete structures. Jour-
to assess the global performance of the system in nal of Structure and Infrastructure Engineering, Special
terms of base shear force and displacement ductility Issue of 1st International Symposium on Life-Cycle Civil
(Biondini et al., 2010). Seismic strength is com- Engineering (paper accepted for publication).
pared with design actions corresponding to prescribed CEB, 1992. Durable Concrete Structures Design Guide.
Bulletin 183.
peak ground accelerations, and seismic indicators,
CEN-EN 1998-1, 2004. Eurocode 8: Design of Structures
such as overstrength and global behavior factor, are for Earthquake Resistance. Part 1: General rules, seismic
investigated over the structural lifetime. actions and rules for buildings. European Committee for
The results confirm that the lifetime seismic perfor- Standardization, Brussels.
mance of concrete bridges is affected by the diffusive Paulay, T., Priestley, M.J.N., 1992. Seismic Design of Rein-
attack of aggressive agents that induce the deteriora- forced Concrete and Masonry Buildings. John Wiley &
tion of concrete and corrosion of reinforcement (CEB, Sons.
486
Structural robustness and redundancy of deteriorating concrete bridges
F. Biondini
D.M. Frangopol
ABSTRACT are investigated in this paper by taking uncertain-

ties into account. The effects of the damage process
In structural design the terms robustness and redun- are evaluated by using a proper methodology for
dancy are often used as synonymous. However, they life-cycle assessment of concrete structures exposed
denote different properties of the structural system. In to diffusive attacks from environmental aggressive
fact, structural robustness can be viewed as the ability agents (Biondini et al. 2004). Time-variant measures
of the system to suffer an amount of damage not dis- of structural robustness and structural redundancy are
proportionate with respect to the causes of the damage developed at the global level with respect to the loads
itself. Structural redundancy can instead be defined as associated to the first local failure and to the structural
the ability of the system to redistribute among its mem- collapse (Biondini and Frangopol 2008).
bers the load which can no longer be sustained by some The proposed approach is illustrated through the
other damaged members after the occurrence of a local application to the assessment of time-variant structural
failure. robustness and structural redundancy of a concrete
Moreover, structural robustness and redundancy are arch bridge. The results demonstrate that robustness
usually investigated with respect to damage suddenly and redundancy are different structural properties,
provoked by accidental actions and abnormal loads. which may exhibit opposite trends over time, both in
However, damage could also arise gradually in time deterministic and probabilistic terms, depending on
from aging of structures. Depending on the damage the damage scenario.
propagation mechanism, such kind of damage may
also involve disproportionate effects and alternate load REFERENCES
redistribution paths. These effects are particularly rel-
evant for bridge structures due to their environmental Biondini, F., Bontempi, F., Frangopol, D.M., Malerba,
exposure. Notable events of bridge collapses due to P.G., 2004. Cellular Automata Approach to Durability
the environmental aggressiveness and related phe- Analysis of Concrete Structures in Aggressive Environ-
nomena, such as corrosion and fatigue, include for ments, Journal of Structural Engineering,ASCE, 130(11),
example the Silver Bridge in 1967, and the Mianus 17241737.
River Bridge in 1994. For these reasons, structural Biondini, F., Frangopol, D.M., 2008. Probabilistic Limit
Analysis and Lifetime Prediction of Concrete Struc-
robustness and redundancy should be considered as
tures, Structure and Infrastructure Engineering, 4(5),
key factors for a rational approach to life-cycle design 399412.
of deteriorating structure and infrastructure systems. Biondini, F., Restelli, S., 2008. Damage Propagation and
It is therefore of great interest to investigate the rela- Structural Robustness, First International Symposium
tionship between robustness and redundancy and its on Life-Cycle Civil Engineering (IALCCE08), Varenna,
evolution in time under a progressive deterioration of Italy, June 1014. In Life-Cycle Civil Engineering,
the structural performance. F. Biondini and D.M. Frangopol (Eds.), CRC Press.
Structural systems under progressive damage have Biondini, F., 2009. A Measure of Lifetime Structural Robust-
been investigated to identify suitable measures for ness, ASCE-SEI Structures Congress 2009, Austin, TX,
USA, April 29May 3.
structural redundancy (Frangopol and Curley 1987)
Frangopol, D.M., Curley, J.P., 1987. Effects of Damage
and structural robustness (Biondini and Restelli 2008). and Redundancy on Structural Reliability, Journal of
More recently, the time factor has been explicitly Structural Engineering, ASCE, 113(7), 15331549.
included in a lifetime scale for a time-variant measure Okasha, N.M., Frangopol, D.M., 2010. Time-variant redun-
of structural robustness (Biondini 2009) and structural dancy of structural systems, Structure and Infrastructure
redundancy (Okasha and Frangopol 2010). Engineering, 6, 279301.
Based on these approaches, the time-variant robust-
ness and redundancy of deteriorating concrete bridges
487
Life-cycle analysis of a new composite material for bridge pavement

waterproofing
M. Crispino & F. Giustozzi

Politecnico di Milano, Milano, Italy
G. Flintsch
Virginia Tech, Blacksburg, USA
ABSTRACT
One of the most critical issues related to bridge man-

agement and preventive maintenance is deck water-
proofing.
Infiltration of meteoric and rain-wash water into
the bridge asphalt pavement through unsealed cracks
could produce a series of chain events that leads
towards an accelerated decay of the bridge struc-
tural adequacy. Actions like crack sealing and general
waterproofing are very important; they can in fact slow
down the deterioration and reduce the life-cycle cost.
An effective protection system against water is
one of the key factors for a functional and efficient
bridge management plan. However, benefits associ-
ated with such maintenance activities are seldom Figure 1. Savings due to preventive maintenance with
correctly accounted and, in many instances, they are waterproofing membrane (2 membranes/cycle).
not even considered at all. Consequently many road
bridge authorities do not include preventive mainte- Costs were discounted and summed in order to obtain
nance activities in their management infrastructures the Life Cycle Cost for each alternative; results are
programs. Recent research has resulted in the develop- expressed through a Saving Factor (Fig. 1). The
ment of new technologies in the field of crack sealing result, derived from both economical analysis and
and pavement waterproofing. In this context, a new practical experience, showed significant benefits due to
composite material for waterproofing interlayer mem- the adoption, on bridge deck pavement, of preventive
branes, to be cold applied (with great benefit for maintenance treatment with waterproofing membrane,
environment and workers) was studied and also tested allowing a remarkable overall saving.
in full scale experiments by Politecnico of Milano
researchers. This paper presents the main characteris- REFERENCES
tics and performances of this new technology, applied
to bridge pavements. FHWA Federal Highway Administration, 1998. Life Cycle
The Authors carried out a Life Cycle Cost Anal- Cost Analysis in Pavement Design, US Department of
ysis in order to estimate the long term benefits of Transport.
Preventive Maintenance strategies (PM ), making Hall K. & al., 2003. Guidelines for Life-Cycle Cost Analy-
use of a waterproofing interlayer membrane vs a Do- sis of Pavement Rehabilitation Strategies, Transportation
Research Board.
Nothing alternative (DN ). The analysis was devel-
Biondini F, Frangopol D., 2006. Probabilistic Service Life
oped considering a sample road bridge (with unitary Assessment and Maintenance Planning of Concrete Struc-
length), with a transversal section constituted by two tures, Journal of Structural Engineering.
traffic lanes for each direction. Agency and User costs Crispino, M. 2008. Improving Pavement Waterproofing in
were defined considering reconstruction, preventive Italy, 23rd ARRB Conference Research Partnering with
maintenance operations and residual service life value. Practitioners, Adelaide.
488
The fatigue limit states of the AASHTO LRFD Bridge Design

Specifications
D.R. Mertz
University of Delaware, Newark, Delaware, USA
ABSTRACT Table 1. LRFD Specifications fatigue live-load load factors.
Recently, a second fatigue limit state was included limit-state load live-load load
in the AASHTO LRFD Bridge Design Specifica- combination factor
tions (AASHTO, 2010). This limit state, termed the
Fatigue I 1.50
Fatigue I limit state, was added to explicitly dif-
Fatigue II 0.75
ferentiate infinite-life fatigue design from finite-life
fatigue design. The original fatigue limit state is now
termed the Fatigue II limit state and provides finite-life
designs. The live-load load factors of the two fatigue- The Fatigue II live-load load factor of 0.75, from
and-fracture limit-state load combinations excerpted Table 1, times the fatigue load yields an effective
from LRFD Table 3.4.1-1 are given in Table1 below. truck which replicates fatigue damage of the entire
The load factors of the fatigue-and-fracture limit- distribution of truck traffic.
state load combinations represent multipliers which The additional fatigue limit state has no impact
transform the fatigue load of LRFD Article 3.6.1.4, on the design of steel and concrete highway bridges.
the HL-93 design truck with a fixed rear-axle spac- The original fatigue resistance equation with its lower
ing of 30 feet, into nominal fatigue loads for use in bound has been replaced with two separate fatigue
the two fatigue-and-fracture limit-state load combina- resistances equations. The real impact is on the percep-
tions. They do not represent uncertainty of the nominal tion of bridge designers. Now when they are providing
fatigue loads. an infinite-life design, they explicitly know it.
The Fatigue I load factor of 1.50, from Table 1, when
multiplied by the stress range due to the passage of
the LRFD fatigue truck produces the maximum stress REFERENCES
range considered for fatigue design. This maximum
stress range for fatigue design has been characterized AASHTO, LRFD Bridge Design Specifications, Fifth Edi-
as the stress range with a 1 in 10,000 probability of tion, American Association of State Highway and Trans-
being exceeded. portation Officials, Washington, DC, 2010.
489
Reliability assessment of reinforced concrete beams rehabilitated

with CFRP sheets
C.M. Paliga
Federal University of Pelotas, Pelotas, Brazil
M.V. Real
Federal University of Rio Grande, Rio Grande, Brazil
A. Campos Filho
Federal University of Rio Grande do Sul, Porto Alegre, Brazil
S.M.C. Diniz
Federal University of Minas Gerais, Belo Horizonte, Brazil
ABSTRACT Table 1. Probability of failure and reliability index.
In a time of greater demands posed by a deteriorated Undamaged Damaged Retrofitted

infrastructure in both developed and developing coun-
r Pf Pf Pf
tries, it is clear that the design of retrofit/ strengthening
systems for bridges is of utmost importance. Nowa-
3 2 106 4.61 6.9 103 2.46 1.0 106 4.75
days, the utilization of externally bonded carbon fiber- 1 1.1 104 3.70 1.8 102 2.10 7.5 105 3.79
reinforced polymers (CFRP) sheets has found a large 1/3 1.2 103 3.04 3.9 102 1.76 1.1 103 3.07
acceptance in the rehabilitation of reinforced concrete
(RC) structures. This fact is easily understood since
been calculated through a simplified procedure. Two
these materials present a number of favorable charac-
different procedures are used in the reliability analysis.
teristics such as lightweight, relatively easy to install,
Reliability indexes (and corresponding failure
noncorrosive, and high tensile strength that make them
probabilities) are obtained for undamaged, damaged,
suitable for use as retrofit/ strengthening systems.
and retrofitted beams (Table 1). These results are com-
Accordingly, guidelines for the design and construc-
pared and discussed in the light of: (i) the reliability
tion of externally bonded FRP systems for strengthen-
procedure used in the analysis, (ii) the adequacy of
ing concrete structures have been proposed, e.g. FIB
the design scheme used in the repair of the beam,
Bulletin 14 (2001), ACI 440.2R (2008), and JSCE
and (iii) the safety evaluation of existing versus new
23 (1997). Design codes for new structures have
structures. It is shown that reliability levels can be
been calibrated via probabilistic methods; similarly,
significantly overestimated depending on the selected
design recommendations for the rehabilitation of exist-
reliability procedure.
ing structures shall be based on such methods. The
implicit reliability levels attained must be assessed; REFERENCES
additionally, these levels shall satisfy a prescribed
target reliability index. ACI Committee 440, 2008. Guide for the design and con-
In this study, the reliability assessment of RC struction of externally bonded FRP systems for strength-
ening concrete structures. American Concrete Institute.
beams, retrofitted by externally bonded CFRP sheets Fderation Internationale du Bton, 2001. Externally bonded
is performed. The flexural retrofit has been designed FRP reinforcement for RC Structures. Bulletin 14, Lau-
according to the ACI 440.2R (2008) guidelines. For sanne, Switzerland.
comparative purposes, the resulting reliability levels Japan Society of Civil Engineers, 1997. Recommendation
for undamaged and damaged beams are also evaluated. for design and construction of concrete structures using
An important task in the reliability analysis process continuous fiber reinforcing materials. JSCE, Concrete
is the realistic description of the beam resistance. It is engineering series 23.
well known that the debonding of CFRP sheets plays Paliga, C. M., 2008. Anlise probabilstica de vigas de con-
an important role in the performance of retrofitted creto armado recuperadas flexo, atravs do mtodo de
Monte Carlo utilizando um modelo de elementos finitos (in
beams. As such, a nonlinear finite element model, Portuguese). Porto Alegre: PPGEC/UFRGS, PhD Thesis.
capable to deal with the debonding process and cou- Paliga, C.M., Campos Filho, A., Real, M.V., 2007. Finite ele-
pled to a Monte Carlo simulation procedure, is used ment model for numerical analysis of strengthened rein-
in the determination of the statistics of the beam forced concrete structures. IBRACON Structural Journal.
failure loads (Paliga 2008). Failure loads have also v.3, p. 177200.
490
Time varying risk modeling of deteriorating bridge infrastructure

for sustainable infrastructure design
A.S. Rao, M.D. Lepech, A.S. Kiremidjian

Stanford University, Stanford, California USA
X.Y. Sun
Zhejiang University, Hangzhou, Zhejiang, P.R. China
Visiting Scholar 20092011, Stanford University, Stanford, California USA
ABSTRACT
Time varying risk assessment models represent a new

approach for characterizing risk due to multi-hazard
environments, such as extreme seismic events and
long-term structural deterioration, that vary indepen-
dently over time. The proposed method quantifies the
impact of time-dependent deterioration of reinforced
concrete elements on seismic risk for a single infras-
tructure element over time through shifting structural
fragility curves. A more comprehensive, probabilistic
reinforcing steel deterioration model is presented that Figure 1. Column failure fragility with uniform corrosion,
accounts for the effects of uncracked and cracked con- (Concrete cover = 51 mm).
crete covers through the inclusion of pitting corrosion
in cracked concrete. When coupled with a probabilis-
tic seismic response model for reinforced concrete
structures, the result is a comprehensive assessment
tool for the structural fragility of infrastructure sys-
tems exposed to multi-hazard environments over long
periods of time.
The effects of including pitting corrosion increase
the probability of failure of RC elements throughout
their service life, most notably after 25 years. At a
moderate age for bridge structures (50 years) the prob-
ability of failure of an RC bridge column during a
seismic event with Sa = 0.8 g increases by 2.4 times Figure 2. Column failure fragility with pitting corrosion
as a result of accounting for pitting corrosion over (Concrete cover = 51 mm).
previous models as seen comparing Figures 1 and 2.
Such results may point toward an existing systemic
underestimation of the fragility of infrastructure stock
potentially exposed to chloride-induced corrosion in
seismic areas.
Results also show that designers can have a signifi-
cant effect on reducing the multi-hazard risk profile
investigated in this study by altering basic design
parameters such as concrete cover. As seen compar-
ing Figures 2 and 3, the increase of cover thickness
from 51 mm to 127 mm virtually negated the effects of
chloride-induced corrosion thereby preserving the ini-
tial structural fragility curve throughout the expected Figure 3. Column failure fragility with pitting corrosion
100-year service life. (Concrete cover = 127 mm).
491
Study of remaining fatigue life of Brazilian concrete bridges
J.F.S. Rodrigues & P.A.O. Almeida

Universidade de So Paulo, Escola Politcnica, So Paulo, Brazil
J.R. Casas
Universitat Politcnica da Catalunya, Barcelona, Spain
ABSTRACT
In this paper a parametric study on the fatigue life time

of short-span reinforced concrete roadway bridges is
performed. For this study a simulation model has been
developed, which includes the most significant sources
of uncertainty in the definition of traffic action on
bridges as well as of structural response.
From experimental data collected on dynamic tests,
curves for natural frequencies and dynamic amplifi-
cation factors were provided, (Fig. 1). Impact factors
obtained from the tests with higher speeds were found
larger than corresponding values recommended by
bridge codes. In addition to dynamic tests, concrete
samples were extracted from bridges under investiga-
tion. Modulus of elasticity and compressive strengths Figure 1. Experimental DAF-S parameter relationship.
of existing concrete were obtained from specimens
extracted from bridges. Dimensional checking of
bridges under investigation was also carried out. Val- To fatigue analysis, resistance and solicitation were
ues measured on sites were compared with dimensions coupled in the Limit State Function, and reliabil-
specified on bridge design, and respective differences ity indexes were derived using FORM (First Order
were registered. Reliability Method).
For the parametric study, a common structural type The results of the current investigation show that
of bridge built on Brazilian roadway net since 1950 fatigue reliability index can be lower than target
were adopted, a two girder-slab reinforced concrete reliability indexes, mainly in shorter spans bridges,
bridge, with 3 spans, ratio of 0.25 : 1 : 0.25, and free between 7 and 10 meters. Besides that, a substan-
edges on extremity spans. tial increase in the reliability index is observed with
To represent traffic action, real traffic data registered increasing reinforcement ratio. For new structures
on Brazilian roadways were considered. Natural uncer- design this fact suggests a border line of cross-section
tainties inherent to the definition of traffic action, as stiffness to be applied on these kind of structures on
type of vehicles, intervals between vehicles arrivals, the aim of reach 100 year lifetime required by codes.
its total weight and its velocity, were adopted as This border can be investigated with the procedures
random variables. developed in this study.
Monte Carlo simulations were performed with main Due to low values of fatigue reliability obtained in
spans of 7, 10, and 13 meters, within the range of this study, simulations to analyze Ultimate Limit State
investigated bridges, and reinforcement ratios of 1, 1.5 should be performed.
and 2%. Geometrical and mechanical characteristics
of bridges were considered as random variables based
on site survey. REFERENCES
A statistical definition of the S-N curves and the Crespo-Minguilln, C. (1996): A Reliability based methodo-
Miner summation at failure was adopted to deal with logy to define the limit state of decompression in
the uncertainty in the prediction of the number of prestressed concrete bridges. Ph.D. Thesis. Universitat
cycles to failure. Politcnica de Catalunya, Barcelona.
492
SS18: Optical monitoring techniques for
bridge maintenance & safety
Organizers: S. Sumitro & H. Matsuda
Real-life applications and experiences with fiber optic bridge

monitoring installations
K. Chandler
Chandler Monitoring Systems Inc., Lawrenceville, Georgia USA
T.W. Graver
Micron Optics Inc., Atlanta, Georgia, USA
A. Mndez
MCH Engineering LLC, Alameda, CA, USA
1 ABSTRACT on Raman or Brillouin scattering methods have been

developed along with pertinent interrogation elec-
We present a brief overview of real-life bridge struc- tronics and instruments, many of which are already
tural health monitoring (SHM) installations using fiber commercially available.
Bragg grating sensing technology. We review sev- The technical requirements imposed by many SHM
eral project installations and describe the associated applications combined with the increased demand for
successes, challenges and lessons learned for each practical, reliable, field-portable and low cost equip-
application. In general, project successes are coupled ment to perform on-line measurements, has led to
to improved sensing tools: better sensor packages, a new wave of commercial product and a push for
simpler and less expensive instrumentation, improved the development of new optical fiber sensor solutions.
installation techniques, and more efficient data anal- However, can fiber optic sensing address the applica-
ysis tools. Some shortcomings are the direct result or tion needs and provide an economical, effective and
poor project planning and communications. Particu- reliable bridge monitoring alternative? Owners must
lar attention is given to the benefits and economics of manage and ensure the safety of their bridgeseven
instrumenting civil structures when and how it pays. as their use might extend well beyond their original
Over the last few years, fiber optic sensors (FOS) design lifetime. Traditionally, most structures rely on
have seen increased acceptance and widespread use strict maintenance procedures, visual inspections, and
for industrial sensing and structural health monitoring very few sensors. Furthermore, maintenance opera-
(SHM) applications in composites, civil engineering, tions are expensive; visual inspections can miss critical
aerospace, marine, oil & gas, and defense. Given problems; and conventional sensors often fail in harsh
their EM immunity, intrinsic safety, small size & environments. Hence, how can this new technology be
weight, and capability to perform multi-point and satisfactorily proven in the field?
multi-parameter sensing remotely make of FOS an This and other questions are aimed to be answered
attractive, flexible, reliable and unique sensing solu- by way of reviewing a number of real-life bridge instal-
tion. One of the most common applications is for lation projects and their associated practical lessons
strain/stress sensing, but a variety of other param- learned, challenges and successes from each bridge
eters such as temperature, pressure, magnetic field, installation.
voltage, chemical species and others, can also be mea- In general, many successes are coupled to improved
sured using them. And, nowadays, a variety of rugged FOS tools: better sensor packages, simpler and less
sensors and interrogation instruments are commer- expensive instrumentation, improved installation tech-
cially availableoffering attractive performance, ease niques, as well as more efficient data analysis tools.
of installation, reliability and reasonable pricing. Some shortcomings are the direct result or poor project
A variety of discrete fiber strain sensors based planning and communications. Particular attention is
on Fabry-Perot (FP) cavities and fiber Bragg grat- given to the benefits and economics of instrumenting
ings (FBGs), as well as distributed techniques based civil structures when and how it pays.
495
Base study on measure of transformation of concrete structures

used by digital image correlation method
H. Goda & M. Hibino

Kyushu Institute of Technology, Kitakyushu, Japan
M. Uchino
Fukuoka Industrial Technology Center, Kitakyushu, Japan
K. Yamaguchi
Kyushu University, Fukuoka, Japan
H. Matsuda
Nagasaki University, Nagasaki, Japan
ABSTRACT
This study is the basic research for proposing a mainte-

nance method for the structures using full visual field
non-contact measurement technique. By using digi-
tal image correlation method, bending behavior of a
steel and concrete composite girder was measured. The
digital image correlation method is to take a surface
pattern before and after the deformation by a digital
camera and to obtain amount and direction of deforma-
tion of the specimen from the brightness distribution
of the digital image obtained. The specimens are steel-
concrete composite girders using the perfobond-rib Figure 1. Load-deflection curve.
shear connector (PBL: Perfo-Bond Leisten) as shear
connector. A scope of 1000 mm 600 mm was pho-
tographed by a digital camera. In order to reduce
variations in the image caused by factors other than
deformation, multiple images were overlaid. In addi-
tion, deformations in the vicinity of the target point to
be measured were averaged.
It is found that about the same level of accuracy as
that by contact type measurement was obtained by the
measurement of displacements such as deflection or
relative displacement. With regard to the strain mea-
surement, although it has larger variations than the
Figure 2. Strain distribution of composite girder by DICM.
measurement by gauge, its qualitative behaviors were

able to be evaluated.
REFERENCES
Uchino, M., 2001. Two-directional simultaneous measure-
ment system of the surface deformation using speckle
interferometry method. Proceedings of the APCFS &
ATEM01. 863839.
Uchino, M., 2006. Discussion of StrainAnalysis using Digital
Image Correlation Method. JSEM collaborated workshop.
Picture 1. Specimen and camera. No.6-1. 3639.
496
Development of approximation process of existing action stress

of pre-stressed concrete by stress relief technique
K. Hida
K&T Consultant Co. Ltd, Tokyo, Japan
Y. Ito
Saga University, Saga, Japan
M. Uchino
Fukuoka Industrial Technology Center, Fukuoka, Japan
T. Okamoto
Keisoku Research Consultant Co. Ltd, Hiroshima, Japan
P. Sumitoro
Smart Structures, Illinois, United States
H. Matsuda
1 INTRODUCTION
A concrete structure has high durability and it has been

believed that a maintenance is unnecessary. However,
in some concrete structures, by degradation of damage
from chloride attack, carbonation, an alkali aggregate
reaction, etc., it cracks and rebar corrosion etc. has
appeared notably in recent years.
Especially the pre-stress concrete structure (follow-
ing PC structure) built over varieties, such as a highway
bridge, a railroad bridge, and a building, has intro-
duced pre stress into the structure with PC tendon. PC
bridge in which it is located near the seashore shows
the example which PC tendon fractured by damage
from chloride attack to photo 1.1. Photo 1. PC Tendon fractured by chloride attack1) .
Degradation, such as carbonation, chloride attack
and an alkali aggregate reaction, reduces the dura-
bility ability of a concrete structure. A degradation REFERENCES
factor permeates into concrete and it is begun to cor-
rode rebar and PC tendon in concrete. If degradation Kenichi Hida, Non-destructive test (Rebar corrosion), Pre-
stress concrete, Vol.48, No.3, 2006.
advances further, by breaking PC tendon, pre-stress Kenichi Hida, Repair and Strengthening on A PC Bridge
will decrease and a load-proof performance will also Deteriorated Due to Chloride Attack in Coastal Environ-
decrease. ment, Proceedings of the First International Conference
Therefore, the suitable measure against PC structure of Asia Concrete Federation, Thailand, October, 2004,
is made by presuming and predicting reduction of pre- 11121121.
stress. And the safety of PC structure is intentionally The non-contacting Full-field Optical stress and strain Mea-
maintainable. surement committee of degradation / hardening process of
However, the technology that adequately measures concrete, The school about the new method of measur-
of pre-stress of the PC structure under the present sit- ing the infrastructure structure by the optical measuring
method, Japan, December, 2006, 137143.
uation has not been established. Therefore, authors
combined the stress relief technique and the full-field
measurement method and developed the measuring
method of the existing action stress.
497
Development of on-board image measurement system for actual

running and application to wall surface surveys of structures
H. Kurashige, S. Nishimura, K. Hara & J. Kato

Keisoku Research Consultant Co., Hiroshima, Japan
T. Omae & H. Yamasaki

Kankou Co., Osaka, Japan
ABSTRACT that the images acquired from the system were

clearer.
Many inspections of linear structures such as rail-road Centralized management of a huge amount of
tunnels and monorail track beams have been carried images using time axes
out by close visual inspections, but the surveys spend The authors synchronized more than one CCD
a lot of time, and weigh on the original repair sched- camera, and managed hundreds of thousands of
ule. As an alternative technique for these close visual high-resolution images acquired using time axes.
inspections, a method using video cameras has been We calculated the travel distance that changed
reported whereby video images of surfaces of struc- successively using the optical encoder installed to
tures are converted into still images to create developed the moving vehicle, had the images segmented
images while the vehicle is in motion. However, this automatically, and synthesized them.
technique inevitably causes image quality degradation. Developed image creation system (image correc-
In addition, it cannot synchronize several cameras, tion, segmentation, and connection)
and takes time for post-processing. In contrast, indus- The authors constructed a geometric correction
trial cameras used in imaging tests at manufacturing system to create developed images that maintain
lines have smaller sizes and higher precision and frame uniform quality from images having different pho-
rates, and can be connected directly to PCs. tographic resolutions, and created precise devel-
However, to apply these industrial cameras to oped images of curved surfaces and cross sections.
inspections of public structures, it is necessary to solve By changing the system configuration according
their adverse problems in photographing conditions to each subject, photographing keeping a designated
(Table 1). quality is possible.
This time, the authors revealed the characteristics of Extraction of cracks
the industrial cameras through laboratory experiments There is still room for improvement in post-
to solve these problems, and developed an on-board processing including crack extraction, hence we will
image measurement system for actual running that continue to develop new systems.
does not cause image quality degradation.
We verified the system on monorail track beams
and at road tunnels, and confirmed that it can acquire REFERENCES
clearer images than those acquired from conven- Study on evaluation techniques of soundness of road tunnels
tional video cameras, and report the results in the Study group for new urban society technological fusion
post-processing in about a half period of time. and creation, in April 2009. Research project on evalua-
tion techniques of soundness of road tunnels http://www.

Acquisition of high-definition images
kkr.mlit.go.jp/road/shintoshikenkyukai/02.html Research
The authors compared the images acquired from result 7 Research on evaluation techniques of soundness
this system with those acquired from conven- of road tunnels.
tional video cameras by verifying the system at For example, Japan Concrete Institute: Concrete diagnosis
monorail track beams and tunnels, and confirmed technique, 2002 Foundation, pp.89, 2002.
Table 1.
Outside light conditions Camera-to-subject distance Travel speed Number of equipment
Manufacturing Can be controlled to constant Almost constant Can be controlled at a Single camera
line conditions constant speed
Infrastructure Change between day and night, Distance changes depending Travel speed changes. More than one camera is used. Vibration-
survey and depending on places on cross sections. proof and small-size cameras are desirable.
498
Development of a remote laser scanning system for continuous

monitoring of cable-stayed bridges
T. Miyashita
Nagaoka University of Technology, Japan
N. Miyamoto & S. Umemoto

Keisoku Research Consultant Co., Japan
Y. Fujno
University of Tokyo, Japan
ABSTRACT measurement points and mechanical rotation of the

LDV together with the TS. Furthermore, by using
In Japan, continuous investments in infrastructures the rotating motor of the TS, the scanning angle of
have formed a huge stock. However, since the infras- the system becomes 90 degrees.
tructures were constructed rapidly and emphatically Measurement procedure using the developed sys-
in high economic growth period, the numbers of the tem follows below.
degraded infrastructures increase rapidly. Therefore,
the development of quantitative and efficient moni- 1) Reflection tapes or prisms for surveying are
toring techniques is strongly required (Fujino et al. attached on measurement points.
2007). 2) LDV laser beam is adjusted so that the reflection
This situation hold true for existing bridges. More- levels of the laser at each point attain the highest
over, according to geographical features of Japan, there level.
are many long-span bridges such as the Tatara bridge, 3) TS measures three-dimensional coordinates of
the Akashi Kaikyo bridge and so on. For these large measurement points and stores their information.
structures, a large amount of labor is required for 4) Measurement starts. TS rotates to a memorized
maintenance. The expensive cost of its inspection will point, and then LDV conducts vibration measure-
become a big burden in the future. ment automatically.
In order to reduce the cost, vibration-based struc- 5) When LDV measurement finishes, TS rotates to the
tural health monitoring is quite promising. In fact, next measurement point. 4) and 5) are repeated for
tensile forces of cables in long-span bridges have been all measurement points.
measured using vibrations. In vibration-based struc-
tural health monitoring, accelerometers have been In this study, at first, the developed system was
usually utilized. They are attached to selected mea- evaluated at selected bridges. Then, in order to exam-
surement positions of a structure and wired to other ine the practicality of the system, it was applied into
equipments such as amplifiers and PC. At long-span measuring vibration of the Tatara bridge. As a result,
bridges, measurement works involve the installation of the developed system is expected to apply continuous
sensors at high ground levels. Therefore, it is strongly vibration monitoring of the whole long-span bridges.
needed to reduce the dangerous works and improve By using this system, most of cables, girder, and pylon
their efficiency. in the long-span bridge can be monitored quickly,
In this research, to improve the safety for work of easily and safely.
vibration measurement in a long-span bridge, a non-
contact and remote laser scanning system, which is the
combination of a Laser Doppler vibrometer (LDV) and REFERENCES
a total station (TS), was developed. Fujino, Y. and Siringoringo, D. M., Bridge monitoring in
Although this system is simple because the LDV is Japan: the needs and strategies, Proceeding the Fourth
simply attached on the TS, it makes possible to conduct International Conference on Bridge Maintenance, Safety,
automated remote vibration measurement continu- and Management, 2008, pp. 4952 (Full paper is enclosed
ously due to internal storage of the coordinates of in CD-ROM).
499
3D profile measurement and buckling simulation on thin-walled

cylindrical shells under compression by utilizing 3D-digital
image correlation method
C. Morita, C. Zhao, A. Demizu, K. Makino & H. Matsuda

Structural Engineering, Nagasaki University, Japan
K. Ichimiya
Civil Engineering, Oita National College of Technology, Japan
ABSTRACT
Cylindrical shells are widely used in the most diverse

branches of civil engineering technologies because of
they combine light weight with high strength. How-
ever the buckling strength and post-buckling behav-
ior prediction are extremely complex especially for
thin-walled ones. Initial geometric imperfections are
normally known to have a strong detrimental on the
buckling strength and to be the reason of large dis-
crepancy between experimental buckling loads and
theoretical predictions. Moreover random imperfec-
tions in real shells are not known at the stage of design
and may take a complex form. In order to investigate
the effects of geometric imperfections on the buckling
strength and post-buckling phenomena of thin-walled
cylinders, these investigations were carried out both
experimentally and numerically. Figure 1. Load-displacement curve of specimen No.2.
In present paper, a general and simple approach
is proposed for modeling the geometric imperfec-
tion of thin-walled shell specimens by utilizing 3-D This approach for the 3d profile measurement of
digital image correlation method (DICM), an opti- thin-walled shell structure is an efficient and accurate
cal non-contact and full-field measurement. Realistic method and would be conduced to the maintenance
geometric imperfection of thin-walled shell specimens work of realistic thin-walled structures. By which, the
were obtained and applied in numerical simulation, realistic imperfection information can be employed
the thickness of specimens used in this experiment is to form the numerical mesh, therefore the simula-
0.131 mm and the ratio of external diameter (R) to tion results agree very well with the experimental
thickness (H ) R/H is 250. Procedures of scanning simulation. As to the realistic structures, combin-
on the geometric imperfection of thin-walled shell ing this dialog and the 3d profile measure approach,
specimen and its application in numerical simulation the real imperfection information of structures can
are described detailed in this paper. The experimental be obtained, thereafter by employing the numerical
investigation agrees well with the numerical simu- simulation; rationalization proposal can be obtained
lation on both the strength and buckling behaviors. for the maintenance work of realistic thin-walled
Therefore by using this effective approach the potential structures.
of strength capacity can be tapped in real thin-walled
shell structures and it is demonstrated the visualization
of buckling behaviors with initial random geometric REFERENCES
imperfections by using this digital image correlation
technique. Arbocz J & Hol JMAM. 1991. Collapse of axially compressed
cylindrical shells with random imperfections. AIAA J 29:
This investigation shows the sensitivity of buckling 224756.
behaviors of thin-walled shell on the geometric imper- Yamada S. 1998. Nonlinear Analytical Confirmation on the
fection clarify the quantized classification on the effect Application of Reduced Stiffness Method rof the Buckling
of geometric imperfection as follow figure. of Axially compressed Cylinders. Japan. JSME.
500
Strain analysis method using multi-rosette analysis by digital image

correlation method
M. Uchino
Fukuoka Industrial Technology Center, Fukuoka, Japan
T. Okamoto
Keisoku Research Consultant Co, Hiroshima, Japan
K. Hida
K&T Consultant Co, Tokyo, Japan
Y. Ito
Saga University, Saga, Japan
P. Sumitoro
Smart Structures, Rantoul, United State
H. Matsuda
ABSTRACT
A new analysis method of strain distribution using

the digital image correlation method (DICM) will be
developed in order to estimate stress concentration
and residual stress (Uchino, M. et al., 2009). We will
present the relationship between the stress or strain
concentration around the hole and the distance change
of the hole diameter. As shown in figure 1, for the
one-dimensional tension of an infinite thin plate with
a through hole, the change ratio of the hole diameter
along the x-axis equals the tangential strain at the bor- Figure 2. Results of tensile test of the aluminum plate with
der of the hole on the y-axis and then the change ratio a 2 mm diameter hole in the stress of 15 MPa.
of the hole diameter along the y-axis equals the tan-
gential strain at the border of the hole on the x-axis. around the hole is estimated by this distance change
The changes of the hole diameter and the ring geometry (drx : along x-axis, dry : along y-axis). For an application
around the hole were used for the new analysis method of this method, the one dimensional tension experi-
of stress concentration and residual stress. Specifically, ments in the aluminium plate with the center hole and
the distance change between two points that is sym- the stress release experiment in the concrete block by
metry with respect to a center point of the hole will be a core drilling method were performed to evaluate the
measured by using the DICM. The stress distribution proposed method. Here, the results of tensile test in
the case of the center hole diameter was 2 mm using
the multi rosette analysis at the tensile stress of about
15 MPa is shown in figure 2. The analysis result using
the finite element method (FEM) is also shown in this
figure. The experimental results (drx , dry ) of the multi
rosette analysis correspond with the analysis results of
the FEM.
REFERENCE
Uchino, M., Okamoto, T., Hida, K., Ito, Y., and Mastuda, H.,
2009. Investigation of Strain Analysis Method around a
Hole in a Plate Using Multi-Rosette Analysis by Digital
Image Correlation Method (in Japanese), Journal of the
Japanese Society for Experimental Mechanics (JJSEM),
Figure 1. Definition of deformation state at the border of
9(2): 96102.
the through hole.
501
Deflection measurement for bridges with frequency-shifted feedback laser
S. Umemoto, M. Fujii, N. Miyamoto & T. Okamoto

Keisoku Research Consultant Co., Hiroshima, Japan
T. Hara & H. Ito

Tohoku University, Sendai, Japan
Y. Fujino
The University of Tokyo, Tokyo, Japan
ABSTRACT Table 1. Results of verification.
Full-field optical measurement technology, which is Standard Measurement

Measurement Deviation for 50 mm Deviation
the focus of the authors efforts, is one field of Struc- Distance (mm) disp. (mm) (mm)
tural Health Monitoring technology. Full-field opti-
cal measurement technology is a technology which 50 m 0.1280.156 49.929 0.071
utilizes the properties of light to grasp informa- 300 m 0.1320.158 49.976 0.024
tion on objects of interest in 2- or 3-dimensions, 500 m 0.1370.169 50.119 0.119
and has several important advantages, including 1000 m 0.1490.181 49.986 0.014
1) remote/noncontact measurement, 2) collection of
information on numerous points within a short period
of time, and 3) visualization of that information.
The frequency-shifted feedback laser (hereinafter, FSF
laser) developed by Tohoku University, which is one
such technology, is an ultra-high accuracy measure-
ment technology (Fig.1), and because of its principle
of measurement, it has the distinctive features that
noncontact distance measurement is possible and mea-
surement accuracy is not dependent on the measure-
ment distance. Furthermore, because the maximum
sampling frequency of the FSF laser is 1000 Hz, mea-
surement of dynamic displacement behavior is possi-
ble. The authors have carried out various verification
studies with the aim of applying the FSF laser to Struc- Figure 2. Result of deflection measurement (Truss Bridge).
tural Health Monitoring, and made improvements in
for actual bridges are shown in Figure 2. Based on
the FSF laser at Tohoku University reflecting the
these results, it is possible to apply the FSF laser to
results. This paper reports the results of verification of
actual bridges. In particular, the FSF laser is expected
the distance and displacement measurement accuracy
to demonstrate its effectiveness in measurements of
of the new FSF laser in long distance measurements,
long bridges and bridges with high piers.
and also presents examples of application of the FSF
laser to measurements of bridge deflection in service.
The results of verification of the distance and dis- REFERENCES
placement measurement accuracy of the new FSF laser
at measurement distance up to 1000 m are summarized Umemoto, S, et al., 2007. Verification report of high accuracy
in Table 1. And the results of deflection measurement measurement technology by long distance and noncon-
tact with the FSF laser optical coordinates measurement
system. The 3rd International Conference on Structural
Health Monitoring of Intelligent Infrastructure, Vancou-
ver, November.
Umemoto, S, et al., 2008. The application of the frequency-
shifted feedback laser optical coordinates measurement
system for field measurement of bridges in service.The 4th
International Conference on Bridge Maintenance, Safety
and Management, Seoul, July.
Figure 1. The external appearance of the FSF Laser.
502
SS19: Implementation of bridge management
administration in Japan
Organizers: H. Furuta & E. Watanabe
Asset management of bridge structures in Osaka prefectural government
H. Ito & R. Tanaka

Department of Urban and Public Works, Osaka Prefectural Government, Osaka, Japan
H. Furuta
Department of Informatics, Kansai university, Takatsuki, Japan
1 INTRODUCTION 4 CONCLUSIONS
The Osaka Prefectural Government has been build- In 1999, around 900 bridges with more than 15 m
ing an asset management system for bridge structures. span length had been inspected in Osaka prefecture.
This system attempts to maintain existing bridges in Using the inspection data, the integrity level of each
good conditions and extend the remaining lives by bridge was calculated, which was used to make the
introducing the concept of Life-Cycle Cost (LCC). It ordering of repair actions. In the determination of the
is, then, necessary to achieve a rational plan of resource repair order, the importance of bridges was taken into
allocation through the normalization of annual budget. account as well as the damage states. The importance
In this paper, the asset management system developed was given by the significance level of the road involv-
by the Osaka Prefectural Government is described. ing the bridge. The importance level was classified into
three categories.
LCC calculation presents 3.5 million US dollars/
2 PRESENT STATE OF BRIDGES IN OSAKA year necessary for maintenance. Because of the present
PREFECTURE financial difficulty of Osaka prefecture, it is difficult
to secure the budget for the maintenance. Thus, the
There are 811 bridges (more than 15 m span) in Osaka necessary resource may be partly appropriated from
prefecture. The age of most bridges is 40 years, and other budget of repair works such as pavement, paint-
it will become 60 years after 20 years. The ratio of ing or other repair actions or may be managed to be
bridges with more than 50 years will become 56% secured through the more efficient utilization of road
from 9%. Then, it is necessary to equalize the annual infrastructures.
budget and change from the essentail maintenance to The improvement of painting has started to extend
the preventive maintenance in order to establish a ratio- the interval of repainting. Various actions have been
nal maintenance program under the limited amount of performed to obtain the understanding of tax payers for
budget. the importanece of maintence works throught the site
explanation meetings. The repair strategy and meth-
ods have been integrated and standardized to realize a
3 BRIDGE MANAGEMENT SYSTEM IN rational allocation of financial and human resources.
OSAKA PREFECTURE The movement of asset management of bridge
structures has spread all over Japan. Many local gov-
This system attempts to maintain existing bridges in ermnents established a forum of asset management
good conditions and extend the remaining lives by engineers, in which active discussions and experience
introducing the concept of Life-Cycle Cost (LCC). It and knowledge have been exchanged. The appeal for
is, then, necessary to achieve a rational plan of resource the necessity of preventive maintenance has started
allocation through the normalization of annual budget. and the movement has grown up.
505
Evaluation of maintenance cost estimation and feedback to the BMS
H. Kawaragi
Aomori Prefectural Government, Aomori, Japan
M. Kaneuji
Kajima Corporation, Tokyo, Japan
E. Watanabe
Osaka Regional Planning Institute, Toyonaka, Osaka, Japan
ABSTRACT
The Aomori Bridge Management System (ABMS)

has been developed in 2004 and 2005, and the Five
Year Bridge Management Plan was established using
ABMS.
The accuracy of maintenance cost estimation has
been evaluated in order to feedback the results to
ABMS. The accuracy of maintenance cost estima-
tion is vital, because the Budgetary Plan which is
the main part of Bridge Management Plan is solely
dependent on the maintenance cost estimation. If the
accuracy of maintenance cost estimation derived from Figure 1. Ratio of estimated and actual costs.
BMS is poor, the Bridge Management Plan will lose
dependability and it becomes difficult to keep PDCA
management cycle of BMS rolling. was limited. In other case, original plan was to replace
The investigation on the actual cost of the mainte- the bridge bearings, but in the actual rehabilitation
nance works against the estimated cost calculated by work, refreshing method was used to reduce long-term
ABMS was made on 52 bridges whose maintenance maintenance cost instead of replacing bridge bearings.
works were completed in the year of 2007. There were 5 bridges whose actual cost was more
Among 52 bridges, there were 9 bridges which were than 150% of estimated cost. Two of them were
not originally in the list of maintenance works but were steel girder bridge, whose original maintenance mea-
included in the 5-year action plan because the seismic sure, partial painting, was changed to whole painting
retro-fit was planned. In such cases, some maintenance because the old paint had PCB and it was necessary to
works were conducted along with seismic retrofit. In take all PCB off.
other cases, the originally planned maintenance action Taking those two groups out, the remaining group of
was cancelled after detailed investigation. Therefore, 30 bridges had actual cost with 50% to 150% of orig-
cost of newly added maintenance works was subtracted inally estimated cost. The group of 18 bridges, which
from the actual cost, and the maintenance works which is 35% of 52 bridges, had actual cost with 75 to 125%
was planned but not conducted were also subtracted of estimated cost. The mean value of cost ratio (actual
from the estimated cost. After making above men- cost/estimated cost) was 74%.
tioned modification, estimated cost and actual cost Considering the fact that the cost estimation is
were compared. based only on the visual inspection data from peri-
As shown in Figure-1, there were 17 bridges whose odic inspection, the accuracy of cost estimation is in
actual cost was less than 50% of originally estimated the satisfactory level.
cost. The main reason for the big decrease in cost was After three years of using ABMS, we are satisfied
the change of maintenance measures. For example, with the accuracy of the LCC estimation, and we do
original plan was to take off concrete with high concen- not feel that we need to change various data includ-
tration of chloride ion by water-jetting, but in the actual ing cost data at this moment. The feedback to ABMS
rehabilitation work the water-jetting was not used will be made after the completion of 5-year bridge
because the area of high concentration of chloride ion management plan.
506
Risk based inspection strategy considering structural redundancy of long

span bridges
N. Mashima & K. Sugioka

Osaka Business and Maintenance Department, Hanshin Expressway Company Limited, Osaka, Japan
H. Kobayashi & H. Kanaji

Construction Division, Hanshin Expressway Company Limited, Osaka, Japan
H. Oishi
Planning Division, Hanshin Expressway Management Technology Center, Osaka, Japan
K. Kaito
Graduate School of Engineering, Osaka University, Osaka, Japan
ABSTRACT
Generally, long-span bridges are large in scale and con-

tain numerous inspection points. Most of them are also
marine bridges with poor accessibility, which makes
maintenance, especially inspection, more difficult.
Some of the existing inspection gantries on long-
span bridges are approaching a time for costly renewal
on which decisions need to be made in a very near
future. For the purpose of rationalizing long-span
bridge inspections, this study investigated how to
determine optimum inspection intervals for long-span
bridges utilizing risk assessment, with a focus placed
on whether to use the inspection gantries or not.
Figure 1 shows an example result of 60-year LCC anal-
ysis for a single-plane cable-stayed bridge in relation to Figure 1. LCC vs. risk on a single-plane cable-stayed
the average expected risk obtained. Both LCC and risk bridge.
were lower than the current inspection policy (inspec-
tion at 7-year intervals using inspection gantries) at
the inspection intervals of 3 or 5 years with no-use of
inspection gantries.
Since long-span bridges tend to have a complicated
structure with high redundancies, it is not necessarily
reasonable to weigh all members equally in inspec-
tions. In an attempt to identify locations and members
to which higher inspection priority should be given,
the authors carried out an entire bridge analysis using
a network arch bridge model with corrosion caused
to various members and joints. Figure 2 shows verti-
cal displacements of the arch crown, respectively, with
live load applied to the center of the span. As shown
in this figure, when corrosion was caused in the arch
springings ((1)-a) or in the entire arch rib ((1)-c), limits
were exceeded and displacement rapidly increased at
a remaining section rate of 40%. Members with higher Figure 2. Arch crown displacements by member corrosion
risk of leading to overall deterioration were identified. pattern.
507
Approach for bridge management using BMS in West Nippon Expressway

Company Limited
T. Matsui, Y. Wada & S. Sakai

Nippon Expressway Research Institute Co., Ltd., Tokyo, Japan
T. Yasuzato
West Nippon Expressway Co., Ltd., Osaka, Japan
ABSTRACT
Under the circumstance that expressway bridges

should be deteriorating in the near future, the Nippon
Expressway Company Limited (NEXCO) developed
the bridge management system (NEXCO-BMS) in
2003 for the preventive bridge maintenance and the
system has been applied to bridge management in
NEXCO since then to keep the safety and security of
the expressway. This system is expected to support the
well-planned maintenance by determining bridge con-
dition, predicting future deterioration, and selecting
optimal timing and method for repair and/or reinforce-
ment. Since the integrated maintenance database was Figure 1. Lifecycle cost compared Preventive and Correc-
built such as for inventory data and the inspection tive actions (Case B).
records and repair/improvement history evaluation
data, we know the transition number of deteriorated Table 2. Scenarios of optimizing annual budget (Case A).
bridges which need repair and/or reinforcement by
using several kinds of prediction formula based on Items Scenario 1 Scenario 2 Scenario 3
engineering knowledge. So far the function of this sys-
tem has been partially improved for supporting the well Budget limit 2 billion 3 billion
per year yen yen
plan and maintenance work enforcement of the actual
Lifecycle cost 51.7 billion 48.8 billion 44.7 billion
project that suited the practical work in 2008. In this (45 years) yen yen yen
paper, outline of NEXCO-BMS and actual simulation Ratio of LCC 1.16 1.09 1.0
case in West Nippon Expressway Company Limited Ratio of project 1.7 2.8 15.6
are described. concentration*
By using improved NEXCO-BMS, several cases of
simulation might be calculated accurately. The finding *Ratio of maximum annual budget and average annual budget
obtained from the simulation is as follows (Table 1).
As for scenario of the corrective maintenance
Case A, if calibration ratio is not high, same deterio- As for scenario of the preventive maintenance
ration tendency might be occurred based on similar Case B, preventive actions is superior to corrective
environmental factors, deterioration curves and so one in the view point of lifecycle cost, also cal-
on. ibration such as deterioration curve, deterioration
mechanism is needed (Figure 1).
Table 1. Simulation cases in West NEXCO. Result when optimizing annual budget applied to
Case A, total lifecycle cost is the most economical
Case Case A Case B on available budget free case, nevertheless Ratio
of project concentration is highest. It indicates
Service life (year) 45 23 that budgeting case increases both total lifecycle
Number of member 445 90 cost and ratio of deteriorated members because of
Start year of simulation 2009
Term of simulation (year) 45
changing strategic optimal planning; on the other
Maintenance level Grade 1 to 4 hand, project concentration will be distributed.
(Table 2).
508
Preventive maintenance and technical development on long-span bridges
M. Okuda, I. Yamada & Y. Hasegawa

Honshu-Shikoku Bridge Expressway Company Limited, Kobe, Japan
ABSTRACT
This paper describes the preventive maintenance and

the technical development on the Honshu-Shikoku
Bridges (HSB) in order to inspect and maintain the
long-span bridges in healthy conditions for more than
200 years as a long-term program, continuing and
developing the bridge engineering in Japan.
The HSB, including 17 long-span bridges, were
completed and opened to traffic by 1999 as the three
permanent links between Honshu and Shikoku, i.e.,
two of four major islands in Japan. One of the most Figure 1. Dry-air injection system for Akashi Kaikyo
famous bridges is the Akashi Kaikyo Bridge, which Bridges.
was completed in 1998 with the worlds longest span
of 1991 meters. As the toll road project, the HSB
were designed and constructed by the advanced bridge
technologies, considering typhoons, earthquakes, tidal
currents and corrosive environments.
For the preventive maintenance of long-span
bridges, the dehumidification system for the main
cable, called as the dry-air injection system (Figure
1), has been applied to all 10 suspension bridges since
1998. The dry-air injection system is operated effec-
tively in order to enhance the durability of main cables.
In addition, the coating management system is under
way in order to reduce the life-cycle costs of steel struc-
tures. The coating thickness, especially the surface
coat and the intermediate coat, is periodically checked
at designated positions on the long-span bridges, eval- Figure 2. Automatic repainting on main tower, using auto-
uating the coating durability and determining the matic coating machine and magnet-wheeled gondola.
overall repainting. long time even though Japanese road policy will be
In the technical development of long-span bridges, changed in the future.
the automatic coating machine and the magnet-
wheeled gondola (Figure 2) were developed for steel
REFERENCES
box girders and steel towers in order to inspect the
coating durability and to repaint the deteriorated coats K. Kawaguchi,Y. Nagao & T. Sugimoto, Repainting of Seto-
safely and efficiently. On the other hand, the vacuum- Ohashi Bridge for Increase of Durability, ICSBOC 2006,
wheeled gondola was also developed for concrete New York, USA
structures in order to inspect the surface durability T. Sugimoto, K. Kawaguchi, & T. Hanai, Maintenance of the
with no scaffolding. In addition, a non-destructive Seto-Ohashi Bridges, REAAA 2006, Manila, Philippine
test, called as the electromagnetic flux method, was M. Nakamura, Y. Hasegawa, T. Hanai & T. Sugimoto, Pre-
developed and applied to the investigation of existing ventive Maintenance of the Honshu-Shikoku Bridges,
IABSE 2006, Copenhagen, Denmark
suspenders on suspension bridges, checking the area
T. Mohri, Y. Hasegawa & K. Sakai, Operation of Dry Injec-
reduction of suspenders and revising the maintenance tion System on Honshu-Shikoku Bridges, ICSBOC 2008,
program. Takamatsu, Japan
By continuing the preventive maintenance and tech- Y. Hasegawa, S. Kurino & Y. Fujisawa, Preventive Main-
nical development on long-span bridges, the HSB shall tenance of the Suspender Ropes on the Honshu-Shikoku
be operated and managed in sound conditions for a Suspension Bridges, REAAA 2009, Incheon, Korea
509
Risk evaluation and financial analysis for road maintenance

on urban expressway based on H-BMS
Y. Sakai
M. Jido
Chuo Fukken Consultants Co., Ltd., Osaka, Japan
H. Furuta
Department of Informatics of Kansai University, Osaka, Japan
K. Kobayashi
Graduate School of Management, Kyoto University, Kyoto, Japan
ABSTRACT On the other hand, uncertainty of deterioration pro-

cess is treated as risk, and the relationship between
Hanshin Expressway Public Corporation had been pri- this deterioration risk and maintenance/rehabilitation
vatized in 2005 and has responsibility for maintaining cost is verified. Based on the risk evaluation result, this
service level of civil structures at sound condition and study tries to calculate financial information which can
completely paying off its debt within 45 years after pri- be used to making debt payment plan.
vatization. Therefore, it is an important issue to estab- Finally, for continuous improvement of the busi-
lish maintenance and rehabilitation plan considering ness process, extraction of the high priority monitored
the payment of the debt. structures is discussed based on relative evaluation
This study addresses on Hanshin Expressway asset method of deterioration rate of each structure using
management system considering internal control and benchmarking analysis.
risk management system. Concretely, first of all, busi-
ness process of Hanshin Expressway is constructed in
terms of internal control and risk management.
510
Study for bridge renewal and repair by Osaka municipal government
E. Watanabe
H. Furuta
Kansai University, Takatsuki, Japan
T. Yokota, Y. Komatsu & S. Nagahashi

Public Works Bureau, Osaka Municipal Government, WTC Building, Nankoh Kita, Suminoe-ku, Osaka City,
Osaka, Japan
ABSTRACT deterioration rate of the bridges. However, in view

of the fact that the old bridges built before the Sec-
The Osaka Municipal Office has been operating the ond World War were designed using different codes
Bridge Management System (OBMS) since 2005. from now with respect to loads, earthquake-resistant
OBMS executes the optimum planning for the pre- design and river conditions and do not correspond to
ventive bridge maintenance through the periodical the present design codes. Thus, the judgment on the
inspection, efficient data acquisition and the prior- bridge replacement only on the basis of the bridge
ity rule taking into the historic and civic importance. health assessment may be misleading. From this stand-
According to a trial calculation, the total budget in point, the final decision was proposed by adding the
coming 30 years can be reduced by 50% if the pre- functionality in addition to the health assessment.
ventive maintenance is executed on 100 old bridges In conclusion, as a result of the Replacement judg-
by implementing the renewal of limited number of ment matrix and the LCC analysis, 10 of bridges
bridges, for example, only 10 bridges in comparison were selected for the replacement in addition to those
with the renewal of every one of these 100 old bridges. bridges which were to be renewed that are budgeted
Recently, Osaka City is in a serious financial situa- by different budget plans.
tion. Although it is still desirable to build new bridges, This paper explains the basic concepts regarding
renew old bridges or retrofit seismically vulnerable the project on the replacement of old and historic
bridges, the cost reduction by means of new main- bridge (OCPEWF, PWBOMG). Thanks to the well-
tenance technologies is above all essential since the preserved drawings, documents of summary of works
City of Osaka has a limited bridge budget. and records of works on the bridges built before the
Osaka Municipal Office is anxious to enthusiasti- Second World War, the project has been promoted
cally develop the project of preventive maintenance; quite smoothly. Thus, the effort and tradition of the
it found out that the renewal or replacing old bridges bridge engineers of the Osaka Municipal Office must
should be kept minimum. Therefore, the countermea- be highly respected. The importance of the succession
sure for the old bridges by the Osaka Municipal Office of the important documents and information to the next
may be classified into the preventive maintenance for generations is becoming more and more important.
the majority of old bridges but only a limited number
of renewal or replacement.
Osaka city had made the first selection of one hun- REFERENCES
dred old bridges that have been in service for more
than 70 years since before the World War II (excluding OCPEWF: City of Osaka Public Engineering Works Foun-
dation. Bridges in Osaka City. History of bridge tech-
those bridges which are targets of special replace-
nology in Osaka City (in Japanese) http://www.osaka-
ment or removal budget plan), to perform the detailed udce.or.jp/bridge/chap1/index1.htm
analysis through the Replacement judgment matrix. PWBOMG: Public Works Bureau of Osaka Municipal Gov-
Through this screening process, only a limited num- ernment (in Japanese) http://www.city.osaka.jp/kensetsu/
ber of bridges were selected for the LCC analysis to shigoto/hashi/top.htm
see if they are worth being considered for preventive Yokota, T., Komatsu, Y. and Nagahashi, S. 2008. Replace-
maintenance or the replacement (Yokota 2008). ment plan of old bridges by Osaka Municipal Government.
It is a general practice to decide the mainte- Bridge and Foundation Engineering, 42(8): 3638 (in
nance plan based on the field inspection data and the Japanese).
health assessment of bridge taking into account the
511
SS20: Challenges for enhancing bridge security
Organizer: S.R. Duwadi
Recognizing and reducing vulnerabilities of transportation infrastructure
S.R. Duwadi
Federal Highway Administration, McLean, Virginia, USA
ABSTRACT DOT 2009 National Transportation Statistics, Bureau of

Transportation Statistics, Department of Transportation,
Incorporating robustness and resiliency into highway http://www.bts.gov/publications/national_transportation_
infrastructure design has evolved over time through statistics/index.html#introduction).
Bridges WP, Bridges for the 21st Century, Focus Area
lessons learned from each hazard event. As a result
Bridge Safety, Reliability and Security White Paper,
of hurricanes, earthquakes, flooding and scour there Safety, Reliability and Security Team, Federal Highway
have been a greater understanding of the impact of Administration, 2004, unpublished.
these on the infrastructure and its vulnerability. Solu- Chase, S. B; Duwadi, S.R; Hooks, J.M, Getting Ahead
tions have been developed over years through research of the Curve, Public Roads, November/December
and development effort, although there are still chal- 2003, Vol. 67, No. 3, Federal Highway Administration
lenges that remain. The understanding & designing for http://www.tfhrc.gov/pubrds/03nov/01.htm.
security however is relatively new, and therefore mul- Methodology 2009 Methodology for Assessment of Natural
titude of challenges lay ahead. Unlike natural hazards Hazard Vulnerability in U.S. Coastal Zone Using Remote
Sensing, 2009 http://freshhealthybody.com/methodology -
where effort has been spent over decades to understand
for - assessment - of - natural-hazard-vulnerability-in-u-s-
behavior, effort to understand behavior of structures to coastal-zone-using-remote-sensing/.
explosive loadings or other security risks is relatively NCHRP Project 12-72, Blast-Resistant Highway Bridges:
new and still evolving. The importance of recognizing Design and Detailing Guidelines, Final Report, 2008,
and defining vulnerabilities to each hazard event is crit- Unpublished, National Cooperative Highway Research
ical for developing solutions to keep these structures Program, Transportation Research Board.
functional after any event. This paper will discuss the
importance of recognizing vulnerabilities of highway
bridges to natural and manmade hazards and present
a summary of solutions that have been developed to
mitigate these hazards.
REFERENCES
FHWA 2006 Conditions and Performance Report, Status
of the Nations Highways, Bridges, and Transit, Federal
Highway Administration, Department of Transportation,
2006 http://www.fhwa.dot.gov/policy/2006cpr/index.htm.
515
Bridge design relevance and efficiency of protective measures for bridge

structures under severe loading
M. Noeldgen & W. Caspari

Schuessler-Plan Engineering Consultants, Duesseldorf, Germany
J. Krieger
Federal Highway Research Institute, Bergisch-Gladbach, Germany
ABSTRACT shows that this scenario does not necessarily lead to

an abrupt failure or progressive collapse of the global
The research project SKRIBT (Protection of critical structure as detailed analysis of maximized dynamical
bridges and tunnels in the course of roads) is con- internal forces shows. The analysis is based on safety
cerned with the development and demonstration of factors analog to the extraordinary load combination
a comprehensive methodology for the evaluation of in Eurocode 1 and compared to the materials yield
infrastructure subjected to severe natural and man- strength. Against this background the components
made hazards. The presented paper is focused on the remain linear elastic. It can be shown that damage
resilience of bridge structures as a part of this general stays localized and only local damages caused by
analysis and the relevance and efficiency of protective the explosion have to be reconstructed. The result-
measures under severe loading conditions. ing bearing capacity limits the allowable traffic loads
Starting with a thorough investigation of potential to 60% of the original loads until the bridge is fully
and available measures and an identification of rel- restructured.
evant scenarios the extraordinary loading cases are The described damages are classified in a five step
integrated in the semi-probabilistic design approach categorization of damage and evaluated in a compre-
provided in the current Eurocode standards. The sub- hensive approach together with all relevant affected
sequent analysis of those actions and structural resis- aspects for the user and the environment.
tance is demonstrated by means of a stay-cable bridge Once the comprehensive evaluation system for the
example. criticality and vulnerability is established, the variety
Albeit the loss of a single cable has to be taken into of measures (see table 1) can be evaluated using the
account in regular bridge design neither the loss of same methodology. Thus the procedure enables for the
a cable group nor the significant additional dynam- evaluation of the relevance and efficiency of protective
ical effects possibly caused by man-made hazard are measures for bridge structures under severe loading
covered in European standards. The presented analysis conditions.
516
Mitigation of terrorist threats to structural components
J. Ray, V. Chiarito, B. Walker & J. Minor

US Army Engineer Research and Development Center (ERDC)
Permission to publish was granted by the Director, Geotechnical & Structures Laboratory, ERDC
ABSTRACT Previous research and mitigation efforts for build-

ings (where most efforts were concentrated prior to
Many significant structures, including buildings, 911) always assumed some reasonable degree of
bridges, tunnels, etc. are threatened by terrorist attack. bomb standoff, and distance, even a small distance,
There are unlimited possibilities as to the types of is by far the most effective means of blast mitiga-
terrorist threats that could be brought against these tion. Unfortunately, many structure owners do not have
structures. Based upon past history, the predominant this luxury as enforcement of any reasonable stand-
threat of concern has been vehicular bombs, or using off will essentially require closure of the asset. Thus,
the current term, vehicle-borne improvised explosive much more effort must be given to blast hardening of
devices (VBIEDs). these critical structures and understanding the load-
There are many approaches to terrorist threat mit- ing and response phenomena associated with these
igation, including the four Ds that are commonly severe blast pressure environments very close to the
defined as: Deterrence, Detection, Denial, and physi- explosive.
cal Defeat. A complete mitigation plan for any struc- Currently, the engineering approaches to structural
ture will incorporate all four of these components. hardening against blast are very widespread, with some
However, with a vehicle bomb threat, the first three Ds being sound and others being poor or improperly
are often of limited benefit in that many of our nations applied for the given situation. Much of the prob-
critical structures, such as bridges and tunnels, demand lem arises from inexperience in the arena of explosion
unlimited vehicular access by their very nature, leav- effects and blast loading as many engineers are just
ing Defeat, or more accurately, hardening, as the being introduced to this unique field. As a result, there
only option. is much confusion and mis-information. It is the pur-
Even worse, critical structural components are often pose of this paper to provide a basic overview of
directly adjacent to and within inches of high-traffic explosive blast mitigation techniques for basic struc-
areas, making them extremely vulnerable to detona- tural components that are applicable to any structure
tions from VBIEDs. Further compounding the prob- type. Mitigation techniques are categorized and dis-
lem, these structural components are often very weak cussed in terms of the manner in which they affect
in the direction of applied loading from a VBIED. For the most basic law of motion, Newtons second law.
example, columns are predominantly designed for ver- Results from recent high explosive tests conducted
tical axial loadings. Yet, a VBIED would generally by the authors are presented and used to show the
detonate at such a location that its extreme pressure effectiveness of the varied concepts.
loadings will be applied laterally on the column, in its
weakest structural direction.
517
SS21: Structural monitoring of bridges:
Hong Kongs experience
Organizers: Y.-L. Xu & M.C.H. Hui
Structural health monitoring and safety evaluation of Stonecutters Bridge

under the in-service condition
K.Y. Wong
Bridges & Structures Division, Highways Department, The Government of Hong Kong Special Administrative Region
ABSTRACT evaluate the stability of the bridge structural sys-

tem when the monitoring criteria are exceeded.
A structural health monitoring and safety evaluation The structural performance limits at the service-
system (SHM&SES) is deployed by Highways Depart- ability limit state should be used as the base-line
ment (HyD) to monitor the structural health condition of structural health monitoring.
of Stonecutters Bridge under its in-service condition (2) Performance limit for monitoring and evaluation
and to evaluate its structural safety when exceedance The performance of the bridge structural sys-
of monitoring criteria are identified in certain loca- tem to be monitored by the SHMS is devised in
tions and/or components.The SHM&SES is composed accordance with the structural performance lim-
of 4 integrated systems, namely, structural health its defined by BS5400 at the serviceability limit
monitoring system (SHMS), structural health rating state; whereas the structural safety to be evalu-
system (SHRS), structural health evaluation system ated by SHES is devised in accordance with the
(SHES) and structural health data management sys- structural performance limits defined by BS5400
tem (SHDMS). This paper outlines the functional and at the ultimate limit state.
architectural requirements of the SHMS, SHRS, SHES (3) Codified requirements in monitoring and
and SHDMS. The modification works done on pre- evaluation Apart from structural health motor-
vious SHMS in HyD for enhancing the operational ing and safety evaluation, the key functions of
efficiency of SHM&SES are also described. development and deployment of the SHMS&SES
A summary of structural health monitoring and are: (a) to improve/enhance the current practice
safety evaluation for long-span bridges is outlined as of bridge inspection and maintenance from local
follows: and subjective condition to global and objective
condition; and (b) to provide data and information
(1) Codified performance limits Structural health
for updating/amending the contemporary bridge
monitoring refers to the usage of on-structure
design manuals, standards and codes which are
instrumentation system to monitor the perfor-
the standards for future works of: (i) design and
mance of the bridge structural system; whereas
construction of new bridges and (ii) maintenance
structural safety evaluation refers to the usage of
and rehabilitation of old bridges.
analytical tools such as finite element software to
521
Structural health monitoring of Stonecutters Bridge under

the construction stage
K.Y. Wong
Bridges and Structures Division, Highways Department, The Government of Hong Kong Special Administrative
Region
D.M.S. Yau & M.C.H. Hui

Formerly Major Work Project Management Office, Highways Department, (Currently Development Bureau)
The Government of Hong Kong Special Administrative Region
ABSTRACT 2. The stress monitoring in key locations of bridge

showed that the strength limits had not been
The Stonecutters Bridge is a two-cable-plane cable- exceeded during the construction stage.
stayed bridge and carries dual-3 highway traffics. It 3. The stay cable force monitoring shows that 92.4%
is currently the worlds second cable-stayed bridge (207/224) of the stay cables are within 10% of the
with the main/central span length exceeding 1 km. analysis/design results.
In the bridge construction stage, Highways Depart- 4. The Scruton Number of most stay cables, basing on
ment deployed a structural health monitoring system to damping ratios measurement on selected samples
investigate the variation of structural health conditions (1 out of 4) of stay cables, are less than 10. This
of the bridge under various stages of construction, with supports the necessity for installation of dampers
particular reference to the erection stages of steel deck in stay cables for reducing the potential risk of rain-
segments at main/central span. This paper presents the wind-induced vibrations on stay cables.
monitoring works in six aspects: (i) objectives and
scope of monitoring scope, (ii) monitoring equipment
and facilities, (iii) monitoring base-lines, (iv) bridge
geometry profiles monitoring, (v) stay cable force and
damp-ing ratio monitoring and (vi) steel deck segment
strain/stress monitoring.
Finally the monitoring works of Stonecutters Bridge
under the construction stage are summarized as fol-
lows:
1. The as-built bridge geometry profile basically sat-
isfies the upper limit as stated in the Construction
Speci-fication for Stonecutters Bridge.
522
Thermal behaviors of Tsing Ma Suspension Bridge
Y. Xia, B. Chen, Y.Q. Bao & Y.L. Xu

Department of Civil and Structural Engineering, The Hong Kong Polytechnic University, Hong Kong, China
ABSTRACT
The Wind And Structural Health Monitoring System

(WASHMS) for Tsing Ma Bridge has been operated
since 1997. The WASHMS is devised to carry out the
monitoring of four categories of parameters, namely,
environmental status, traffic loads, bridge features, and
bridge responses. This paper studies on temperature
distribution of the Tsing Ma Suspension Bridge via
both field monitoring and thermodynamic analysis. Figure 2. Monthly ambient temperature in 19972005 (T1).
With appropriate simplification, the entire large-
scale bridge can be divided into several thermal fields
and each component can be modeled with fine finite
element models. With appropriate thermal boundary
conditions, transient thermal analysis is conducted in
the present study to predict the temperature distri-
bution of the deck plate, bridge section, and bridge
tower at different time and at different seasons. Vari-
ation of the temperature data at different time is
obtained and compared with the measurement data
from the WASHMS. The comparison shows a very
good agreement between the model prediction and
field measurement. Consequently the FE models are
verified.
The numerical results provide thorough information
regarding the temperature distribution of the bridge.
The temperature data can be employed to calculate
the thermal responses of the bridge, which can be
used to evaluate the environmental effect on bridges
performance quantitatively.
The Tsing Ma Bridge with a total span of 2132 m
is a suspension bridge as shown in Fig. 1.
Fig. 2 shows a clear and fairly stable cycle of
monthly ambient temperature variation. Figure 3. Temperature fields of tower section at 13:00 pm.
Fig. 3 illustrates the temperature distribution across
the tower section at 13:00 pm. It is clearly shown that
the structural temperature in the region near the surface REFERENCES
is much higher than those inside the tower.
ANSYS Inc. 2004. ANSYS Manual 2004. Southpointe, PA,
USA.
Branco, F.A., Mendes, P.A. & Mirambell, E. 1992. Heat of
hydration effects in concrete structures. ACI Materials
Journal 89(2): 139145.
Churchward, A. & Sokai, Y.J. 1981. Prediction of tempera-
tures in concrete bridges. Journal of Structural Division
ASCE 107: 21632176.
Figure 1. Position of temperature sensors in Tsing Ma
Suspension Bridge.
523
SHMS-based bridge rating method for long span cable-supported bridges
Y.L. Xu, Y. Zheng, Z.W. Chen & Y. Xia

Q. Li
A.X. Guo
K.Y. Wong
Bridges & Structures Division, Highways Department, Hong Kong
ABSTRACT relatively accurate condition evaluation of the bridge

and to reduce uncertainties involved in the existing rat-
Long span cable-supported bridges begin to deterio- ing method. The procedures for determining relative
rate once they are built and continuously accumulate weights and fuzzy synthetic ratings for both critical-
damage during their service life due to natural hazard ity and vulnerability are then suggested. The fuzzy
and harsh environment such as typhoons, earthquakes, synthetic decisions for inspection are made in consid-
vehicles, temperature and corrosion. To ensure the ser- eration of the synthetic ratings of all structural compo-
viceability and safety of long span bridges, bridge nents. The SHMS-based F-AHP bridge rating method
rating systems are often adopted by bridge man- is finally applied to the Tsing Ma suspension bridge in
agement authorities as guidance in determining the Hong Kong as a case study.The field measurement data
time intervals for inspection and the actions to be recorded by the SHMS installed in the bridge for the
taken in the event of defects being identified. Most past ten years are used to update the existing bridge
of the currently-used bridge rating methods are based rating system. The effects of different comparison
partly on engineering analysis and partly on prac- matrices on the inspection decision are investigated.
tical experience. Kushida et al. (1993) proposed a The results show that the effects of relative weights
membership function to quantify the subjective uncer- from different comparison matrices are small. For the
tainty included in empirical knowledge on bridge bridge components concerned, the time intervals for
rating. Aktan et al. (1996) integrated analytical and inspection are either 1 year or 2 years. The results from
experimental research to assess global conditions and the case study indicate that the proposed bridge rating
evaluate the serviceability and safety of bridges. Liang method is feasible and can be used in practice for long
et al. (2001) set up an evaluation multiple layer fuzzy span cable-supported bridges with SHMS.
method for evaluating the damage stage of existing
bridges. Sasmal et al. (2006) developed a systematic
procedure and formulations for rating existing bridges REFERENCES
using fuzzy mathematics.
On other hand, structural health monitoring tech- Aktan, A.E., Farhey, D.N., Brown, D.L., Dalal, V., Helmicki,
A.J., Hunt, V.J. & Shelley, S.J. 1996. Condition assessment
nology gains a rapid development recently. Struc-
for bridge management. Journal of Infrastructure Systems
tural health monitoring systems (SHMS) have been 2(3): 108117.
designed and installed in a number of long span bridges Kushida, M., Tokuyama, T. & Miyamoto, A. 1993. Quan-
to monitor their serviceability and safety. However, tification of subjective uncertainty included in empiri-
there is insufficient link between the bridge rating cal knowledge on bridge rating. Uncertainty Modeling
method and the SHMS to fulfill the common goals. and Analysis, 1993. Proceedings, Second International
In this regard, this paper presents a SHMS-based Symposium on.
bridge rating method for long span cable-supported Liang, M.T., Wu, J.H. & Liang, C.H. 2001. Multiple layer
bridges. The fuzzy based analytic hierarchy process fuzzy evaluation for existing reinforced concrete bridges.
Journal of Infrastructure Systems 7(4): 144159.
(F-AHP) is first employed, and the hierarchical struc-
Sasmal, S. Ramanjaneyulu, K. Gopalakrishnan, S. & Lak-
ture for synthetic rating of each structural component shmanan, N. 2006. Fuzzy logic based condition rating of
of the bridge is proposed. The criticality and vul- existing reinforced concrete bridges. Journal of Perfor-
nerability analyses are performed largely based on mance of Constructed Facilities-ASCE 20(3): 261273.
the field measurement data from the SHMS to offer
524
A method for stress concentration factor determination of welded steel

bridge T-joints under moving load
X.W. Ye, Y.Q. Ni, & J.M. Ko

K.Y. Wong
Bridges & Structures Division, Highways Department, Hong Kong
ABSTRACT (TMB). The strain data of the pre-allocated measur-

ing points are acquired and the hot spot strain at the
Fatigue is one of the most important failure modes of weld toe is determined by a linear regression method.
metallic components of aircraft, civil structures, and The SCF is then calculated as the ratio between the
mechanical systems. Fatigue life prediction for struc- hot spot strain and the nominal strain which is derived
tural components subject to oscillating stresses with from the measured data from desired strain gauge. To
constant amplitude is given by the well known Whler take full account of the effect of predominant factors
curve, also called S-N curve. In the real applications, on the scatter of SCF, the experiments are carried out
however, the nominal stress approach is not suitable for under different moving load conditions. The statisti-
structural joints if the object detail is complicated and cal properties and probability distribution of SCF are
incomparable to any classified joints, or the loadings achieved, which reveal that SCF for the welded steel
are complex to make it difficult or impossible to deter- bridge T-joint conforms to a normal distribution. The
mine the nominal stress. An alternative method for the proposed method will benefit the fatigue life and relia-
fatigue analysis of complicated welded steel joints is bility assessment using measurement strain data from a
the hot spot stress approach which is more accurate long-term structural health monitoring (SHM) system
and reliable than the nominal stress approach. since the sensors for strain measurement are usually
When using the hot spot stress method for fatigue not installed at the most critical locations due to the
design and fatigue life prediction of welded struc- limitations of implementation technology and specific
tures, a critical issue is focusing on how to determine field conditions.
the stress concentration factor (SCF) for the welded
details. In general, the determination of SCF can be
achieved by means of finite element analyses, lab- REFERENCES
oratory experiments, or field tests. Research efforts
have been devoted to investigating the SCF deter- Chan, T.H.T. Zhou, T.Q. Li, Z.X. and Guo, L., 2005. Hot
spot stress approach for Tsing Ma Bridge fatigue evalua-
mination (Karamanos et al. 2000, Fung et al. 2002,
tion under traffic using finite element method. Structural
Gho et al. 2003, Chan et al. 2005, Gao et al. 2007), Engineering and Mechanics, 19, 261279.
and most of them are related on the welded tubular Fung, T.C. Soh, C.K. Chan, T.K. and Erni, 2002. Stress
joints and seldom are devoted to investigating the SCF concentration factors of doubler plate reinforced tubular
properties for the welded plate joints of large-scale T joints. Journal of Structural Engineering, ASCE, 128,
civil engineering structures, especially for the cable- 13991412.
supported steel bridges. Furthermore, because of the Gao, F. Shao,Y.B. and Gho, W.M., 2007. Stress and strain con-
considerable effects and sources of uncertainties dur- centration factors of completely overlapped tubular joints
ing the determination of SCF either by finite element under lap brace IPB load. Journal of Constructional Steel
Research, 63, 305316.
analyses or by experimental measurements, an inves-
Gho, W.M. Fung, T.C. and Soh, C.K., 2003. Stress and strain
tigation into the stochastic properties of SCF is of vital concentration factors of completely overlapped tubular
necessity. K(N) joints. Journal of Structural Engineering, ASCE,
The purpose of this research is to seek a method 129, 2129.
for determining the SCF and its stochastic characteris- Karamanos, S.A. Romeijn, A. and Wardenier, J., 2000.
tics for a typical welded steel bridge T-joint composed Stress concentrations in tubular DT-joints for fatigue
of two perpendicular steel plates, which is perceived design. Journal of Structural Engineering, ASCE, 126,
by conducting full-scale model experiments of a rail- 13201330.
way beam section of the suspension Tsing Ma Bridge
525
Long-term performance of structural health monitoring system in bridges
S. Zhu, W.A. Shen & Y.L. Xu

ABSTRACT
As a cutting-edge technology for infrastructural haz-

ard mitigation, structural health monitoring systems
(SHMS) have gained worldwide acceptance which is
witnessed by a number of long-term SHMS installed
on those recently-built long-span bridges over the past
two decades. Being a basic component in SHMS, sen-
sor systems play an important role in data acquisition
and transmission. Although SHMS systems are gener-
ally intended for long-term service, very rare attention
has been paid to the long-term performance of sensor
systems. It has been observed recently that the lifetime
of many installed sensors may not be as long as ser-
vice life of bridge structures, which requires proper
inspection, maintenance or complete replacement of
sensor systems during the lifetime of bridges. The tra- Figure 1. Time-variant performance index ( = 0.5).
ditional optimal design approaches based on initial
performance and cost may be greatly compromised
by neglecting these factors. system are generally shorter than the median life of
In light of this limitation, this study focuses on the individual sensors, and both are greatly affected by
health of SHMS in bridges. To assess the functional- the prescribed performance threshold in failure crite-
ity of sensor systems, a performance index is defined rion. It is also found that a large dispersion of sensor
based on the observation errors of modal response. life adversely affects the median life and the reliability
Using Monte Carlo simulation, the time-variant per- of the sensor system with no maintenance actions.
formance index of sensor systems can be obtained The discussion of the MTTF and AMP of the sensor
based on the lognormal life distribution of individual system will shed light on more rational design strat-
sensors. The optimal placement of a vibration sen- egy of SHMS in futurean optimal design method of
sor system composed 100 sensors is first conducted sensor systems in terms of their life-cycle cost instead
for Tsing Ma Bridge located in Hong Kong, and of the initial cost.
such a system is employed in the case study of long-
performance sensor system performance. The simula-
tion results clearly demonstrate the deterioration trend REFERENCES
of the sensor system performance. A performance
index threshold is adopted as the failure criterion for Kammer, D.C. 1992. Effect of noise on sensor placement
the sensor system. Subsequently, two important char- for on-orbit modal identification of large space structures.
Journal of Dynamic Systems, Measurement, and Control,
acteristics, the median-time-to-fail (MTTF) and the
114: 436443.
average maintenance period (AMP), are discussed in Xu, Y.L., Ko, J.M. & Zhang, W.S. 1997. Vibration stud-
detail. The effects of two key variables, the disper- ies of Tsing Ma Suspension Bridge. Journal of Bridge
sion in sensor life and failure threshold of performance Engineering, 2(4): 149156.
index, are also examined within the procedure through Yang, G. 2007. Life cycle reliability engineering, John
parametric studies. The MTTF and AMP of the sensor Wiley & Sons, INC. Hoboken, NJ, USA.
526
SS23: Performance-based design for steel structures
Organizers: S.-H. Kim & J.-S. Kong
Determination of load actions for performance-based design
H. Furuta
Department of Informatics, Kansai University, Takatsuki, Japan
N. Sato
Department of Civil Engineering, Chuo University, Tokyo, Japan
M. Kawatani
1 INTRODUCTION and effect of action. Chapter 5 is the analysis of actions

and their combination, and Chapter 6 is associated
Recently, performance-based design has gained atten- with the classification of actions, and the definition
tion as a design method that can provide more rational of terminology is given in Chapter 7. Part I has five
structural designs. In the performance-based design, appendices, first of which is the combination of load
it is possible to allow any design as long as it satis- actions and the second is uncertainties involved in load
fies the requirements of performance. This means that actions and design loads. The third one is associated
designers can try to produce new innovative structures with the establishment of action model based on sta-
under more flexible environments. Needless to say, it tistical approachs. The fourth and fifth are accidental
is necessary for practical design cases to solve such load actions and the present status of actions in some
issues as how to prescribe, guarantee, and evaluate the international design codes.
required performance, and prove the rationality and
optimality of the design obtained.
In establishing process of performance-based 3 CONCLUSIONS
design, design loads have not been sufficiently dis-
cussed from the viewpoint of performance-based The outline of the content of the guideline for Actions
design. Task committee on Actions on Civil Engineer- on Civil Engineering Structures for Performance
ing Structures for Performance-Based Design Codes Based Design Codes (Guideline for Code Writers)
(Guideline for Code Writers), Japan Society of Civil published in March, 2008 from Japan Society of Civil
Engineers has completed the report on load actions Engineers was summarized in this paper. In the estab-
for performance-based structural design, which con- lishment of performance-based design, design loads
sists of two parts; Part I is related to the objectives, have not been sufficiently discussed from the view-
framework, classification of load actions, and techni- point of performance-based design. A new concept of
cal terms, and Part II describes the detail explanation Action was introduced to determine the design loads,
of various load actions. In this paper, the outline of its which can be available for the design load modeling.
content will be introduced and some problems will be It is believed that this guideline could contribute to
discussed for its practical use. the achievement of performance-based design. Fur-
thermore, the guideline could provide useful infor-
mation for various structures such as steel, concrete,
2 PART I GENERAL FRAMEWORK composite and geotechnical structures, in order to
make a common consensus for their mutual under-
In Part I, general framework of actions is described. standing. In addition, code writers can refer some
In Chapter 1 the objective of the guideline is made references and accumulated data to specify the loads.
clear, and Chapter 2 describes basic policy which The guideline was completed by the joint task
covers description principle, description scope, and committee for guideline of design load for civil engi-
description policy. Chapter 3 is dedicated to applica- neering structures. The authors express their sincere
tion range and past treatment in the past design codes. gratitude for all committee members for their great
In Chapter 4, framework of actions is given by cov- contributions.
ering fundamental condition of design, action factor,
529
A study on vibration performance estimation of footbridge using human

body model
S.-H. Kim, K.-I. Cho & M.-S. Choi

Yonsei University, Seoul, Korea
ABSTRACT
In recent decades there has been a trend towards to

improve mechanical characteristics of materials used
in footbridge construction. Engineers are now design-
ing lighter, more slender and more aesthetic structures.
As a result of these construction trends however, many
of these footbridges are becoming more susceptible to Figure 1. Time histories of response from human body
vibration when subjected to dynamic loads. In most model R.
cases the vibrations of footbridges lead to serviceabil-
ity problems, i.e. the inconvenience of the pedestrians
Table 1. Maximum responses for each analysis model.
or in some extreme cases a bridge may no longer
be used and has to be closed. The human-induced Classical
dynamic loading occurs frequently and it is often Human Human random
regarded as dominant load for footbridges because it body body walking
sometimes obstructs pedestrians to walk conveniently. model S model R model n
Therefore, a three dimensional analysis program is
developed in this study using the human body model. Maximum 18.249 12.81 4.87
This program can effectively generate pedestrian exci- displacement (mm)
tation for the dynamic behavior of a footbridge under Maximum 3.70 1.84 1.00
human-induced load. acceleration (m/s2 )
A new human body model is proposed in this study.
It is comparatively easier to consider these effects in
the human body model. Also, it has advantages to
apply complicated loading conditions such as loads human body model R and S compared to the clas-
from crowded pedestrians. sical random walking model n. Such results indicate
Therefore, dynamic analysis is performed for the that the human-structure interaction force and the iner-
crowded pedestrians using human body model. Three tial forces of human masses somehow affected the
different walking models are compared, namely, the responses whole bridge.
randomly walking human body model (human body
model R), the synchronized model (human body model
S) and the classical randomly walking model (classical REFERENCES

random walking model n) which was proposed by the
Matsumoto et al. (1978). Matsumoto, Y., Nishioka, T., Shiojiri, H., and Matsuzaki, K.
(1978). Dynamic design of footbridges IABSE Proceed-
Pedestrian density is set to be 1.5 pedestrians/m2 as ings P-17/78, pp. 115.
it represents crowded walking (Schlaich, 2002). Fig- Schlaich, M. (2002). Planning conditions for footbridges Pro-
ure 1 indicates the example of responses from human ceedings of the International Conference on the Design
body model R. As shown in table 1, larger displace- and Dynamic Behavior of Footbridges, Paris, France. pp.
ment and acceleration responses can be found in both 4052.
530
I-girder with discrete torsional bracings: Lateral-torsional buckling

and torsional vibration
C.T. Nguyen & H.E. Lee

School of Civil, Environmental & Architectural Engineering, Korea University, South Korea
ABSTRACT (Hibbit, Karson & Sorensen 2001) and those obtained

by previous researchers (Trahair 1993, Yura 2001).
The lateral-torsional buckling (LTB) strength of an From the results, it is found that the proposed equations
I-girder can be increased by using an adequate brac- show good correlation with finite element analyses
ing system. Torsional bracing can be divided into two regardless of the number of bracing points and are suc-
categories: continuous and discrete torsional bracing. cessfully verified. Finally, a reduced formula for the
For the LTB strength of beams with continuous tor- total stiffness requirement is proposed for the design
sional bracing, Taylor & Ojalvo (1966) developed the purpose.
exact solution for the critical moment of beams. Their
solution is adapted for discrete torsional bracing by
summing the stiffness of each bracing along the span REFERENCES
and dividing by the beam length to obtain an equiva-
lent continuous bracing stiffness (Trahair 1993, Yura Taylor, AC & Ojalvo, M. 1966. Torsional restraint of lat-
eral buckling. Journal of Structural Division ASCE 92(2):
2001). Trahair (1993) proposed a stiffness requirement
115129.
for beams with a mid-span torsional restraint based Trahair, N.S. 1993. Flexural-torsional buckling of structures.
on a numerical approximation. Valentino & Trahair London: E & FN Spon.
(1998) performed several solutions for beams with Yura, J.A. 2001. Fundamentals of Beam Bracing. Engineering
midspan torsional bracing under various loading con- Journal AISC First Quarter: 1126.
ditions. Thus, it seems that a significant amount of Valentino, J. & Trahair, N.S. 1998. Torsional restraint against
study is still required for torsionally braced beams, elastic lateral buckling. Journal of Structural Engineering
especially for those with multiple bracing points. ASCE 124(10); 12171225.
The analytical solutions for LTB strength and Nguyen, C.T. & Moon, J.H. & Le, V.N. & Lee, H.E. 2009.
Lateral-torsional buckling of I-girders with discrete tor-
the torsional stiffness requirement are successfully
sional bracings. Journal of Constructional Steel Research
developed (Nguyen et al. 2009). This paper presents 66(2): 170177.
equations to determine the critical moment, torsional Chen, W.F. & Lui, E.M. 1987. Structural Stability-T heory and
natural frequency and stiffness requirements of I- Implementation. New York: Elsevier Science Publishing
girders with discrete torsional bracings under uniform C., Inc.
bending moment. For the I-girders subjected to lin- Bazant, P.Z. & Cedolin, L. 1991. Stability of Structures-
ear moment gradient, the equivalent moment factor Elastic, Inelastic, Fracture, and Damage Theories. New
Cb in terms of the number of bracing points is pro- York: Oxford University Press.
posed. The critical moment and the torsional natural Meirovitch, L. 1967. Analytical methods in vibrations.
London: Collier Macmillan Publishers.
frequency are derived for the arbitrary number of
2001. ABAQUS Standard users manual version 6.2.
bracing points using Rayleigh-Ritz method (Chen & Michigan: Hibbit, Karson and Sorensen Inc.
Lui 1987, Bazant & Cedolin 1991) and Lagranges 2001. Manual of Steel Construction, Load and Resistance
equation (Meirovitch 1991), respectively. factor Design 3rd Edition. American Institute of Steel
The proposed equations are then compared with Construction (AISC).
the results of finite element analyses using ABAQUS
531
Reliability evaluation of RC piers of highway bridge using time series

multi-state system approach
W. Shiraki, H. Inomo & M. Matsushima

Faculty of Engineering, Kagawa University, Kagawa, Japan
S.Yabe
Ryusho Industrial Co., Kagawa, Japan
M. Dogaki
1 INTRODUCTION evaluated. Further, their results are compared with the

results obtained by a conventional reliability method
The reliability-based design method has been intro- with considering only binary performance states such
duced as a design methodology of structures in Japan as normally operating or completely break down. The
after the Hanshin-Awaji Big Earthquake Disaster. advantage of the reliability evaluation using a time
In the conventional reliability-based design method, series MSS approach is verified.
however, a structure has only two performance states The time series MSS approach makes it possible to
whether normally operating or completely break down. evaluate the reliability of RC bridge pier RC bridge
In a lot of systems, the change of state happens with pier considering a cycle of performance deterioration
not only two states but also multiple states including a such as initial state, corrosion initiation, craze ini-
partial failure according to the passage of times. The tiation and lost carrying capacity. The conventional
system that considers the multiple states changing with reliability evaluation method cans only two perfor-
time is known as a time series MSS (Multi-State Sys- mance deterioration such as initial state and lost carry-
tem) (Lisnianski, A & Levitin,G. 2003). If such a time ing capacity state. Also, as the MSS approach has
series MSS is introduced in evaluation of structural three reliability indices such as availability, expected
reliability, we may consider the change of multiple per- performance and performance deficiency, diversity
formance levels by the occurrence of crack, progress evaluation of reliability can be possible. These
of corrosion, deterioration of strength, so on. merits take advantage for maintenance problem of
In this study, the reliability evaluation of with mul- structures.
tiple performance levels is performed using a time
series MSS approach. Concretely, multiple perfor-
mance states of piers due to salt damage such as the REFERENCE
occurrence of crack of concrete, progress of corrosion
of reinforcement, deterioration of carrying capacity Lisbiabski, A. & Levitin, G. 2003. Multi-State System Relia-
of member are modeled by the time series MSS, and bility Assessment, Reliability and Engineering Statistics -
then the reliability of piers subjected to seismic load is corrosion Vol.6,World Scientific.
532
Reliability based design optimization of steel box girder bridge under

corrosion attack
N.V. Son, M.C. Jeong, S.H. Kim & J.S. Kong

School of Civil, Environmental and Architecture Engineering, Korea University, South Korea
ABSTRACT
Recently, it is reported that the construction cost of

steel bridge lower than concrete bridge cost. And the
steel bridge structures are developed rapidly in the
world. This trend created a motivation for finding the
optimal solution which minimizes the total construc-
tion cost and maximizes the structural performance.
Reliability-based design optimization (RBDO) is con-
sidered as the integrated analysis of reliability analysis
and optimization algorithms. Total cost of bridge is
expected to minimize under predefined target relia-
bility index that related to uncertainties and RBDO
can be satisfied these conditions. During the service
lifetime, uniform corrosion can occur on the surface
of steel box girder bridge and lead to the structural Figure 1. Optimum total cost versus reliability index for
strength reduction and damage (Czarnecki & Nowak, ultimate and span length.
2008). This paper presents a method to solve RBDO
of steel box girder bridges using a computer program respect to span length from 40 m to 70 m are plot-
which integrates the Matlab optimization toolbox and ted versus the allowable reliability index for ultimate
a reliability analysis subroutine. Moreover, the effect strength in the range from 3.5 to 5 in Figure 1. Consid-
of environmental agents is considered in terms of mod- ering this Figure, the optimal cost steadily increased
eling the uniform corrosion phenomena which occur according to the span lengths and target reliability
on the box surface during the structural lifetime. indices. It can be seen that this results is appropri-
An simply supported PC box girder bridge is con- ated with the target reliability index that suggested
sidered with span length L = 4070 m, subjected to by AASHTO code. Moreover, the proposed method
self-weight dead load and truck live load HL93 follows to solve RBDO of PC box girder bridge by using
AASHTO LRFD code. The results of design variables optimization toolbox and code development is sim-
are shown in Table 1 which X1 , X2 , X3 are thickness ple, flexible, practical and appropriate with the bridge
of top, bottom flange and the thickness of web respec- engineering design and other field.
tively, X4 is the depth of web. The total costs with
REFERENCES
Table 1. Results of optimum design variables. Czarnecki, A. A., Nowak, A. S. 2008. Time-variant reliability
profiles for steel girder bridges. Journal of Strutural safety,
Design 30 (2008), 4964.
variables L = 40 m L = 50 m L = 60 m L = 70 m Kwang-Min Lee, Hyo-Nam Cho and Cheol-Jun Cha. 2006
Life-cycle cost effective optimum design of steel bridge
X1 18.10 18.23 18.40 18.68 considering environmental stressors, Engineering Struc-
X2 25.15 28.28 31.26 34.14 ture Journal, 12521265.
X3 18.24 18.35 18.52 18.77
X4 1704.3 2113.5 2537.7 2976.5
Total cost 390.96 436.7 483.22 530.6
533
Performance-based durability design of weathering steel bridge
E. Yamaguchi
Kyushu Institute of Technology, Kitakyushu, Japan
ABSTRACT the index of corrosion loss rate diminution B, are given.

The JSSC activity has provided formulas of evaluating
In the recent climate of the budge cut for infrastruc- B from A so that the corrosion loss can be estimated
ture investment, the weathering steel bridge is getting once A is obtained. The evaluation of A is then the key
popular in Japan, and in fiscal 2006, the weathering to the assessment of atmospheric corrosiveness.
steel bridge accounts for over 30 mass% of steel con- Two formulas have been proposed for evaluat-
sumed for the construction of steel bridges. However, ing A. The variables of those formulas include not
the development of abnormal rust in the weathering only the effect of the air-born salt but also the other
steel bridge is reported occasionally. The understand- influential factors such as wetness and temperature.
ing of the durability design of the weathering steel But those meteorological data at the specific con-
bridge among bridge designers and bridge owners does struction site are not always available: the data at a
not seem adequate, and the durability design may need nearby meteorological observatory may have to be
be tuned up as well. used instead.
For the durability design of the weathering steel Alternatively, a simple means of evaluating A at the
bridge, it is essential to see if the atmospheric envi- construction site has been proposed. The method has
ronment in which to construct the bridge is not very been termed the button test.
corrosive. To this end, the study based on the long- The button test is essentially an exposure test, using
term exposure test has indicated that such a mild button-shaped weathering steel test specimens. After
environment is defined by the value of the air-born- one-year exposure, the weight losses of the test speci-
salt-deposition rate less than 0.05 mdd (Public Work mens will be measured to yield the site-specific value
Research Institute 1993) and has further specified the of A. The cost for conducting the button test could be
areas of those environments where the weathering steel even lower than that of the air-born-salt-deposition-
is applicable. To that end, Japan is divided into 5 zones: rate measurement.
for each zone, the distance from the sea is specified
individually, beyond which the unpainted JIS-SMA
weathering steel highway bridge can be constructed
without conducting the investigation of the air-born- REFERENCES
salt-deposition rate. Design Specification for Highway Japan Road Association 2002. Design Specifications for
Bridges in Japan (Japan Road Association 2002) has Highway Bridges, Part II Steel Bridges. Tokyo: Japan
employed the distance requirement from the sea in Road Association.
the commentary. The distance requirement appears JSSC 2006. Potentials and New Technologies of Weathering
satisfactory, yet there is room for improvement. Steel Bridges. JSSC Technical Report 73. Tokyo: Japanese
A problem of the distance requirement is its sole Society of Steel Construction.
dependence on the air-born salt while it is not the only JSSC 2009. Applicability Assessment and Prevention of Cor-
factor that influences the rust development. Taking rosion of Weathering Steel in Bridges. JSSC Technical
Report 86. Tokyo: Japanese Society of Steel Construction.
the other influential factors into consideration, more Public Work Research Institute of Ministry of Construc-
sophisticated corrosion-prediction methods have been tion, The Kozai Club (Iron and Steel Mill Products
proposed in the JSSC activity. Association) & Japan Association of Steel Bridge Con-
The corrosion loss of steel can be computed if two struction 1993. Report on Application of Weathering Steel
coefficients, the first-year corrosion loss A (mm) and to Highway Bridges (XX).
534
SS24: Steel bridge rehabilitation
Organizer: M. Sakano
Rehabilitative design/build of a railroad bridge
W.J. Castle
W.J. Castle, P.E. & Associates, P.C.
ABSTRACT
The Roebling Steel Mill located in Florence Town-

ship, New Jersey is part of an EPA Superfund site and
is in the mist of decontamination operations. How-
ever, a creek bisects the property and the only viable
access to the other side is an abandoned railroad
bridge. In October 2008, W.J. Castle, P.E. &Associates,
P.C. (CASTLE) was retained to perform an in-depth
inspection of the bridge to determine the overall con- Figure 1. Typical section.
ditions. The EPA wanted to convert the railroad bridge
into a roadway bridge able to support vehicles up to
110,000 lbs. Our analysis of the site determined that stringers. The voids and/or spalls above water in the
the bridge could be utilized in the proposed design concrete abutments were filled with standard epoxy
with minor repairs required to the components. concrete while the void below water was first covered
CASTLE was awarded the contract for the with composite material before concrete was poured
design/build of the bridge. Upon review of the pro- to fill the voided area.
posed cost of rehabilitation versus the cost of rebuild- For both speed and ease of construction, the
ing from scratch, it was determined that repairing the guiderail posts and all connection hardware were
existing bridge would significantly save both time and attached to the fascia beams prior to be lifted in place.
money. Our initial inspection revealed several small Also the channel diaphragms were attached to one
voids or spalls on the abutments above water and one beam and once the second beam was in place, was
large void below water. Also, the existing steel super- connected to both preventing any potential rolling or
structure exhibited areas of extensive deterioration and displacement of the beam.
corrosion. A full length bearing plate anchored into the con-
Using both our engineering and construction per- crete abutments was utilized due to the minimal
sonnel, we ensured that the designs for the repairs maneuverability.Also, portions of the cheekwalls were
would not only exceed all safety factors and capacities removed to accommodate the new wider spacing of the
but would be easily implemented in the field. beams.
In the original configuration, the six steel stringers All existing steel components were painted with
were closely spaced at the center of the abutments, black epoxy paint except the galvanized repair plates
however the client required a minimum horizontal and guiderails. Construction of the reinforced concrete
clearance of 13 6 to accommodate their vehicles. deck was completed the first week in January and the
This change in layout prohibited the reuse of the exist- bridge was opened by January 23, 2009.
ing diaphragms and bearing plates. While the bridge The total cost of the Project including under-
design was being finalized, the field crew was remov- water inspection with repair to the abutment was
ing the steel superstructure to begin cleaning and $170,000.00 and the Project was completed within 3
repairing of the steel. months.
New steel bent plates were specially designed and
fabricated to reinforce the deteriorated ends of the steel
537
Strengthening design for two-span steel-concrete composite bridges

strengthened by external tendons

Hanyang University, Seoul, Korea
ABSTRACT
In general, existing bridge strengthening as an alter-

native to complete replacement or construct a new
one can provide an effective and economic solution
(Conner et al., 2005). The post-tensioning with exter-
nal tendons has been considered an effective method
of strengthening or rehabilitating existing bridges
(Troisky, 1989). The advantages of this technique
are to enlarge the elastic range of bridge behavior,
to increase the ultimate load capacity of bridges,
and to improve the fatigue and fracture strength
of bridge components (Saadatmanesh et al. 1989a;
1989b; 1989c). In addition, this technique is easy
to perform and convenient to maintain because the
tendons are exposed outside of bridges. For this rea- Figure 1. A plate girder bridge strengthened with external
son, the post-tensioning with external tendons has tendons (unit: mm).
been widely applied to various types of bridges as a
means of strengthening existing bridges (Harajli 1993,
Ng 2003).
The main objective of this paper is to propose a new
rating equation considering the increment of the ten-
don force due to live loads of bridges in order to deter-
mine optimum numbers of strands in external tendons
and the initial tendon force. The post-tensioning with
external tendons is used for strengthening of steel-
concrete composite bridges. Analytic expressions con-
sidering the increment of the initial tendon force are
derived using the virtual work principle for configu- Figure 1 shows longitudinal and cross section of design
rations of straight and draped tendons under external example for plate girder bridge strengthened with
loads. Based on these analytical expressions, a new external tendons.
rating factor equation is introduced considering the
initial tendon force and its increment under external
loads. A systematic procedure is illustrated to deter- REFERENCES
mine the number of strands in external tendons and
the initial tendon force using the proposed rating equa- Conner, R.J. & Fisher, J.W. 2005. Fielding testing of
tion. A design example is also given to demonstrate the orthotropic bridge decks. International Journal of Steel
Structures, KSSC 5(3): 15982351.
effect of the proposed equation on increasing the load-
Harajli, M.H. 1993. Strengthening of concrete beams by
carrying capacity of existing steel-concrete composite external prestressing. PCI Journal 38(6): 7688.
bridges. Eq. (1) and Eq. (2) represent a new rating fac- Saadatmanesh, H., Albrecht, P. & Ayyub, B.M. 1989a. Ana-
tor and initial tendon force proposed in this paper. Eq. lytical study of prestressed composite beams. Journal of
(3) represents the number of required strand (Nt ). Sturctural Engineering, ASCE 115(9): 23642381.
Troitsky, M.S., Zielinski, Z.A. & Nouraeyan, A. 1989. Pre-
tensioned and post-tensioned composite girders. Journal
of Structural Engineering, ASCE 115(12): 31423153.
538
Assessment and corrosion performance enhancement of weathering

steel highway structures
N. Damgaard, S. Walbridge, C. Hansson & J. Yeung

University of Waterloo, Waterloo, ON, Canada
ABSTRACT
Herein, three studies undertaken to investigate sev-

eral issues related to the assessment and corrosion
performance enhancement of weathering steel high-
way structures are presented. The first examines the
effects of corrosion product and local pitting on plate
thickness measurements obtained using a standard
ultrasonic gauge. The second evaluates the potential
of metallizing, as a means for protecting elements
of existing corroded structures. The third study com-
prises a series of reliability analyses of weathering steel
composite box girder overpasses with bottom flanges
corroding uniformly at various rates. Several of these
analyses simulate plate thickness measurement or met-
allizing application events occurring part way through
the service life. Figure 2. Reliability analysis results for urban corrosion.
In Figure 1, mass gain/loss results are presented for
uncoated and metallized weathering steel specimens
subjected to laboratory corrosion testing according to and 40 years for the metalized specimens. Base on
the SAE J2334 regimen (SAE 2002). It is estimated results such as this, it is concluded that metallizing is an
that these tests simulated at least 15 years of service effective way of improving the corrosion performance
in a marine environment for the uncoated specimens of weathering steel structures.
Figure 2 shows the results of a structural reliability
analysis, performed using limit state functions based
on (CSA 2006), wherein the bottom flange thickness of
a 40 m span, simply-supported box girder overpass is
reduced over time using an assumed corrosion model
for urban service conditions. Based on analysis results
such as this, it is concluded that the structural relia-
bility over time is highly dependent on the assumed
corrosion rate. This assumed rate has a very high
level of uncertainty associated with it, making in-situ
thickness measurement a vital part of any structural
assessment.
REFERENCES
Canadian Standards Association (CSA). 2006. CAN/CSA-
S6-06: Canadian Highway Bridge Design Code.
Society of Automotive Engineers (SAE). 2002. SAE J2334:
Figure 1. Weathering steel corrosion test results. Cosmetic Corrosion Lab Test SAE Handbook.
539
Analytical study of fatigue repair of coped beams using carbon fibre

reinforced polymers
T.A. Holden & J.J.R. Cheng

University of Alberta, Department of Civil and Environmental Engineering
C.C. Lam
University of Macau, Department of Civil and Environmental Engineering
M.C.H. Yam
Hong Kong Polytechnic University, Department of Building and Real Estate
ABSTRACT Table 1. Results of case study.
Fatigue and fracture behaviour are important consid- Fatigue %

erations in determining the condition of steel bridge Case Repair n ETR SIF Life Increase
girders. The existing microscopic cracks in girders
NH None 80 N/A 1 86,200
and stringers can propagate into large cracks due to
NL None 40 N/A 1 689,500
repeated loading. The presence of large cracks will SHL1 Long. 80 0.132 0.93 108,100 25%
reduce the load-carrying capacity of the girders. If left SHL2 Long. 80 0.264 0.90 117,600 36%
unattended, unstable crack growth can occur, causing SHA1 Angled 80 0.132 0.78 179,200 108%
catastrophic failure of the structure. SHA2 Angled 80 0.264 0.71 244,000 183%
A fibre reinforced polymer (FRP) crack bridging SLL1 Long. 40 0.132 0.93 864,800 25%
repair eliminates fastener holes or weld stress risers SLL2 Long. 40 0.264 0.90 941,000 36%
that are produced using traditional repair methods. SLA1 Angled 40 0.132 0.78 1,433,800 108%
The FRP repair also has good fatigue resistance, high SLA2 Angled 40 0.264 0.71 1,952,000 183%
strength-to-weight ratio, good durability, corrosion
resistance, and formability to complex contours.
This paper presents an analytical study of the effect conditions. They take into account bridge loadings and
of an FRP repair on the stress intensity factor at the accessibility of the repaired members.
crack tip, and its applicability to use on bridges. The linear elastic fracture mechanics was employed
A finite element analysis study was carried out to to predict the increase in fatigue life of a repaired mem-
determine the effect of an FRP repair on the stress ber based on the repair based change in stress intensity
intensity factor at the crack tip of a fatigue crack. factor at the crack tip.
For this study, a three layer technique developed by Table 1 shows the results of the analytical study.
Naboulsi and Mall (1996). The percentage increase in fatigue life was identical
When compared to the unrepaired specimens, the for the high and low stress ranges. The study showed
average reduction of stress intensity factor (SIF) on the that a longitudinal repair is significantly less effective
patched side was about 61% for specimens with effec- than a perpendicular repair. Provided there is clear-
tive stress ratio (ETR = Efrp tfrp /Es ts ) of 0.16 and 53% ance to apply the repair perpendicular to the fatigue
for specimens with ETR of 0.107. On the unpatched crack, FRP repairs show to be a very effective means
side, SIF showed a maximum increase of 14%. of extending the fatigue life of an aging structure.
An analytical case study was undertaken to evaluate
the effectiveness of an FRP patch applied to a fatigue
crack in the web of a coped beam. A coped beam has REFERENCES
a complex combination of bending and shear stresses
Naboulsi, S. and Mall, S. 1996. Modeling of a cracked metal-
at the cope location. The specimens tested by Yam and
lic structure with bonded composite patch using the three
Cheng (1988) were modelled using the SIF data from layer technique. Composite structures. 35, 295308
the finite element study. Yam, M.C.H. and Cheng, J.J.R., 1988. Fatigue strength of
Three parameters were considered. These were: coped steel beams. Structural Engineering Report No.
stress range (), ETR, and orientation of FRP 160, Univ. of Alberta, Dept. of Civil & Environmental
repair. The parameters were chosen to simulate field Engineering.
540
CFRP repair of fatigue cracks and bonding behavior subjected

to cyclic load during curing
F. Lin, H. Nakamura & K. Maeda

Tokyo Metropolitan University, Tokyo, Japan
H. Suzuki
Meisei University, Tokyo, Japan
T. Irube
TTK Corporation, Tokyo, Japan
ABSTRACT
Recently, many fatigue damages have occurred in steel

bridges because of the influence of the increase in
traffic volume and passing of heavy vehicles. It may
be difficult to repair and reinforce the fatigue cracks
because they are often initiating in narrow locations,
where some welded members are crossing each other.
The Carbon Fiber Reinforced Plastic strip (it is here-
after called CFRP) is recognized to be a useful material
for repair and reinforcement of existing steel struc-
tures, because of its lightweight, high strength and
excellent corrosion resistance. In addition, the boning Figure 1. The relationship between strain and number of
method for CFRP strips is very simple and is also easy cycles.
to use in situ. However, In the repair of steel bridges
in service using CFRP strips, not only steel members,
but also the CFRP strips and epoxy resin adhesives on
repair work are subjected to cyclic stress and vibra-
tion generated by traffic load. The aim of this study is
to investigate for the adhesion characteristics and the
effects of the fatigue crack repair using CFRP strips
and epoxy resin adhesives under dynamic load during
curing.
In this study, investigation was carried out exper-
imentally. First, adhesion behavior under cyclic load
during curing was investigated using coupon speci-
mens of flat steel plates with bonded CFRP strips.
Next, Effects of the fatigue crack repair under the
similar condition were examined using coupon spec-
imens of flat steel plates with a through crack. The Figure 2. The relationship between crack length and number
sinusoidally varying loads were applied to the speci- of cycles.
mens at frequencies of 0.01 and 3.0 Hz for 24 hours.
The curing temperatures were 40 C. on static bond strength by tensile testing. Next, as
As a result, first, as shown in Figure 1, when the shown in Figure 2, in the effects of fatigue crack repair
specimens were subjected to cyclic load during curing, under cyclic load, while the epoxy resin adhesive was
it was observed that epoxy resin adhesives have cured curing, fatigue crack propagated slightly. However, it
in several hours and CFRP strips have been bonded at was found that the fatigue life after curing could be
mean load. It means that the strains on CFRP strips sufficiently improved by high cycle fatigue testing.
were nearly zero at mean load. After unloading, ten- Therefore, the CFRP repair in service conditions was
sile stresses were also introduced to the steel plate. confirmed feasible for practical use.
It was found that the tensile stress did not influence
541
An analytical study on the fatigue cracking of the top flange in steel

railway through truss stringers
K. Matsumoto
Railtech Co., Osaka, Japan
Y. Niwa & T. Nakayama

West Japan Railway Co., Osaka, Japan
M. Sakano
INTRODUCTION
In steel railway through truss stringers, the fatigue

cracks have been detected along the corner of angle
steel in the top flange (see Photo 1). In the previous
study, it is reported that the compressive stress is high
at the both inside and outside of the top flange through
measuring the transverse stress on the back of the top
flange under the sleepers in the stringer. By the results
of previous study, we could expect that both inside and
outside of the top flange were displaced downward, and
we considered that the behaviour of the top flange was
controlled by the contact condition between the sleeper
and the top flange. Although it can be understood that
one side of the top flange is displaced downward, it
has not yet been verified the mechanism of deforma- Figure 1. Transverse distribution of Transverse stress on
top flange.
tion that both inside and outside of the top flange are
displaced downward at the same time.
The principal results obtained through this study are
In this study, the deformation down the both sides
as follows;
of the top flange at the same time is simulated through
finite element analysis. (1) In the case for the flat type or the convex type in the
face shape of the top flange, the inside of the top
flange was deformed down and outside of the top
flange was deformed up. In the concave type, the
both sides of the top flange were deformed down
at the same time.
(2) The mechanism of the behaviour that the both
sides of the top flange were deformed down at
the same time is inferred the sleeper contacted
strongly with the edge of the top flange because
the both edges are higher than the center in the
face shape of the top flange (see Fig. 1).
(3) Stress condition on top flange in stringer can be
simulated by changing the face shape of top flange.
REFERENCES
R. Daito, K. Matsumoto, Y. Niwa, M. Sakano.2008. Out-
of-plane bending behavior of the top flange under sleep-
ers in steel railway though truss stringer. Journal of
Photo 1. Fatigue crack. Constructional Steel in Japanese 16: 703710.
542
A study of longer service life of a multispan simple steel railway bridge
Y. Niwa & T. Nakayama

West Japan Railway Company, Osaka, Japan
K. Matsumoto
Railtec, Osaka, Japan
M. Sakano
A. Koshiba
BMC, Chiba, Japan
ABSTRACT bridge recently. We analyzed tendency of cracks, mea-

sured stress at angle corner of stringer upper flange,
The average use years of steel railway bridges in Japan and examined touch condition between sleeper and
exceeds 60 years, and there are some bridges that is upper flange. By the results of the examinations, it
used for 100 years or more. In the future, concentration is known that out-of-plane bending occurs at upper
of renewal of steel bridges will come caused by cor- flange angle of stringer. And by the tendency of cracks,
rosion, fatigue and so on. Given this situation, bridges it is considered that stringer is a weak point on fatigue.
its total length is long (for example it strides over a So, it is possible that the amount of fatigue crack
wide river) need large cost of operation and main- increases continually. Therefore, we have decided to
tenance. It tends to be also difficult to rebuild those strengthen inspection especially aiming at fatigue and
bridges by some restrictions. Therefore, it is valuable exchange stringer members on a medium-range plan.
to make maintenance scenario of bridge for longer ser- And, we practiced a test construction.
vice life. We selected a bridge striding over a wide In the past, the thought of Scrap & Build was
river in an actual important railway route as a model common and a lot of bridges were rebuilt. However,
case and make several examinations focusing on cor- recently, thought of Longer Service Life of infrastruc-
rosion, function of bearing and fatigue. The outlines tures such as bridge etc. has become major gradually.
of the examinations are as follows. In such a situation, we think this work can contribute
Concerning corrosion, we examined the present to longer service life of the model bridge.
condition of films and steel corrosion, and the level
of corrosive environment (the amount of salt adhered
to members). By the results of the examinations, it REFERENCES
is considered that the environment where the model
bridge locates is not corrosive. While it is possible Hidehiko ABE, Toshio WADA, Kunio ISIGURO & Naoki
that films peel off widely where total film thickness is IWASHITA. 1993. Twist Fatigue of Neck-weld of Rail-
relatively thick or mill scale of steel member is fragile. way Girders. Journal of Structural Engineering. Vol.39A.
pp. 959970. Japan: Japan Society of Civil Engineers.
Therefore, we have decided to examine the condition
Railway Technical Research Institute(ed.) 2005. Manual
of former films and determine surface cleaning level of Painting 2005 -Design and Application for Steel
before every repainting. Structure-. Japan: Railway Technical Research Institute.
Concerning function of bearing, we marked lines to Ryo DAITO, Kentaro MATSUMOTO,Yuichiro NIWA,Taishi
upper bearings and lower bearings of reformed bear- NAKAYAMA & Masahiro SAKANO. 2008. Out-of-plane
ings in 6 spans, and measured movement of bearings by bending behavior of the top flange under sleepers in steel
variations in temperature for a year. By the result of the railway through truss stringers. Symposium on Structural
examination, it is considered that reformed bearings Steel Construction 2008. Tokyo, Japan: Japanese Society
(sliding bearings) move smoothly at present. How- of Steel Construction.
Ryo DAITO, Taishi NAKAYAMA, Takuya KONDO &
ever, its durability is unknown. Therefore, we have
Yuichiro NIWA. 2009. Inspection of Reformed Movable
decided to continue to measure movement of bearings Bearings of Steel Railway Through Truss Bridge. Japan
by variations in temperature twice a year. Society of Civil Engineers 2009 Annual Meeting. 1471.
Concerning fatigue, a lot of cracks have occurred Fukuoka, Japan: Japan Society of Civil Engineers.
along angle corner of stringer upper flange in the model etc.
543
Development of XFEM analysis code for simulation of fatigue crack

propagation in steel structure
K. Shibanuma, H. Aoi & T. Utsunomiya

M. Sakano
Y. Natsuaki
Japan Bridge Association, Osaka, Japan
ABSTRACT
Recently, the fatigue damages of the steel floor systems

of the aged bridges have been reported. It is therefore
urgently required to clarify the cause of the damages
and to rationalize the maintenance such as inspection,
repair and reinforcement.
The numerical simulation of the behaviors of
fatigue crack in the steel floor systems is effective Figure 2. Numerical model of the orthotropic steel deck
specimen.
for the clarification of the mechanism of fatigue crack
propagation and the rationalization of maintenance for
the aged steel bridge.
The XFEM proposed Belytschko and his cowork-
ers (Belytschko & Black 1999) is a numerical method
employs the local enrichment function, which enables
the approximation allowing the reproduction of singu-
larity or discontinuity in the local parts of the domain,
though it is formulated in the framework of the FEM.
As a feature of the XFEM, the crack can be modeled Figure 3. Comparison of the results of fatigue propagation
independently of finite element meshes. Therefore, the path in the numerical simulation and fatigue test.
XEFM has a potential ability to solve the problem
Using this simulation code, the behavior of a fatigue
of vexatious complication in the modeling of crack
crack through thickness of a plate in a three dimen-
propagation using the standard FEM.
sional structure can be quantitatively evaluated. The
In the present study, in order to evaluate the behav-
performance of the developed PU- XFEM code is
ior of a fatigue crack in the local part of large-scale
evaluated through its applications to the numerical
civil engineering structure, we develop a fatigue crack
simulations of the fatigue crack propagation in the
simulation code based on the general-purpose FEM
intermediate floor beam of I-girder bridge and the
analysis software ABAQUS through the implementa-
orthotropic steel deck specimen using bulb rib.
tion of the PU-XFEM approximation (Shibanuma &
It is concluded that the developed PU-XFEM analy-
Utsunomiya 2009). In particular, multiple-nodes are
sis code is useful for the quantitative evaluation on the
introduced to increase nodal degrees of freedom.
path and rate of the fatigue propagation including the
estimation of possibility of crack propagation through
thickness of a plate in a three dimensional structure.
REFERENCES
Belytschko, T. & Black, T. 1999. Elastic crack growth in finite
elements with minimal remeshing, International Journal
for Numerical Methods in Engineering 45: 602620.
Shibanuma, K. & Utsunomiya, T. 2009. Reformulation of
XFEM based on PUFEM for solving problem caused
by blending elements, Finite Elements in Analysis and
Figure 1. Modeling of crack in the PU-XFEM. Design, 45(11): 806816.
544
Development of health monitoring indicators for steel bridges

and its implementation to sensing systems
K. Sugisaki, Mas. Abe & Mak. Abe

BMC Corporation. Mihama-Ku, Chiba, Japan
M. Shimamura
Research and Development Center of JR East Group, Kita-Ku, Saitama, Japan
ABSTRACT
Nowadays, there is a prevailing expectation for Struc-

tural Health Monitoring (SHM) based on real-time and
online evaluation technologies to improve efficiency
and reliability of maintenance of infrastructures. The
target of this paper is to develop a practical SHM
method, i.e. a sensing system which is defined by
indicators based on the structural performance.
According to Maintenance Standards for Railway
Structures in Japan, performance items include a)
safety: load-carrying capacity, durability of fatigue
resistance, running safety and stability, public safety, Figure 1. Representative long-term inclination angle of the
b) serviceability: train operating, running comfort, c) steel bride pier with clinometers and temperature.
restorability. These items are related to the perfor-
mance indicators such as, stress and displacement. Recording of data are conducted 120 minutes by a
Especially displacement is the key parameter for trigger of the amplitude of accelerometer.
health monitoring because it is closely related to run- One of the results, figure 1 shows in a ten-month-
ning safety. Moreover, in design, displacement is the long record of long term inclination of steel bridge pier
indicator which is verified not to exceed the limit state. by clinometers. Inclination of piers shows reaction on
To evaluate strucutural performance, indicator of changing temperature, therefore static inclination has
displacement are useful. But, direct measurement of daily cycle. Focus on the long term trend, daily mean of
displacement generally requires fixed reference point, inclination are described along average daily temper-
which makes it difficult and costly. To implement to ature. Since large inclinations are caused irregularity
sensing system, estimation method for determination of gauges, it keeps up under regular limits. Threshold
of the maximum displacement from inertial measure- value which is calculated from the design limit value
ments such as inclinometer, and accelerometer are of angular rotation on the truck surface is also shown in
needed. To this end, field test has been executed at figure 1. With consideration of seasonal periodicity or
a plate girder railway bridge which has eight spans daily periodicity, it is possible to judge the soundness
with length of 8 22 m. The No.1 span and No.1 of piers form the magnitude relation of them.
pier are chosen. The measurement includes displace-
ment, inclination, acceleration, and temperature. Since
bridges are flexible and displace under various loads, REFERENCES
monitor both dynamic and static displacements are
important. Therefore, our measurements include two Masato Abe,Yozo Fujino: Bridge Monitoring in Japan, Ency-
ways which show below, clopedia of Structural Health Monitoring chapter 125
pp. 21312148, John Wiley & Sons Inc 2009.3
Shot term continuous displacement measurement Yosuke Masui, Osamu Suzuki, Masato Abe, Masahiko
executed for 24 hours, using ring type displacement Samizo, Hiroyuki Furukawa, Makoto Shimamura : Devel-
transducers and other sensors as shown heading 3. opment of Multifunction Scour Monitoring Device, For
Long-term inertial measurement of piers has Railway Bridge Piers, Part 2, The Fourth International
Conference on Scour and Erosion 2008 (ICSE-4 2008)
been executed using inclinometer, accelerometer.
545
Retrofit and advanced investigation on fatigue cracks

penetrating orthotropic steel deck plates
H. Sugiyama, K. Sugioka & A. Tabata

S. Tsukamoto & K. Utsunomiya

ABSTRACT
The crack that penetrates through the deck plate occurs

because the stress concentrates on the weld root. This
concentration of stress is caused because the deck
plate bends locally by the weight of the car. It is seri-
ous damage that causes the pavement damage and the
road cave-in when this crack progresses. However,
orthotropic steel deck with trough rib is difficult to
detect fatigue cracks by visual inspections or Mag-
netic Particle Testing (MT) under the decks without
pavement removal on steel deck plates. Then, it paid
attention to PAUT to be able to expect a high detec-
tion performance as the inspection technique, and the
detection performance was confirmed to the crack of
a real bridge.
Figure 2. Comparison between SFRC pavement and asphalt
The result and the crack investigation imitative (deck side toe between cross ribs).
chart investigated by PAUT to the crack of penetra-
tion through the deck plate are shown in Figure 1. A confirmation in the direction of the crack signal in each
left image shows the bridge axis right-angled direction section of the bridge axis like this, and the unpenetrated
section in the deck plate, and a direction of the deck crack can be presumed to be length. Consequently, all
plate thickness and a horizontal axis show the result of crack length was estimated to be 356 mm. When the
the survey between 35 in investigation angle and 80 shape of the penetration crack that generated it in the
in the spindle in a bridge axis right-angled direction. steel plate deck was compared, it became a result that
The part that enclosed it with a left image shows the was about 7 mm in all crack length shorter. It was able
crack, there is a crack signal up to the board thickness to be confirmed PAUT is to be able to investigate the
12 mm that becomes upper of the deck plate from the crack here in high accuracy.
lower side of the deck plate, and it can be presumed Next, the guss asphalt was exchanged for the steel
the penetrated crack. The penetration crack length fiber reinforced concrete (SFRC) which increases the
of the direction of the bridge axis in the continuous bending stiffness of the thin deck plate against the pen-
etrated crack. Stress reduction effect using the SFRC
was confirmed in the orthotropic steel deck bridge.
The stress measurement measured the strain at
5 mm position from each weld toe for the stress etc.
of a deck plate and trough rib weld surrounding. The
load was executed with the test car adjusted to 245 kN
in total weight. One example of the result is shown in
Figure 2. According to this, the stress range reduction
to about 16% by the maximum value in the weld deck
plate side toe of the deck plate and trough rib.The stress
range reduction of about 50% was confirmed between
cross ribs. As for the stress range in the deck plate weld
toe at a cross rib reduced more than 50%. The improve-
ment of the fatigue durability by the SFRC pavement
Figure 1. Example of sector scanning image of penetrated can be expected for the penetrated crack through the
crack. deck plate.
546
Study on improvement of fatigue strength of out-of-plane gusset welded

joints by attaching GFRP
H. Suzuki
Meisei University, Hino, Tokyo, Japan
N. Inaba
Central Nippon Expressway, Numadu, Shizuoka, Japan
Y. Tomita
East Nippon Expressway, Chiyoda-ku, Tokyo, Japan
K. Shito
East Nippon Expressway, Misato, Saitama, Japan
ABSTRACT
In this study, in an attempt to examine the effects of

GFRP on the improvement of the fatigue crack ini-
tiation life Ni and fatigue crack propagation life Np,
fatigue tests were conducted for out-of-plane gusset
welded joints to which GFRP were attached using the
primer. The effectiveness of GFRP for reducing the
stress at the toe of round weld was also investigated
analytically. The effectiveness of GFRP for controlling
the fatigue crack propagation rate was also considered
from a viewpoint of fracture mechanics.
The dimensions of the specimen are shown in
Figure 1. Figure 2 shows S-N curves indicating the
regression curves of respective specimens. The regres-
sion lines of GN and GF were close to each other in Figure 2. S-N curves (regression curves).
the high stress range. The fatigue life of GF was much
longer than that of GN in the lower stress range. In (1) It was confirmed that fatigue life could be
the high stress range, therefore, attaching GFRP may improved by attaching GFRP to the round weld
not be so effective as expected for strengthening. It of the out-of-plane gusset welded joints.
has, however, been confirmed that attaching GFRP is (2) The improvement of fatigue life through the
effective in the lower stress range. attachment of GFRP using the primer was
The conclusions obtained are described below. attributed in this study to the reduction of the crack
propagation rate and the improvement of crack
propagation life by GFRP.
Figure 1. Specimens.
547
Study on improvement of the fatigue durability by filling of mortar

in U-shaped rib of orthotropic steel deck
A. Tabata, Y. Aoki & Y. Takada

H. Onishi & S. Matsui

Osaka University, Osaka, Japan
ABSTRACT between U-shaped rib and deck plate were shown

approximately 30% on the deck plate side of the
This paper reports to study on the method of retrofitting center of the U-shaped rib, also U-shaped rib
orthotropic steel deck without need for traffic restric- web was about 40% comparison with the original.
tion for eliminating fatigue cracks originating from Especially It was remarkable the stress reduction
the weld between the deck plate and U-shaped rib and effective of the intersection of the U-shaped rib and
preventing fatigue damage. We proposed the reinforce- crossbeam (Figures 2, 3).
ment method as filling of mortal in U-shaped rib and 2. As a result of the wheel load running test for the
installation of the splice plates between the U-shaped non-retrofitting model of the specimen, it was con-
rib as shown in Figure- 1. The static loading test and the firmed the crack at the intersection of the U-shaped
wheel load running test were carried out in order to ver- rib and deck plate which occurred the non-welding
ify the fatigue durability and the stiffening effects by point of the longitudinal welding of the U-shaped
the retrofitting. The main conclusions and discussions rib and deck plate at 600,000 cycles of the load
are as follows, running.
3. It was confirmed that the stress decrease effect of
1. As for the result of the static loading test, the
the measurement points due to the reinforcement
stress reduction effective ratio due to reinforce-
are stabilized to the test end respectively, in addi-
ment for local stress of neighborhood welding
tion, the crack on the deck plate after the test ended
doesnt have much progress comparing with the
original(Figure 4).
Figure 1. Retrofitting method.
Figure 3. One example for stress influence line.
Figure 2. Location of stress measurement. Figure 4. One example for change of stress.
548
Study of fatigue assessment of orthotropic steel decks in consideration

of variations in locus distribution of wheels
Y. Takada
Hanshin Expressway Management Center, Osaka, Japan
A. Tabata
Hanshin Expressway Corporation, Osaka, Japan
T. Yamaguchi
Osaka City University Professor, Department of Civil Engineering, Osaka, Japan
ABSTRACT ni : Number of repetitions of stress range

m: m = 3 grade of a fatigue design curve
Recent considerable increases in traffic intensity and
wheel loads are causing fatigue cracks in orthotropic In Fig-1, lane space marks are made into the start-
steel decks. Under traffic loading, in particular the ing point(x = 0), and the location of a run wheel is
effect of local wheel loads, longitudinal welds between expressed with LT . The stress range generated in target
deck plate and trough are subjected to local transverse weld joint located in x = LT can be expressed with the
bending moments and are susceptible to fatigue cracks. function (LT (x)) of the load location of transverse
The stress in trough to deck plate welds is strongly direction. That is, LT (x) is an influence line.
influenced by actual-working load and run location. Next, the equivalent stress range in consideration
Then, in consideration of the variation in locus dis- of run frequency distribution of vehicles is examined
tribution of a wheel, the method of carrying out fatigue [1]. Instead of stress range which acts on the joint in a
assessment of orthotropic steel deck analytically are formula (2) LT (x), is used.
studied. In this study, the stress function is computed The frequency function of formula (3) is used
by the influence-line load of the vehicles load of a instead of the number of repetitions of a stress range.
transverse direction. Locus distribution of the wheel is
given with the frequency function of the normal distri-
bution.The equivalent stress range in consideration of
Therefore, the equivalent stress range in consideration
this locus distribution is calculated based on the linear
of the distribution of the run location of the transverse
damage rule, and fatigue assessment are performed.
direction of a wheel to the location of welding by which
examination by reference is carried out is given by the
1 MODELING OF RUN FREQUENCY
formula of the 3rd square average of the integration of
DISTRIBUTION
the frequency function of a normal distribution, and a
stress function.
Distribution of the run location of the transverse direc-
tion of vehicles can be expressed with a normal dis-
tribution. The frequency function is given by average
and standard deviation by the following formula. Where,
eLT : The equivalent stress range in consideration
of the variation in the run location of the
transverse direction of a wheel.
Where,
: Standard deviation of frequency distribution of With reference to the formula (4) mentioned above,
a run location. calculation of the equivalent stress range uses numer-
: distance from left side lane space marks end to ical integration to a generating stress range. By
the maximum frequency location of centre of the the calculation program, the smoothing interpolation
wheel in slow lane. function which is the 3rd equation is determined from
The equivalent stress range e which is a stress which the data of a generating stress range, and numerical
does equivalent fatigue damage with the same number integration is repeated in the section length of 1 mm.
of repetitions as a change amplitude stress is given by
the linear damage side by the following formula. REFERENCE
[1] Yoshihiko Takada ,Akiko Tabata, Examination in con-

sideration of distribution of the tire locus of a fatigue
Where, evaluation, Jsce, 1-069, pp. 137138, 2009.9
i : Stress range which acts on a weid joint (MPa)
549
Experimental study on the root fatigue life improvement for the reasonable
design of steel pier seismic reinforcement
T. Tominaga & K. Matsuoka

Nippon Steel Corporation, Chiba, Japan
C. Miki
Tokyo Institute of Technology, Meguro, Japan
Y. Hirabayashi
Metropolitan Expressway Corporation, Chiyoda, Japan
ABSTRACT
Recently, many cracks have been found at the cor-

ners of beams to column connections of steel bridge
piers. The majority of these cracks are fatigue cracks
caused by incomplete penetration defects, as well as
weld cracks related to those defects.
Because of the complexity of plate arrangements
at beams to column connections of steel bridge piers,
there exist zones where complete penetration is dif-
ficult, especially where welding lines intersect from
three different directions (Miki et al. 2003). As a result,
incomplete penetrations and weld cracks are often
formed at the corners of beams to column connections
of steel bridge piers.
For the current retrofitting to such damaged areas,
the method of large coring has been developed and
applied. This is a method of removing damaged zones,
in which cracks and weld defeats converge, using a
100 mm diameter large coring, after reinforcing the Figure 1. Description of a large coring method and bolted
beams by attaching bolted plates. This can improve splice reinforcement.
the fatigue life of the detail efficiently in a short
construction time. fatigue life of the detail with open cracks after large
However, this method leaves a difficulty: weld roots coring. Since the likely crack initiation point is an open
are exposed after removing the area with defects by root, this study proposes several new treatments to the
coring. While weld roots, where present, are always open root and experimentally examines their effects
closed in welded steel structures, in the method of large on the structural improvement of the fatigue life.
coring incomplete penetration defects are visible from In this study, root edge drilling, pin shooting and
outside. In this case, the stress concentration at the cover welding are tried to improve root fatigue perfor-
edge of the defect is larger compared to the case of mance. And all these method are conducted together
the closed space. Also the corner of a beam to column with ultrasonic impact treatment to control residual
connection is the point at which working stress from stress at surfaces and toes. Each of these meth-
dead and live loads is very large, and coring increases ods showed fatigue improvement, but especially root
this working stress. Since stress concentration is usu- drilling with cover welding was efficient.
ally larger than before and working stress can still be
large, there is a risk of the return of root cracks. And
they can induce brittle fractures during earthquake. REFERENCE
But if the fatigue performance of the open root Miki C. & Hirabayashi Y. & Tokita T. & Konishi T. &
could be improved, it would be possible to simplify Yaginuma Y. 2003. Beam-column connection details
or even omit the reinforcement by bolted plates in the of steel pier and their fatigue damage mode, Jour-
case of small working stress. This study, thus, aims at nal of structure mechanics and earthquake engineering,
realizing reasonable seismic design by improving the No. 745/1-65, pp. 105119 (in Japanese).
550
Effect of reinforcing method against fatigue cracking of orthotropic steel

deck with bulb ribs
D. Yamaoka & M. Sakano

Y. Natsuaki
Japan Bridge Association, Osaka, Japan
S. Nonaka, Y. Nakagawa & K. Nakamura

Kinki Regional Development Bureau, Ministry of Land, Infrastructure and Transport, Osaka, Japan
ABSTRACT
Recently, thousands of fatigue cracks have been

detected in orthotropic steel decks in Japan. Of these,
fatigue cracking in welded joints between bulb ribs and
lateral ribs is the most frequent type found in the Kan-
sai area. Fatigue cracks are classified into 4 types, as
shown in Fig. 1 (Tabata et al. 2007). At the intersection
of the bulb rib and the lateral rib, the crack that propa-
gates from the weldment of the lower part of the slit into
the lateral rib web is d-type, and the other that propa-
gates from the weldment of the upper part of slit into
the deck plate is a-type. These cracks may propagate
into the deck plate and have a bad influence on traffic.
It is important, therefore to grasp such fatigue
cracking behavior. An effective method against the Figure 2. Intersection of bulb rib and lateral rib with angle
fatigue cracks is needed. In this study, we tried to steel reinforcements.
grasp fatigue cracking behaviour in the welded joints
between the bulb rib and the lateral rib by means of
fatigue tests of the orthotropic deck plate specimen
which is the same size and has the same structural
detail as the actual bridge. Also, we verified the effect
of the proposed reinforcing method using angle steels,
as shown in Fig. 2.
Fatigue cracks were initiated at the weld toe of
the upper part of the slit and propagated through the
weldment into the deck plate, as shown in Photo 2. We
confirmed that the fatigue crack detection life of the
Photo 1. Fatigue crack at upper part of Rib 1 slit after 0.7 M

cycles loading.
welded joint between the bulb rib and the lateral rib
was improved more than eight times by applying angle
steel reinforcement.
REFERENCE
Tabata, S., Yamamura, K., Hamada, N., Sakota, H., Sakai,
Y. and Sakano, M. 2007. Experimental study on rein-
forcing method against fatigue damage at welded joint
between bulb rib and lateral rib in orthotropic steel deck,
Figure 1. Fatigue cracks in welded joint between bulb rib Proceedings of 62th Annual Meeting of JSCE, I-003 (in
and lateral rib in the Hanshin Expressway. Japanese).
551
SS25: New developments in bridge design codes
Development and calibration of new reliability-based bridge design

code in Korea
E.-S. Hwang
Kyung Hee University, Yongin, Korea
I.R. Paik
Kyung Won University, Seongnam, Korea
H.-M. Koh
Most internationally-renowned design codes for

bridges have implemented limit state concepts includ-
ing AASHTO and Eurocode. Since the first bridge
design code was introduced in 1962 in Korea, reg-
ular revisions have been made in accordance with
new social and technological needs. However, to meet
with global standard, state-of-art technology and fast-
changing local environments, new bridge design code
is being developed based on reliability or limit states
design concept.
This paper deals with the development of reliability-
based bridge design code and calibration of load and
resistance factors based on new live load model. To
establish statistical characteristics of live loads, truck
weights and traffic patterns information are collected
using WIM system and video recording. Based on
probabilistic analysis, new live load model and their
statistics are determined. Statistical data on strength
of concrete, reinforcing bars and steel members are
collected and used in evaluation of member strength.
Reliability analysis is performed on various types of
bridges. Comparisons are made with current design
code to prove the validity and effectiveness of pro-
posed design code. It can be concluded that proposed
live load model and proposed load and resistance fac-
tors give more consistent reliability indices for various
load effects and span lengths.
Figure 1. Evolution of design code for bridges. Figure 2. Comparison between proposed and current .
555
A study on ductility of high strength steel bridge girder
H.S. Joo, S.H. Cha & H.E. Lee

PhD Candidate, School of Civil, Environmental & Architectural Engineering, Korea University, South Korea
B.H. Choi
Professor, Department of Civil Engineering, Hanbat National University, South Korea
ABSTRACT
It is possible to apply the high-performance

steels(HPS) to practical design not only light weight
structures but also simple structures with simple
weld details. However, from the results of previous
researchers, it is found that the ductility of girder with
HPS is reduced due to the increasing of yielding stress.
Thus, the study on method to guarantee the proper
ductility of girder with HPS is necessary. This paper
investigates the flexural ductility of I-girders in nega-
tive moment region with HSB800(Fy = 680 MPa).The
flexural strengths of bridge girder which assumed to be
made by using 800 MPa class high performance steel
are evaluated based on AASHTO LRFD design code. Figure 1. Results from FEM.
Concrete deck cannot resist in negative moment
regions due to the fact that the tensile stress is reached
in the upper flange. Therefore, negative moment
regions are designed by concept of noncompact sec- REFERENCES
tion. Moment envelope of negative moment region is
very sharp and higher than positive moment region. AASHTO LRFD Bridge Design Specifications (2007)
Therefore, plastic hinge negative moment region can Felkel, J.P., Rizos, D.C. and Ziehl, P.H. (2007). Structural
be generated easier than that in positive moment performance and design evaluation of HPS 70W bridge
region. Thus, it is necessary to provide the sufficient girders, Journal of Constructional Steel Research 63
ductility for negative moment region. (2007) 909921.
Earls, C.J. and Shah, B.J. (2002), High performance steel
This study prescribes an alternative bracing system bridhe girder compactness, Journal of Constructional
to ensure sufficient flexural ductility. The alternative Steel Research 58 (2002) 859880.
bracing system does not require any additional cross White, Donald w. and Barth, Karl E. (1998) Journal of
beam members. The alternative bracing system needs Constructional Steel Research 45 (1998) 241280.
to move the cross beam members to near the bridge Yura, Joseph A., Galambos, Theodore V. and Mayasandra
pier. From the result of A800E model, despite of the K. Ravindra (1978), The Bending Resistance of Steel
satisfaction of AASHTO LRFD(2007) compactness, Beams, Journal of the Structural Division, Vol. 104, No.
the girder cannot attain the desired rotational capac- 9, September 1978, pp. 13551370.
ity. However, By using alternative bracing system, it American Society of Civil Engineers. In: Plastic design in
steel, a guide and commentary. New York: American
is found that the rotational capacity of the girder is Society of Civil Engineers; 1971, p. 80.
increased compare with ordinary bracing system. 2001. ABAQUS Standard users manual version 6.2. Michi-
This study also prescribes Optimum using yield- gan: Hibbit, Karson and Sorensen Inc.
ing range when girder is reached on plastic moment. 2001. Manual of Steel Construction, Load and Resistance
The girder have sufficient rotational capacity when factor Design 3rd Edition. American Institute of Steel
cross beam is placed in optimum . Construction (AISC).
556
Development of the AASHTO guide specifications for bridges vulnerable

to coastal storms
J.M. Kulicki
Modjeski and Masters, Inc.
ABSTRACT force and associated horizontal force and moment

data compiled. This data was then used to develop
In response to the need to learn more about how parametric equations for the various components of
to design and retrofit coastal bridges subjected to wave forces. The resulting equations are complex but
coastal storms such as hurricanes, the FHWA autho- amenable to solution by spreadsheet.
rized a research project entitled Development of A three-level approach was adopted in order to ana-
Guide Specifications and Handbook Retrofit Options lyze the wave forces for a given situation based on 100
for bridges vulnerable to coastal storms. This paper year design conditions. Level 1 involves a conservative
summarizes work that culminated in the 2008 adop- evaluation of the situation using available data for
tion of AASHTOS Guide Specifications for Bridges wind and storm water height. The appropriate sources
Vulnerable to Coastal Storms. for data for the Level 1 analysis would be either site-
A literature study undertaken early in the project specific information on astronomical tide, wind and
showed that while many investigators have studied surge, or the use of the ASCE 7-05 wind maps adjusted
the action of waves on structures, most of the work for 100 year return period in combination with FEMA
concentrated on the vertical components of the struc- storm surge maps. A Level 3 analysis is much more
tures, or structures that differ greatly from a bridge thorough and involves the use of a number of sophis-
superstructure.The existing research on superstructure ticated storm surge and wave computer models and
components concentrated on offshore environments, must be performed by qualified coastal engineers. A
where the design wavelength is much greater than the Level 2 analysis falls somewhere between Level 1 and
length of the structure. Level 3 analyses and considerable latitude is permitted
Physical model tests were conducted in a 6 ft wide, in the choice of how far between Level 1 and Level 3
6 ft deep, 130 ft long wave tank at the University the engineer believes is appropriate for the particular
of Florida. Numerous plots were developed showing situation.
the comparison of wave forces predicted by various Two limit states are recommended for the design
methods and the results of the wave tank experi- of bridges vulnerable to coastal storms: strength and
ments. The basis for a choice of a method to be extreme event. Bridges which are designated as crit-
proposed in the Guide Specification involved evalu- ical/essential by the owner should be designed at
ation of the predicted forces compared to the physical the Strength Limit State to achieve a performance
model test data, the prediction of failures on the level of Service Immediate, or Repairable Dam-
I-10 Escambia Bay Bridges, theoretical complete- age, depending on how quickly they must be passable
ness, and practicality. Based on an evaluation of all after the storm.
these factors, it was decided to proceed with the Service Immediate can be taken to mean that the
Physics Base Model (PBM) developed by Ocean Engi- bridge may be assumed to be sufficiently undamaged,
neering Associates, Inc. (OEA) method. The PBM, stable and aligned to be usable for rescue and recovery
which can be applied to both submerged and sub- forces after a cursory inspection.
aerial structures, had great appeal due to its theoretical Repairable Damage can be taken to mean that
completeness. some repairs could be needed to restore sufficient ser-
The PBM was run for a range of span configurations viceability to put the bridge back in limited use within
and elevations (relative to the storm water level), water the owners criteria for outage duration and after an
depths, and wave conditions and the maximum vertical inspection.
557
Approach for developing calibrated service limit states for the AASHTO
LRFD bridge design specifications: A progress report
J.M. Kulicki & W.G. Wassef

Modjeski and Masters, Inc.
A.S. Nowak
University of Nebraska
D.R. Mertz
ABSTRACT determine how much service life remains in an

existing, possibly deteriorated, bridge, and
The notion of limit state, the boundary between accept- quantitatively determine the effect of retrofit, repair
able and unacceptable performance, is fundamental or maintenance on remaining life,
in the AASHTO LRFD Bridge Design Specifications
(AASHTO LRFD). all with a specified certainty or probability.
The current service limit states of AASHTO LRFD To achieve the objective of developing the appro-
are intended to ensure a serviceable bridge for the priate tools, candidate service limit states have to be
specified 75-year design life. These limit states are evaluated against a set of criterion including:
based upon the traditional serviceability provisions
of the AASHTO Standard Specifications for Highway Is the limit state quantitatively and qualitatively
Bridges. However, they are not calibrated using reli- meaningful?
ability theory. This paper describes ongoing work to Can the limit state be calibrated?
develop calibrated service limit states under project Does a limit state really relate to the service life
SHRP2 R19B. rather than to some other characteristic?
Among others, the current service limit states in Does it provide a method to evaluate the significance
AASHTO LRFD include limits on live load deflection of interventions in extending the service life of the
of bridges, cracking of reinforced-concrete compo- structure component?
nents, tensile stresses of prestressed-concrete com-
ponents, compressive stresses of prestressed-concrete An acceptable probability of exceeding a SLS is
components, permanent deformations of compact much higher than for ULS. If the target reliability index
steel components, slip of slip-critical friction bolted for ULS is T = 3.5 to 4.0, then for SLS, T = 0.0 to
connections, and settlement of shallow and deep foun- 1.0 may be shown to be appropriate.
dations. Some of these service limit states may relate Given a target reliability index, the anticipated
to a specified design life; others do not. Many are calibration procedure is as follows:
presently very deterministic. One of the major objec-
Select representative location characteristics such
tives of the reported work is to develop the tools
as climate.
necessary to:
Select representative traffic characteristics.
design a new bridge to achieve a specified service Identify the design parameters.
life approaching or in excess of 100 years, this Provide for user prediction of performance and
will require identification of service limit states and deterioration.
specification provisions to implement them for a Select the acceptability criteria, i.e., performance
minimum 100 year service life, and development of parameters that are acceptable, and those that are
performance measures for calibration processes, as not.
well as to monitor the performance of new bridges Calibrate the load and resistance factors to meet the
designed with the new service limit state provisions acceptability criteria.
and to gage the performance of older bridges for the Review the developed design parameters, and make
purpose of bridge management. necessary adjustments.
558
Effect of tension stiffening in composite bridges in the light of Eurocodes
G. agoda & W. Radomski

Warsaw University of Technology, Warsaw Poland
M. agoda
Lublin University of Technology, Lublin, Poland
Road & Bridge Research Institute, Warsaw, Poland
ABSTRACT Region a corresponds to the stage when the con-

crete bridge slab is un-cracked, region b corresponds
One of the crucial problems concerning the design to the stage when the initial crack in the slab is
calculations of the composite bridges with continu- formed, and region c corresponds to the stage when
ous structural system is to take into account the cracks the cracked pattern in the concrete slab is stabilized.
which can occur in the concrete deck slab in the regions For each of the above stages the formulae concerning
of negative bending moments (hogging moments). The the relation between the normal force in the con-
above problem can be considered ac cording to the crete slab and the strains in the concrete and its steel
procedure proposed in the Eurocode EC 4, EN 1994-2 reinforcement are given.
(2005). Next, the internal forces in the composite section are
The Eurocode 4 allows to consider the stiffness analyzed without and with the effect of shrinkage. The
of the composite girder with the cracked concrete cracking moment is defined in particular in the men-
slab deck using the diagram normal force strain tioned above regions band c. The additional normal
for tensioned reinforced concrete structural members. force of the concrete section due to tension stiffening
Three fundamental regions of the above diagram can is also given in two fundamental cases without and
be differed, denoted by a, b and c (Fig. 1). with prestressing the concrete slab by tendons.
Finally, the contribution of concrete in tension
between the cracks in the stiffness of a composite
cross-section is presented using the relevant formulae.
As a one of the conclusions, it is emphasized that
the effects of tension stiffening in composite steel and
concrete bridges proposed in EC 4-2 (2005) leads to
the more economical solutions compared to the design
procedures applied so far in many countries, including
Poland. The relevant procedure can be applied both for
designing process of the new composite bridges and
for strengthening of the existing ones where taking
into account tension stiffening of the cracked concrete
slab allows to obtain more economical solution of the
strengthening itself.
REFERENCES
CEN. En 1994-2, 2005: Design of composite steel and
concrete structures. Part 2: Composite bridges.
Gomes Nawarro M., Lebet J.P. 2001 Concrete cracking in
composite bridges: tests, models and design proposals.
Structural Engineering International, 11 (3), 184190.
Figure 1. Normal force vs. mean strain for reinforced

concrete tension member.
559
SS26: WIM-based live loads for bridges
Extreme truck load effect prediction for bridge structural reliability
G. Fu
J. You
ABSTRACT Over the past two to three decades, the economic

development in China has nurtured the establishment
Structural design and evaluation are moving towards of a highway network with a large number of bridges.
reliability based practice. The latest building and However, there are still no nationwide specification
bridge design specifications in the US are exam- provisions for assessing their safe load carrying capac-
ples of this trend. In developing these specifications, ity. The research work reported herein focuses on
risks involved with major design factors were mod- developing reliability based requirements for this pur-
eled as random variables, such as the live load for pose. In this study, weigh-in-motion data for more
the structure, the resistance of structural members, than 7.3 million trucks were gathered from highways
the environmental loads due to the wind and flood, in three provinces of China, continuously over one
etc. The load factors and resistance factors were then to 16 months in 2006 and 2007. The data were pro-
derived by controlling the failure of concern. On the cessed and projected to model the live load spectrum
other hand, this first generation of reliability based over 3-year and 100-year periods, respectively.The for-
specifications still has ample room for improvement. mer is the required bridge inspection interval and the
This paper focuses on one of such issues relevant to latter the bridge design life span, according to cur-
maximum load projection for the intended life span of rent Chinese maintenance and design specifications.
the structure applicable to deriving the truck live load The proposed projection method is shown to be more
model and its load factor for highway bridge design reliable compared with those reported. The resulting
and evaluation codes. load spectra are used to assess the structural reliability
Using short term load data to extrapolate or project of typical Chinese highway bridges at the component
to remote future maximum load is often exercised level. Based on the accordingly selected target relia-
for specification calibration and development. It was bility index, the live load factors for bridge evaluation
indeed done for the AASHTO LRFD bridge design are developed in this study, proposed to be included in
specifications. However, this approach has not been the Chinese national specifications.
adequately evaluated or validated due to lack of long
term data, which is important for the structures safety REFERENCES
governed by these codes. This issue is focused here
for the application to highway bridge design and eval- Fu, G. Variance Reduction by Truncated Multimodal
uation, taking advantage of available long term truck Importance Sampling, Structural Safety, Vol.13, 1994,
weight data obtained using the weigh-in-motion tech- pp.267283
nique. A new extrapolation method is proposed here Kulicki, J., Prucz, Z., Clancy, C.M. D.R. Mertz, and Nowak,
for more reliable projection, based on the understand- A.S. Updating the Calibration Report for AASHTO
ing of importance of the load probability distribu- LRFD Code Final Report for NCHRP 20-7/186, Jan.
2007
tions high tail. This approach significantly reduces
Nowak, A.S. Calibration of LRFD Bridge Design Code,
the mathematical length of extrapolation/projection NCHRP Report 368, Transportation Research Board,
and thus increases its reliability, also shown herein by National Academy Press, Washington D.C. 1999
application examples. van de Lindt,J., Fu,G. Locality of Truck loads and Ade-
This method is also applied here to develop highway quacy of Bridge Design Load, ASCE Journal of Bridge
bridge evaluation requirements for practice in China. Engineering, Sept/Oct. 2005, p.622
563
US bridge formula (FBF-B) and implications of its possible

application in Europe
B.A. Jacob
Laboratoire Central des Ponts et Chausses (LCPC), Paris, France
M. Bereni
Service dEtude sur les Transports, les Routes et leurs Amnagements (SETRA), Bagneux, France
M. Ghosn
City College of New-York, New York, United States
B. Sivakumar
HNTB, One Penn Plaza, New York, New York, United States
T. Kearney
Federal Highway Administration, New York, United States
ABSTRACT 44 t in Europe. Moreover, some European countries

already allowed or experimented with European Mod-
Bridges are designed using conventional load models, ular System (EMS) vehicles with a total length of
such as those defined in the AASHTO and Ontario 25.25 m and a mass limit of 60 t.
bridge codes or the Eurocode 19912, to ensure that To properly manage the changes in HGV weights
bridge members and elements are able to support and dimensions without increasing their damage to
the effects of service trucks and exceptionally loaded existing bridges, it is important to maintain some
vehicles including those due to single truck crossings form of the FBF-B formula to limit the weight-to-
and overtaking events as well as the corresponding length ratio, but the existing formula may require some
fatigue stress cycles To ensure the safety of existing up-dating after 35 years. It is also proposed to inves-
bridge and highway networks, gross vehicle, single tigate the possibility of implementing the FBF-B to
axle and multiple (tandem, tridem) axle set limits the European HGVs truck fleet and study whether it is
are enacted in State or Federal laws (e.g. the 1956 applicable to limiting the damage on typical European
Highway Act) or International directives (e.g. Euro- bridge shapes and topologies.
pean directive 96/53EC). In particular, in response to An investigation is being carried out to study the
the dramatic increase in heavy trucks that occurred in damage caused by the most common US and Euro-
the late 50s and early 60s, and out of concern over the pean trucks to existing bridges for both maximum
damage caused by heavy, single unit trucks, a bridge load effects and fatigue damage. Maximum bending
formula was developed in the US which accounted for moments at mid-spans and on piers, as well as shear
the different levels of road damage caused by differ- forces of simply-supported spans, and continuous 2
ent vehicle configurations and weights including the and 3 span bridges, with span lengths from 10 to 100 m
number of axles and their spacing. In 1974 the Fed- were considered. For each truck n with its maximum
eral Bridge Formula-B (FBF-B) was passed into law to permitted weight Wn, and a weight limit imposed by
ensure that the truck fleet did not produce overstresses the FBF-B of Wbfn, a load coefficient was calculated
on the representative bridge network which was dom- as: cn = Wn / Wbfn. For each truck (and weight Wn)
inated by simply supported superstructure at the time. and each load effect, a damage coefficient Kn with
No change has been made to the US Federal Bridge respect to a reference truck r was calculated, as the
Formula or federal gross and axle weight limits since ratio of the maximum load effect induced by the truck
1974, while the truck fleet, design and volume have n and the reference truck r. Then Kn was compared to
dramatically evolved. the ratio Cn = cn / cr, to assess the homogeneity of the
To reduce GHG and CO2 emissions, road conges- FBF-B weight limit. It is shown that the US bridge for-
tion, and to satisfy the increasing demand of freight mula is adapted to simple supported spans of 20 m and
transport and increase road transport efficiency, studies two continuous spans of 10 m each. Fatigue damage
are carried out in the US and in Europe on the bene- under stress cycles is also considered and a similar
fits and disadvantages of increasing the Heavy Vehi- approach is developed to assess the ability of the FBF-
cles weights and dimensions, and thus their capacity. B to limit fatigue damage. Finally, some proposals are
There is a strong demand to extend the general gross made to adapt the FBF-B for Europe.
weight limit up to 97,000 lbs in the US and to
564
Live load models based on weigh-in-motion data
M. Kozikowski & A.S. Nowak

University of Nebraska, Lincoln, Nebraska, USA
ABSTRACT be discarded. Based on the sensitivity analysis per-

formed on various sites it was observed that removal
The Weigh-In-Motion measurements provide the unbi- of only 0.03% of all trucks from the top of the dis-
ased truck traffic data and it is a rational basis to tribution can cut the maximum load effect by 32%.
develop the statistical model of live load for a code This can lead to the conclusion that all trucks in the
calibration. The objective of this paper is to present filtered database have a great importance in live load
the development of statistical models for live load that prediction and correct filtering criteria are needed.
is applicable to the AASHTO LRFD code calibration. Multiple presence and degree of correlation analy-
The database presented in this study includes truck sis showed that the time of record of the passing truck
records from four states: Florida, Indiana, Mississippi, has to have 0.01 second accuracy otherwise it is dif-
and New York. The total number of trucks exceeds 35 ficult to determine the accurate headway distance. A
millions. For each location, the WIM data represents correlation analysis performed on the available new
one year of truck traffic. For each measured vehicle, the WIM data confirmed previous assumption that about
recorded information includes the gross vehicle weight every 500th truck is on the bridge simultaneously side-
(GVW), number of axles, load per axle, axle spacing, by-side with another fully correlated truck. However,
lane of travel and speed. The raw data was filtered and the expected maximum weight of the fully correlated
preprocessed to ignore any errors in weight per axle, trucks is smaller than the maximum weight of trucks
spacing between axles and speed. An additional filter recorded at the same site. Therefore, a load combi-
was implemented to verify the class of the vehicle. For nation with two fully correlated trucks in adjacent
all truck, the maximum load effect was obtained using lanes does not govern. The governing combination is a
structural analysis procedures. The results were extra- simultaneous occurrence of the extreme truck and an
polated up to 75 year maximum values.The cumulative average truck.
distribution functions of GVW and live load moments
were plotted on the normal probability paper. The plots
show a large variation that indicates that the live load REFERENCES
is a strongly site-specific. For comparison, the truck
survey data used in calibration of the AASHTO LRFD Castillo, E., 1988. Extreme Value Theory in Engineering.
Academic Press, San Diego.
Code is also presented.
Nowak A. S., and K. R. Collins, 2000. Reliability of Struc-
WIM stations are usually hidden from the truck tures, McGraw-Hill,
driver. Probable illegal overweight vehicles that can Nowak A. S., 1999. Calibration of LRFD Bridge Design
cause the maximum load effect are included in the Code, NCHRP Report 368, TRB.
records and can provide the unbiased load spectra. Sivakumar, B., Ghosn, M., Moses, F., and Lichtenstein
Although a WIM data measurement has improved, Consulting Engineers, I., 2008. Protocols for Collect-
there is a need to filter the records. There is no widely ing and Using Traffic Data in Bridge Design. National
acceptable guideline for data filtering and this proce- Cooperative Highway Research Program, TRB.
dure has a major impact on the live load distribution. Scholz, F.W. and Tjoelker, R.A., 1995. Nonparametric Tail
Extrapolation, Simulation Results. ISSTECH-95-015,
Different projects resulted with different filtering cri-
Boeing Information & Support Services, Seattle, WA.
teria. It is understandable to remove the unrealistic
trucks from the population but no heavy vehicles can
565
Study of multiple presence probabilities for trucks using recent

weigh-in-motion data
B. Sivakumar
HNTB Corp., New York, New York, U.S.A
ABSTRACT occurred if the gap between two trucks, that is the dis-
tance between the last axle of the leading truck and
An important parameter that controls the maximum the first axle of the trailing truck, is less than the span
load imposed on the structure is related to the number length. Five WIM sites (ten directional sites) with free-
of simultaneous vehicles on the bridge, which is deter- flowing traffic in New York State were studied by the
mined through data on truck headways under operating NCHRP 1276 Research Team in order to determine
conditions. Accurate headway information should be the maximum multiple presence probabilities for vari-
collected through WIM systems. Simultaneous data on ous truck traffic volumes. Daily truck traffic volume
headways and weights is necessary to determine pos- was classified as light (less than 1000 trucks per day),
sible correlations between truck positions or the lanes average (more than 1000 trucks but less than 2500
they occupy and their weights or other characteristics trucks per day), heavy (more than 2500 trucks but less
such as truck type, size and numbers of axles. Fortu- than 5000 trucks per day), and very heavy (more than
nately, the data needed for multiple presence estimates 5000 trucks per day). Multiple presence probabilities
is presently available and already contained in the raw were compiled for headway separations up to 300 feet,
data files captured by many WIM data loggers. Field in 20-foot increments. Maximum multiple presence
measurements of truck arrival data to a 0.01 second probabilities were obtained for each headway separa-
resolution is necessary to establish multiple presence tion interval. These statistics can be used to simulate
probabilities for a span. multiple-presence events for sites where accurate time
Multiple presence is influenced by traffic volume stamps are not available. The paper will describe the
and span length. When considering multiple trucks on findings of this study. Mr Sivakumar was the Principal
a given span, a multiple presence event is said to have Investigator for NCHRP 1276.
566
MENSUSMONITOR Algorithm implementation for detecting live load

events and assessment of structural effects on bridges
H. Sousa, A. Henriques & J. Figueiras

LABEST, Faculty of Engineering, University of Porto, Porto, Portugal
B. Costa
LABEST, Faculty of Engineering, University of Beira Interior, Covilh, Portugal
ABSTRACT
Nowadays, Structural Health Monitoring (SHM) is a

subject of major interest in the Civil Infrastructures
domain. The monitoring systems used in this domain
are a powerful tool to verify the structures performance
conformity with the advantage to give a valuable
database to study in more detail the real behavior of the
monitored structures (Sousa et al., 2008a). The most Figure 1. Functioning structure of MENSUSMONITOR.
recent monitoring systems allow, in an automatic as
well programmable mode, the observation and register
of a set of parameters. This is possible by predefining Furthermore, although the monitoring systems
reading procedures that are interpreted by the acquisi- deployed on bridges aim to help both the structural
tion systems resulting in a database with the respective integrity assessment and the management of mainte-
sensors readings. The size of those databases can nance/repair operations of large groups of structures,
quickly reach considerable dimensions. Considering they can also be utilized to characterize the cross-
the flexibility of the current monitoring systems used ing traffic, if properly designed for it. As the bridge
in civil infrastructures, the readings sample rate can go is equipped with a kit of sensors, and assuming its
from 0.00001 Hz to 200 Hz, in depending of the sensor behavior remains linear elastic under normal opera-
type and/or of the respective acquisition system. tion conditions, the whole set can act as a balance and
The long-term observation of concrete structures therefore measure traffic characteristics (B-WIM).
is mainly made by following the structure response to This paper presents the implementation of an algo-
environmental actions and the delay effects of the con- rithm for detecting live load events and assessment
crete time dependent behavior, such as shrinkage and of structural effects on monitored bridges. The events
creep, with influence in the stress redistribution. For detection algorithm offer a set of results in form of
this case the sensors sampling rate is low, typical in histograms events that characterize several variables
the order of 1 sample per hour, and by consequence distributions like the number of events detected, the
the volume of data is more easily manageable. How- amplitudes of observed parameters, or even the level
ever the long-term observation of metallic structures of load applied to the structure. The proposed algo-
is different, since the dynamic response is crucial to rithm is implemented in the MENSUSMONITOR
evaluate the fatigue response of these flexible struc- software specifically devoted to the structural health
tures. For these structures, higher sampling rates are monitoring. Grounded in the know-how gathered by
required, typically in the order of 200 Hz, and the the research unit FEUP-LABEST in the recent years,
data management in these cases is more demand- MENSUSMONITOR proposes to give faster and effi-
ing. For both cases, the monitoring systems provide a cient answers to the processing and interpretation of
database with great potentialities to evaluate the struc- records collected by the structural monitoring systems
tural safety. But to have the desired answers in useful deployed on bridges.
time, those databases must be consulted and analyzed
with specific tools to extract the desired knowledge.
567
SS27: High performance concrete lessons of past decades
Organizer: M.L. Ralls
High performance concrete in Washington state
J. Kapur
Washington State Department of Transportation
ABSTRACT particularly important with the increasing demand for

rapid construction (get in, get out, and stay out) and
The Washington State Department of Transportation satisfying environmental requirements to keep bridge
(WSDOT) was very active in the development of high piers out of wetlands and waterways. WSDOT achieves
performance concrete (HPC). WSDOT, as a member high economic value from the inherent cost efficiency
of the AASHTO/SHRP Lead States Team, conducted of precast, prestressed concrete girders constructed
a demonstration project in 1996 through 1998 using with HPC when compared to other alternatives for
HPC to design and construct a three span bridge long spans. WSDOT has developed 95 inch deep pre-
carrying State Route 18 over State Route 516. stressed concrete girders capable of spanning 210
The project presented an opportunity to compare feet.
the standard bridge designs with those using HPC. WSDOT has made several changes to its design
The design comparison proved the economic and long- methodology and specifications to optimize the use
term benefits of using HPC allowing the number of of HPC. In addition, HPC is also used in all bridge
girder lines to be reduced from seven to five, and real- decks, cast-in-places piles, and deck overlays of latex
izing a net cost savings of at least $50,000. As a result, modified concrete, micorsilica modified concrete, and
WSDOT began using HPC in all its precast, prestressed fly ash modified concrete.
concrete bridge girders and has used it for an average The use of HPC improves construction economy
of 20 bridges per year since 1998. When cost savings by permitting longer spans, reducing girder lines, and
are extrapolated to all HPC bridges, significant savings allowing shallow girders that improve vertical clear-
can result. Seven out of ten bridges designed in the ance. Additional benefits include faster construction,
last decade by Washington State have utilized precast, reduced inspection and maintenance times, longer
prestressed concrete superstructure elements. service life and reduced life-cycle cost.
HPC technology has also been used in the develop-
ment of 83 inch and 95 inch deep precast, prestressed Keywords: High Performance Concrete, Super-
concrete super-girders for longer spans. This is girders, rapid construction.
571
HPC lessons learned and future directions
S.N. Lane
Portland Cement Association, Washington, DC, USA
ABSTRACT survey conducted during 20062007, it was found that

16 State DOTs used HPC on up to 10% of their bridges,
The first Strategic Highway Research Program (SHRP) 19 State DOTs used HPC on 10% to 80% of their
brought needed research into High Performance Con- bridges, and 15 State DOTs used HPC on more than
crete (HPC) for Bridges. Once SHRP ended, the 80% of their bridges (Triandafilou 2009).
Federal Highway Administration (FHWA) was then Now, some states are using self-consolidating con-
asked to lead the effort in implementing the research crete for certain bridge applications. This type of con-
results emanating from SHRP. One of the key efforts crete can be very easy to place and improve the quality
of FHWA was working to define HPC in terms of com- of the concrete in areas of dense reinforcing steel.
ponents that could be specified by states, and then Other states are innovating with Ultra High Perfor-
working with individual states to build bridges and mance Concrete in bridge members, taking advantage
bridge components with HPC. of very high compressive strengths.
Initially only five states partnered with FHWA to try This paper summarizes the initial steps and growth
the new HPC technologyTexas, Nebraska, Virginia, in use of HPC in US bridges, the lessons learned, and
New Hampshire, and Washington State. Then more the future direction of HPC for bridges.
states joined quickly as successful construction of
HPC bridges emerged, with some lessons learned, in
the first five states. Soon HPC bridge projects were REFERENCE
also underway in Colorado, Georgia, Alabama, North
Carolina, Ohio, Tennessee, and South Dakota. Addi- Triandafilou, L., Updated FHWA HPC Survey, HPC Bridge
tional states followed suit. More recently, in an FHWA Views, Issue 53, JanuaryFebruary 2009.
572
HPC implementation in Virginia with lessons learned
C. Ozyildirim
Principal Research Scientist, Virginia Transportation Research Council, Charlottesville, Virginia, U.S.A.
ABSTRACT air content is needed (Ozyildirim 2004). High strength

concrete with normal or lightweight aggregates and
This paper discusses the implementation of the high with large 0.6-in strands can be used successfully to
performance concrete (HPC) program in Virginia as it reduce the number of beams and allow longer spans.
addresses workability, durability, strength, and crack For longer spans, LWHPC should be considered to
control. The lessons learned during this program are reduce the superstructure and substructure dead loads
also presented. HPC that has high workability and and foundation sizes. Cracks adversely affect durabil-
is self-consolidating was introduced. HPC with low ity, and can be minimized through proper selection of
permeability for longevity throughout the structure ingredients; mixture proportioning by minimizing the
and high strength in beams was used (Ozyildirim water, cement, and paste content; moisture and ther-
2005). High strength in beams resulted in fewer beams mal control; adequate consolidation and curing; and
and smaller beam cross sections. Lightweight HPC addition of fibers to help to control length and width of
(LWHPC) with improved durability and strength was cracks. More representative test procedures such as the
introduced to reduce superstructure and substructure use of temperature-matched curing are needed. Speci-
loads for longer span structures, thereby reducing fying performance parameters as in ERS is expected
foundation requirements, and to extend the service life. to lead to improved HPC.
For longevity of reinforced structures, crack control HPC implementation is expected to provide longer
was sought through the selection of ingredients, mix- service life, cost savings, reduced construction time,
ture proportioning, moisture and thermal control, and and lower maintenance. HPC has a high potential for
proper placement and curing (Ozyildirim 2007). Mix- improved service; however, care should be exercised
tures were sought with less water, cement, and paste and adequate QC/QA measures used in its production
for reduced shrinkage and contraction. Wet curing of to ensure that these benefits can be achieved.
bridge decks was initiated. Control of thermal cracks
in mass concrete was achieved by large amounts of
pozzolanic materials and a reduction in cement con-
tent. Fiber-reinforced concretes were also investigated REFERENCES
to control the severity and width of cracks. Speci-
Lane, D.S. & Ozyildirim, C. 2000. Combinations of
fications were revised and improved to obtain the
Pozzolans and Ground, Granulated, Blast-Furnace Slag
desired product. The Virginia Department of Trans- for Durable Hydraulic Cement Concrete. VTRC 00-
portation is currently evaluating pilot projects using R1. Virginia Transportation Research Council, Char-
end-result specifications (ERS) to achieve the desired lottesville, VA.
high-quality HPC. Ozyildirim, C. 2004. Air-Void Characteristics of Concretes
Virginias work with HPC has shown that more in Different Applications, TRR 1893, Transportation
workable concretes can be obtained that facilitate Research Board of the National Academies, Washington,
placement and consolidation and provide a smooth DC: 7074.
surface finish. The lower permeability essential for Ozyildirim, C. 2005. History of HPC in Virginia, ACI SP-
228. Presented at the Seventh International Symposium on
durability can be achieved with pozzolans or slag
the Utilization of High Strength/High Performance Con-
cement alone or in combination at a moderate (not low) crete, American Concrete Institute, Farmington Hills, MI:
water-cementitious material ratio (Lane & Ozyildirim 821831.
2000). Pozzolans or slag cement also provide resis- Ozyildirim, C. 2007. Durable Concrete for Bridges. Aspire,
tance to chemical attack and control heat of hydration. the Concrete Bridge Magazine, PCI, Chicago, IL, Summer
Attention to air-void parameters rather than the total 2007: 1519.
573
Status of Texas DOT HPC implementation
K.R. Pruski
Texas Department of Transportation, Bridge Division, Texas, USA
ABSTRACT concrete distress mechanisms, appear to be adequately

mitigated by current and former practices. These prac-
The Texas Department of Transportation (TxDOT) tices are presented along with the subsequent issues
currently has over 33,000 bridges on the state sys- that have arisen when combined with specifying high
tem that it has responsibility to operate and maintain. performance concrete (HPC) is discussed. The most
Over 60 percent of the bridges were constructed at widespread distress mechanism TxDOT addresses by
least 40 years ago. Thus, it can be seen that a long specifying HPC is concrete structure deterioration
service life is required by the inventory of TxDOT caused by reinforcing steel corrosion.
bridges. If one percent of the bridges are replaced TxDOT first used HPC in the early 1990s on
per year, the estimated required service life would be two projects where high strength concrete with an
approximately 100 years. The concrete materials used emphasis on low permeability for increased durabil-
40 years ago typically will not provide this expected ity was specified (Ralls, M.L., Carrasquillo, R.L., &
service life and thus significant maintenance may be Burns, N.H. 1998). The evolution of methodology for
required to keep the system operational.This paper dis- specifying HPC is presented along with discussion
cusses the implementation of obtaining improved con- on the suitability of performance testing for con-
crete specifications to obtain high performing concrete crete permeability (Cox, W.R. & Pruski, K.R. 2003).
structures. It is conveyed that the use of SCMs are benefi-
The significant concrete distress mechanisms are cial to produce HPC including results from research
presented and what is being done to specify concrete to investigating concrete containing fly ash is presented.
counter them. The distress mechanisms includeAlkali- In addition to addressing improved concrete materi-
Silica Reactivity (ASR), sulfate attack, freeze-thaw als, measures including prefabrication and improved
cycling, and reinforcing steel corrosion. Substitut- construction practices beneficial to constructing high
ing supplementary cementitious materials (SCMs), performing concrete structures (Pruski, K.R. et al,
including fly ash, ground blast furnace slag, and silica 2002) are discussed.
fume, for a portion of cement in the concrete mix is
discussed in relation to each distress mechanism.
The occurrence of premature concrete deteriora- REFERENCES
tion attributed to ASR has been observed for over
15 years in TxDOT structures. Because the majority Cox, W.R. & Pruski, K.R. 2003. High Performance Concrete
of the aggregates concrete producers use throughout Structures: A Work in Progress, Transportation Research
the state can be classified as potentially reactive with Board Annual Meeting, Washington, D.C.
respect to ASR, instead of prohibiting their use, con- Pruski, K.R., Medlock, R.D., & Ralls, M.L. 2002. Prefabri-
crete mixtures have been designed to mitigate the cation Minimizes Traffic Disruptions, HPC Bridge Views,
Issue No. 21, May/June, www.cement.org/pdf_files/hpc-
harmful effects associated with using these aggre- 21mayjun02.pdf.
gates. The primary mitigation option that appears to Ralls, M.L., Carrasquillo, R.L., &Burns, N.H. 1998. Opti-
be the most cost effective and fail safe is limiting mization of Structural Design for High Performance
the maximum quantity of cement and including a Concrete Bridges, Transportation Research Record,
substantial amount of SCMs in the mix. Both sul- No. 1624.
fate attack and freeze-thaw cycling, though significant
574
High performance concretes evolution in NH
D.L. Scott
New Hampshire Department of Transportation, Bureau of Bridge Design, Concord, NH, USA
Institute, Gouda, The Netherlands
ABSTRACT Quality Control/Quality Assurance (QC/QA) specifi-

cation rather than a prescriptive concrete specification
The New Hampshire Department of Transportation for the deck.
(NHDOT) now uses a performance based, i.e., quality Today, NHs QC/QA specification includes pay fac-
control/quality assurance (QC/QA), specification to tors that consider strength, air content, water/cement
consistently construct durable concrete bridge decks. ratio, permeability, and concrete cover over the rein-
This specification evolved from NHDOTs participa- forcement. Contractors can earn up to 7% of the
tion as a lead state in the Federal Highway Adminis- concrete bid price for exceptional in place concrete.
trations (FHWA) High Performance Concrete (HPC) QC/QA concrete decks are now standard practice
Bridge Showcase Program, undertaken from 1996 in NH.
through 2000. NHs first HPC bridge, in Bristol, NH, Although at the time of this writing, NHs three HPC
was constructed in 1996. Development of the 8000 psi bridges are, at most, 15 years old, all decks and girders
girder strength showed that too much air content can are still performing well.
make strength gain difficult. However, such strengths
make 12 6 girder spacings achievable. NH also
learned that wide girder spacings have the undesired REFERENCES
consequence of increasing deck falsework costs.
NHs second HPC bridge, also in Bristol, NH was Fratzel, Todd 1996. Evaluation of high performance concrete
constructed in 1999. With this project, the deck false- slabs including in-situ testing at a bridge deck testing
work costs were mitigated by the use of stay-in-place facility. Graduate Thesis, University of New Hampshire,
Durham NH.
partial depth concrete panels spanning the girders. Goodspeed, Charles et al. 1996. High-performance concrete
NHs third HPC bridge, located in Rollinsford, NH, defined for highway structures. Concrete International
uses experimental carbon fiber reinforced polymer, 18(2): 6267.
instead of steel reinforcing steel, as the deck reinforc- Whittemore, M. & Stamnas, P. 2004. Lessons learned in New
ing. This is the first of NHs HPC projects to use a Hampshire. HPC Bridge Views Issue No. 36: 2.
575
SS28: Construction, architecture & testing of soil-steel bridges
Organizers: Z. Manko & D. Beben
Dynamic analysis of soil-steel arch road bridges
D. Beben
Faculty of Civil Engineering, Opole University of Technology, Opole, Poland
Z. Zee Manko
Wroclaw University of Environmental and Life Sciences, Wroclaw, Poland
ABSTRACT
The paper is presented the theoretical consideration of

soil-steel objects (Gimn bridge and Wroclaw bridge
Fig. 1) during dynamic field load tests. The critical
speed magnitudes, dynamic coefficients (DAF), veloc-
ity vibration, vibration frequency were determined in
the paper.
Loads moving across the bridge at high speeds
cause vibration of the bridge structure, impact as a
result of rough road surface and unequal deflections
of the vehicle springs, etc. Dynamic reaction stress Figure 1. Wroclaw Bridge during passing the Kamaz
exerted on elements of the bridge structure as well vehicle on the object with speed of 30 km/h.
as the strains caused by mobile rolling stock become
bigger than similar static loads; i.e. slow placement of
immobile rolling stock of the same or similar weights.
The exact calculation of the bridge dynamics consider- For the dynamic tests, two inertial inductive sensors
ing all the above mentioned factors, the free vibration PEVA 7225 type were fixed at the edge of roadway
of the structure and also possibilities of resonance of (or sidewalk) and reinforced concrete collar and strain
the bridge require exceptionally complicated calcula- gages for strains measurements in the transversal and
tions which do not yet yield an absolute satisfactory longitudinal directions.
solution. Based on strains at characteristic points and cross
Generally, the dynamic coefficient value is related sections of the span, as well as deflections obtained
to the so-called critical speed of the truck and the value at three different points of shell structure the dynamic
of the largest vibration amplitude that occurs. This coefficient values (DAF) were determined for all the
velocity can be calculated using many tests (movement dynamic load variances IXXIV (Giman Bridge) and
of the same load at different velocities across the same IVI (Wroclaw Bridge). On their basis, among other
bridge). The critical speed is defined as a speed during things, the critical speed has been determined on the
which the value of dynamic coefficient is maximum. level vcr. = 60 km/h. It was also noticed that the mag-
During the dynamic tests, a truck was used. The nitude of the DAF were lower in comparison to values
weighed front and rear axle loads together consider- calculated in accordance with the Polish Bridge Load
ably exceeded the total weight of the truck (with its Standard of PN-85/S-10030 (1985) (12% & 23%), the
load included) of circa 300 kN. The speeds of the mov- Eurocode (2002) (15% & 16%) and the BV Bro (2004)
ing truck over the bridges (used earlier in the static tests regulation (19% & 25%) for the Giman Bridge and
were estimated as follows 5, 10, 20, 30, 40, 50, 60 and the Wroclaw Bridge respectively. It should be clearly
70 km/h in both directions. Measurements of dynamic mentioned that the calculated dynamic coefficients
interactions were taken also when the truck was mov- (DAF) were adopted just as the traditional road steel
ing by threshold of 0.03 0.20 m fixed at half way of bridges (which in the experimental tests the values
the bridge distance perpendicular to the longitudinal were obtained mainly much lower), whereas standard
axis of the roadway and during its braking with vari- regulations in this range (in Poland as well as Sweden)
ous different velocities at different distant points of the do not yet relate to the such solutions in a design of
bridge. the new structures.
579
Dynamic testing of a soil-steel composite railway bridge
R. Karoumi, E. Bayoglu Flener & H. Sundquist

The Royal Institute of Technology (KTH), Division of Structural Design and Bridges, Stockholm, Sweden
ABSTRACT greater than the values specified in bridge design

codes. Despite this, due to the high damping involved,
The dynamic response of a long-span arch soil-steel bridges of this kind are believed to be less sensitive to
composite railway bridge is studied. The bridge has resonance problems from passing trains.
a span of 11 m and a rise of 4.3 m (see Figure 1
and Figure 2). Strains, displacements and vertical bal-
last accelerations were measured during passages of REFERENCE
a locomotive at different speeds. The results indicate
that the speed of the locomotive has a large influ- Bayoglu Flener, E., Karoumi, R. 2009. Dynamic testing
of a soilsteel composite railway bridge. Engineering
ence on the displacements, thrusts and moments (see
Structures, Volume 31, Issue 12, Pages 28032811.
Figure 3). The structure was found to be safe when
measured values of moments and thrusts were com-
pared with the live load calculations according to
design codes. However, dynamic amplification fac-
tors as high as 1.45 were obtained for the moments
at the quarter point and this is found to be much
Figure 2. The construction of the steel culvert bridge. This

new bridge was placed on top of an existing masonry arch
bridge with a gap in between, which leaves the old bridge
Figure 1. Sectional dimensions of the culvert. structurally functionless.
Figure 3. Measured moments at the crown centreline for different locomotive speeds.
580
Rehabilitation of old arch bridges using corrugated shell structures
J. Kurzynski
Commercial-Service Bridge Company ZBYSZKO, Klodzko, Poland
Z. Zee Manko
Wroclaw University of Environmental and Life Sciences, Wroclaw, Poland
D. Beben
Faculty of Civil Engineering, Opole University of Technology, Opole, Poland
ABSTRACT
There are many engineering structures located on all

kinds of roads and railways in Europe which needs to
be urgently upgraded or rebuilt or even replaced. A
common result of degradation, caused by the destruc-
tive action of natural forces (e.g. floods), improper
use or belated repairs, of bridges is a considerable
decrease in their load-carrying capacity. This creates
serious dangers to road and railway users. Therefore
the bridges should be quickly and effectively repaired.
Effective solutions entailing minimum costs, includ-
ing social costs resulting from the traffic delays and
road or railway closures, should be sought.
Small and medium-sized bridges and culverts can
be modernized in many different ways. For example,
the damaged bridges can be rebuilt or reinforced or
even replaced on a new structure, however the cost
of the last solution is very expensive. Often the flexi-
ble structures made from the corrugated steel plates
(CSP), with different corrugations dimensions (e.g.
for spirally pipes: 68 12, 100 20, 125 26 mm; Figure 1. Side view on bridge after repair by using the CSP:
and for structures made from plate sheets: 70 13, (a) over railway in Augsburg (Germany), (b) under railway in
Piensk (Poland).
100 22, 150 50, 152 51, 200 55, 380 140,
400 150 mm) are used as a good alternative for the
traditional bridge solution.
Different shell shapes in the roads longitudinal Several successful projects of this kind in Poland have
direction arch, box, circular (pipe and elliptical) made such solutions quite popular among the Pol-
suited to the kind and size of the obstacles and the ish designers who use them to repair and reinforce
soil and hydrologic conditions are also used for this small bridges and culverts. However, the actual design
purpose. The advantages of this solution include: the methods of such flexible structures in Europe are too
short time (a few days only) needed to assemble the simplified because the rational calculation methods are
steel shell structure, the simple design, the relatively complicated and the advanced numerical model should
light deadweight (the structure can be founded on be applied.
weak ground), the long life (often exceeding 80100 The construction works can be carried out very
years) and aesthetic values. Upgrading of engineer- quickly and efficiently (about 510 weeks). The pre-
ing structures by means of corrugated-plate structures sented structural solutions are cheaper (on average
has been known and quite commonly employed on 3040%) than different methods of bridge repairs.
almost all continents (mainly in Canada and USA and The examples of modernization of old road and rail-
in the Scandinavian countries) for many years. The way viaducts (and culverts) in Poland and Germany by
shell structures made from corrugated steel plates were use of shell structure made from corrugated steel plates
first used in USA at the beginning of the 20th Century. are showed in Figure 1.
581
SS29: Lessons learned from instrumented bridges
Organizer: M.Q. Feng
Complementing long term health monitoring of bridges

using numerical models
P.J. Fanning
School of Architecture Landscape & Civil Engineering, University College Dublin, Ireland
M.Q. Feng & H. Gomez

Department of Civil & Environmental Engineering, University of California, Irvine, USA
ABSTRACT
The increasing trend for bridge owners to adopt

long term health monitoring strategies for bridges
under their management, using measured vibration
responses, brings with it a need for associated numer-
ical modeling to complement measured data sets to
explain response trends observed.
The West Street On-Ramp Bridge, constructed in
2000, is a three span post-tensioned reinforced con- Figure 3. Modes shapes 1 & 2.
crete, horizontally curved, over-bridge on the Santa
Table 1. Frequencies & model parameters.
Ana Freeway (I-5) in the city of Anaheim, California,
Figure 1. Measured 3D FE
The acceleration response at eleven sensor loca- Data Set
tions, monitored since 2002, indicate a 5% decrease Year 2002 2009 2002 2009
in the lower frequencies of vibration but the density
of sensors is not sufficient to identify the cause using Edeck (MPa) 20100 20300
measured data alone. A three dimensional finite ele- Ebents (MPa) 28300 22200
ment model, (Fig. 2), was used to complement the F1 (Hz) 2.05 1.90 2.04 1.89
monitoring programme. F2 (Hz) 2.44 2.34 2.71 2.56
F3 (Hz) 2.83 2.64 2.83 2.64
This 3D shell/solid model enabled assignment

of boundary conditions consistent with construction
details. Model input parameters, material properties
and support stiffnesses, were varied to construct a
model response space for the first three natural fre-
quencies. Searches in the response space identified
model parameters for optimum correlation with the
measured frequencies. The first and second mode
Figure 1. West street on-ramp. shapes are illustrated in Figure 3, while the model
parameters found for correlation with 2002 and 2009
measurement data sets are listed in Table 1.
The Youngs Moduli for the deck and bents for
correlation with year 2000 frequencies were approxi-
mately 80% of values proposed by ACI based on their
respective concrete strengths. The reduction in fre-
quency over the period to 2009 is presently attributed
to a reduction in the effective stiffness of the bents
this is the subject of ongoing study as is the effect
of post-tensioning stresses on the various modes of
Figure 2. Shell/solid FE model. vibration.
585
Long-term bridge performance monitoring program in California
M.Q. Feng
Department of Civil and Environmental Engineering, University of California, Irvine, California, USA
L.-H. Sheng
California Department of Transportation, California, USA
ABSTRACT
The Division of Engineering Services at the California

Department of Transportation (Caltrans) started to
instrument bridges in 1999 for their long term per-
formance monitoring and evaluation. As a structure
deteriorates due to aging or suffers from damage
caused by extreme loads such as earthquakes, stiffness
of the damaged structural elements would decrease,
and as a result, the global vibration characteristics of
the structure would change. Therefore, by monitoring
the structural vibration, one can identify the change
in structural vibration characteristics and then further
evaluate the change in the element stiffness.
Three new concrete bridges (Fig. 1) in the Orange
County were instrumented with accelerometers, strain
gauges, displacement sensors and pressure gauges Figure 1. Instrumented bridges in California.
at strategic locations. Traffic-induced vibration data
are periodically collected locally or remotely through
wireless Internet. They are processed with a variety of
signal processing and system identification software the instrumented bridges were analyzed and developed
tools to extract structural vibration characteristics and into a structural stiffness database using a software
to further update their finite element analysis models. platform developed in this study.
All the data were then stored in the developed database This project has demonstrated the potential of using
for statistic analysis. the sensor technology for long term and real-time
A number of system identification methods were structural health monitoring and post-event damage
reviewed and developed in this study for identifying detection.
the structural element stiffness based on measure- The baseline updating methods based on vibration
ment of bridge vibrations caused by traffic and seismic measurement and stiffness identification developed in
excitations. A unique traffic excitation model was pro- this study can also be applied to existing bridges (that
posed for more reliable stiffness identification based represent the majority of the Caltrans inventory) in the
on traffic-induced vibrations. The effectiveness of following two ways; one is for establishing the current
these methods in evaluating seismic damage on a baseline of the bridge for its future damage detec-
bridge structure was demonstrated through seismic tion and deterioration assessment, and the other is for
shaking table tests of a multi-bent multi-column con- assessing the ongoing health if a database of similar
crete bridge model. Long-term monitoring data from types of bridges exist for the comparison purposes.
586
Performance assessment of Jiangyin Bridge using long-term structural

health monitoring data
H.F. Zhou
Structural Health Monitoring Division, PolyU Technology & Consultancy Company Limited, Hong Kong
Y.Q. Ni, Y.L. Guo & J.M. Ko

ABSTRACT formulated for design verification and extreme value

prediction. Another temperature effect considered is
The Jiangyin Bridge is a suspension bridge with the temperature-induced strain variation. A linear rela-
1385 m main span, which crosses the Yangtze River tionship between strain and temperature is observed
in Jiangsu Province, China. It is the first bridge with a and the strain-temperature correlation model has been
main span exceeding 1 km in Chinese mainland, and established for quantitative estimation of the thermal
is also among few bridges in Chinese mainland with strain. Then, the wind characteristics specific to the
structural health monitoring (SHM) systems being Jiangyin Bridge site and the wind-induced lateral dis-
installed in the end of last century. After operation for placement of the deck are investigated. The wind data
several years, the SHM system was found with mal- are analyzed to obtain: (i) mean wind speed and direc-
function in sensors and data acquisition units (DAUs), tion, (ii) wind speed profiles, (iii) turbulence intensity,
and with insufficient sensors to provide necessary (iv) gust factor, and (v) wind spectrum. Verifications
information for structural health evaluation. The SHM of design assumptions and rules of thumb are made,
system has been upgraded by a consortium of The and the wind-induced lateral displacement of the deck
Hong Kong Polytechnic University and Jiangsu Trans- is explored. A linear relationship between transverse
portation Research Institute by repairing/replacing the wind speed and deck lateral displacement is observed
disordered sensors and installing new sensors, renew- and the correlation model between them has been
ing the data acquisition and transmission system, and established
developing a new structural health evaluation soft- for quantitative estimation of the wind-
ware system. The upgraded SHM system comprises induced displacement. Finally, the modal properties
over 170 sensors, including anemometers, GPS, dis- of the Jiangyin Bridge are studied. Modal properties
placement transducers, fiber optic temperature and under normal conditions, including modal frequency
strain sensors, and accelerometers. The system com- and mode shape, have been identified from the acceler-
menced operation in June 2005; since then structural ation data. Analysis of modal properties under extreme
health monitoring data have been continuously col- events, such as typhoon, ship collision, and subsequent
lected, including those obtained during the typhoon damage identification is ongoing.
Matsa and one ship collision accident.
Making use of the long-term monitoring data, per-
formance assessment is carried out for the bridge by REFERENCES
means of multi-dimensional data analysis. This paper
reports the performance assessment with emphasis Choi, C.C. 1983. Wind loading in Hong Kong commentary
on the code of practice on wind effects Hong Kong-1983.
on: (i) analysis of temperature profile and effects;
Hong Kong: Hong Kong Institution of Engineers.
(ii) recognition of wind characteristics and effects; Cook, N.J. 1985. The designers guide to wind loading of
and (iii) characterization of structural dynamic prop- building structures. UK: Butterworth-Heinemann.
erties. First, the temperature profile and its effect on Dyrbye, C. & Hansen, S.O. 1997. Wind loads on structures.
the movement of expansion joint and the variation of New York: John Wiely & Sons.
strain are analyzed. The deck effective temperature Ministry of Transport of the Peoples Republic of China.
is evaluated from the temperatures in a deck cross- 2004. Wind-resistant Design Specification for Highway
section to formulate the temperature profile. Upon this, Bridges. Bejing: China Communication Press.
the thermal movement of expansion joint is studied Simiu, E. & Scanlan, R.H. 1996. Wind effects on structures,
third edition. New York: John Wiley & Sons.
and the displacement-temperature correlation model is
587
SS30: Chinese bridges
Organizer: M.-C. Tang
Major steel bridges for high speed railway in China
Z. Gao
China Railway Major Bridge Reconnaissance & Design Institute Co., Ltd. Wuhan, China
ABSTRACT Table 1. Objective Desig Speeds.
Major steel bridges for high speed railway in China Zhengzhou Jinan
are recently developed. Some of these bridges, located Tianxing Dashen Yellow Yellow
Beijng-Shanghai and Beijing-Guangzhou High Speed Project zhou guan River River
Railways, across Yangtze River and Yellow River,
Objective Design 200 300 350 350
are under construction, include: Tianxingzhou Bridge, Speeds (km/h)
Dashenguan Bridge (Fig. 1), Zhengzhou Yellow River
Bridge, and Jinan Yellow River Bridge. All these
major bridges are multy railway lines or rail-cum-road
bridges, with heavy loading capacity and high railway Yangtze River Bridge, Nanjing Dashenguan Yangtze
operation speeds, therefore, the steel truss structure River Bridge, Zhengzhou Yellow River Rail-Cum-
are adopted for all of them. Some new materials, Road Bridge, and Jinan Yellow River Bridge of
new structures and new workmanships are applied Beijing-Shanghai High Speed Railway. For all the
accordingly. above mentioned bridges, steel truss has been used
Bridges are important structure to high speed rail- with its good rigidity. Some new materials, new
way and railway passenger special line, have main strutures and new workmanships have been applied.
impact on stability of high speed train operation, Objective design speeds for both Beijing-Shanghai
as well as comfort and safety of passengers. Cur- High Speed Railway and Beijing-Guangzhou Railway
rently, Beijing-Shanghai and Beijing-Guagzhou High Passenger Special Line are 350km/h. According to the
Speed Railways are under construction. They have traffic conditions at the bridge location, and the struc-
total 4 major bridges across Yangtze River or Huanghe ture features, the objective design speeds of 4 bridges
River, namely Wuhan Tianxingzhou Rail-Cum-Road above mentioned refer to Table 1:
These bridges have heavy self-weight load, and most
of them are taking several railway lines or carriage-
ways, so the live loads are considerable great also.
Tianxingzhou Bridge undertake 4 railway linesand 6
carriageways, Dashenguan Bridge 4 railway lines plus
2 metro lines, both of them are the bridges with the
largest design loads in the world. Please refer to Table 2
for these bridges span length and design live loads.
According to thelarge span, high Speed and heavy
load features, some new techniques are adopted in the
bridge design.
Figure 1. Dashenguan Bridge.
Table 2. Bridge Spans and Design Live Loads.
Zhengzhou Jinan
Tianxingzhou Dashenguan Yellow Yellow
Bridge Bridge Bridge River Bridge River Bridge
Span(m) 504 2 336 5 168 3 168

Railway lines 2 passenger special 2 high speed railway 2 passenger 2 high speed
and 2 Class I railway and 2 Class I railway special railway and
2 passenger special
Highway lanes 6 / 6 /
Metro lines / 2 / /
Live loads (t/m) 35.1 34.7 19.1 25.6
591
Major bridges in Shanghai
Y.-P. Lin
Shanghai Municipal Engineering Design General Institute, Shanghai, China
ABSTRACT
Shanghai was the first city featuring major long span

bridges in China. The design and construction tech-
niques for long span bridge have undergone significant
development and improvement over the past several
decades.
Bridge construction in the city has kept apace with
a series of impressive achievements. Among these,
some major bridges have attracted particular atten-
tion. The construction of the Nanpu Bridge in 1988 Figure 2. Lupu Arch Bridge.
is regarded as the beginning of modern long span
cable-stayed bridge construction in China, doubling Table 1. Major bridges in Shanghai.
the largest span length in the country at that time.
Span
The 602 m spanYangpu Bridge was under construction Bridge Name Built Year (m) Bridge Type
right after and became the worlds longest span cable-
stayed bridge when it was open to traffic in 1994. Since Nanpu 1991 423 Cable-Stayed
then, design and construction technology for cable- Yangpu 1993 602 Cable-Stayed
stayed bridges have been continuously developed and Xupu 1997 590 Cable-Stayed
improved in China. Shanghais Yangtze River Bridge Shanghai 2009 730 Cable-Stayed
(Fig. 1) and Minpu Bridge are two long span cable- Yangtze River
stayed bridges that incorporate particularly innovative Minpu 2010 708 Cable-Stayed
techniques. Besides Lupu Bridge (Fig. 2) was built and Lupu 2003 550 Steel Box Arch
came into operation in 2003 as the worlds longest arch
span at that time.
Shanghais five cable-stayed bridges and Lupu arch
bridge are described in this paper (Tab. 1), with spe- construction, structural characteristics, technological
cial attention devoted to experience gained during developments, and innovative techniques.
Through the construction of these bridges, the
design and construction techniques of major bridges
have been developed. Compared with other types of
structure, the modern cable-stayed bridge has a rel-
atively short history, while building long span arch
bridge is more difficult. Therefore, many problems
related to design methods and construction technol-
ogy require further study and improvement. Shanghai
will continue to contribute in the development of major
bridges.
Figure 1. Shanghai Yangtze River Bridge.
592
Bridges in Chongqing the bridge capital of China
D. Yin
T. Y. Lin International China, Chongqing, China
W. Xu & X. Duan
China Zhongtie Major Bridge Reconnaissance & Design Institute Co., Ltd. Wuhan, China
ABSTRACT
Being bisected by many large rivers, most notably

the Yangtze River and the Jialing River, the city of
Chongqing needs a lot of major bridges to accom-
modate its ever growing traffic volume. The city has
been very progressive in accepting new ideas and
innovative bridge concepts. The city now has over 50
major bridges, including the worlds longest box girder
bridge, and the worlds largest arch span.
Major bridges built or under constructionm in the
last twelve years were summarized in Table 1. All Figure 2. Renderings of the Tied-Arch Caiyuanba Bridge.
fourteen bridges in the table have span length larger
than 250 m. Its an exhibit of modern bridge types,
innovative construction technology and engineering
creation. It wasnt surprising that the mountain city
is now recognized as the Bridge Capital of China.
Table 1. Major bridges in Chongqing since 1997.
Bridge Name Built Year Span (m) Bridge Type

Figure 3. Renderings of Chaotianmen Bridge.
Lijiatuo 1997 444 Cable-Stayed
Huanghuayuan 1999 250 Concrete Girder
Egonngyan 2000 600 Suspension
Wushan 2005 460 Steel Truss Arch
Shibanpo 2006 330 Hybrid Girder
Caiyuanba 2007 420 Steel Box Arch
Jiahua 2007 252 Concrete Girder
Wujiang 2008 340 Cable-Stayed
Yudong 2008 260 Concrete Girder
Chaotianmen 2009 552 Steel Truss Arch
Jiayue Under Constr. 250 Cable-Stayed
Shuangbei Under Constr. 330 Cable-Stayed
Qiansimen Under Constr. 315 Cable-Stayed
Dongshuimen Under Constr. 445 Cable-Stayed
Figure 4. Renderings of Dongshuimen Bridge.
In this paper, five major bridges recently built or

under construction in the metropolitan area of city
are introduced, including Shibanpo Brdige (Fig. 1),
Caiyuanba Bridge (Fig. 2), Chaotianmen Bridge
(Fig. 3), Qiansimen Bridge and Dongshuimen Bridge
(Fig. 4).
Figure 1. Shibanpo Bridge Hybrid with concrete and
steel box.
593
Cross-river bridges in Jiangsu
Q. You
Jiangsu Provincial Communications Dept, Nanjing, China
L. Ji
Jiangsu Taizhou Bridge CCD, Taizhou, China
X. Dong
Jiangsu Sutong Bridge CCD, Nanjing, China
ABSTRACT Table 1. Major Cross-river bridges in Jiangsu Province.
Jiangsu Province is located in Yangtze River Delta, the Bridge Name Built Year Span (m) Bridge Type
mostly developed area in China. It is also theT-junction
intersection of coastal areas and the Yangtze River Jiangyin 1999 1385 Suspension
Runyang 2005 1490 Suspension
adjacent to Shanghai, Zhejiang, Shandong and other
Taizhou(Fig.2) Under Constr 1080 2 Suspension
provinces with busy land and water traffic. However, Nanjing 4th Under Constr 1418 Suspension
the separation caused by the Yangtze River obstructed Nanjing 2nd 2001 628 Cable-Stayed
the cross-coast exchange and development. Therefore, Nanjing 3rd 2005 648 Cable-Stayed
bridges constructed over the Yangtze River had been Sutong(Fig.1) 2008 1088 Cable-Stayed
the dream of people living on both banks of the river. Nanjing Yangtze 1968 160 Steel Truss
Since Jiangsu Province is located in the middle and River Girder
lower reaches of Yangtze River, the huge water width Dashengguan 2010 336 Steel Truss Arch
and depth as well as the geological complexity make Chongqi Under Constr 185 Concrete Girder
it extremely difficult to construct cross-river bridges
over there. Until 1968, Nanjing Yangtze River Bridge,
the first cross-river bridge along the water section of
450 kilometers in Jiangsu Province, was opened to traf-
fic. In 1980s, with the development of social economy
and the improvement of technology, Jiangsu started to
conduct researches and planning for the construction
of cross-river channels overYangtze River. Therein, 12
positions were selected for the construction of bridges
or tunnels (Fig. 1). Figure 2. Taizhou Bridge.
This article briefly introduces the construction of
cross-river bridges in Jiangsu Province(Tab. 1).
In 2009, the State approved the general plan of
Jiangsu coastal development. To help northern Jiangsu
preferably blend into the Yangtze River Delta and
join the Metropolitan Area of Shanghai, more cross-
river bridges are needed. Recently, Jiangsu Province
has further adjusted planning of cross-river passages,
Figure 3. Sutong Bridge.
confirming that there are 24 cross-river passages in
Jiangsu Province, 11 of which are integral parts of
cross-river passages of Jiangsu Highway. Compared
to first planned location, the geological, hydrologi-
cal and navigation conditions of newly-planned bridge
location are more complicated and the project con-
struction is more difficult. We will make use of our
own and domestic and overseas advanced experience,
go on with the path of development and innovation,
and make contribution to the development of bridges
Figure 1. Cross-river Channels Plan for Express Ways in in our country.
Jiangsu Province.
594
General Session
Organizers: F. Biondini & S. Pakzad
Experimental study on repair welding under static and cyclic loads
Y. Agano & S. Fujihira

Katayama Stratech Corp., Osaka, Japan
S.H. Lee & Y.C. Kim

Joining & Welding Research Institute, Osaka University, Osaka, Japan
ABSTRACT
Plenty of bridges have been damaged such as fatigue

crack, corrosion and deformation due to degradation
of the bridges, unexpected increase of traffic volume
and variation of traffic environment.
So, works such as replacement, improvement, repair
and reinforcement are necessary. Considering the eco-
nomical and social situation in Japan, the most suitable
work has to be selected among them. Fixed structures
such as bridge are required advanced technology that
is suited for their situation.
As the repair/reinforcement work of steel bridges,
the splice bolted joints are mainly used in a site. On the
other hand, welding is not so general. So, the develop-
ment of welding technique has been required one of
the selective works.
At this moment, welding is performed by removing
the loads through the setting up the bents or traffic con-
trol. But, welding under loads without traffic control
is desirable.
The load conditions of bridges can be classified
three types: no load, static load, fluctuated load.
Fitting the member and avoiding the defects are
required in site welding.
In case of welding under no load and static load, Figure 1. Schematic illustration of specimen.
it is possible to prevent of cold cracks and deforma-
tion, using suitable welding conditions. But, in case of
welding under fluctuated load, it is impossible to pre-
vent hot cracks. Hot crack may bring about the collapse I-type joints only in welding under cyclic loads. From
of the structure. the results of a fractography, the confirmed patterns of
About twenty years ago, a judging method of advis- fracture were projections of cellular dendrite and stri-
ability of welding under fluctuated loads was proposed. ations. And it could be known that the magnitude of
And it applies for some bridges. But, it is not widely influenced hot crack.
used by reason, it can be said that the suitable range of
welding for avoiding the hot crack is extremely narrow.
REFERENCE
In order to investigate the possibility of repair by
CO2 arc welding, welding under static and cyclic loads Kim, Y.C., Lee, S.H. and AGANO, Y. 2008. Investigation
were performed. of Soundness of welds in Welding under Static and
As the results, cracks were not observed at the cru- Cyclic loading, Materials Science Forum Vols.580582:
ciform joints. However, cracks were observed at the 671674.
597
Long-term performance prediction of RC bridge slabs

in a marine environment
M. Akiyama
Tohoku University, Sendai, Miyagi, Japan
D.M. Frangopol
I. Yoshida
Tokyo City University, Tokyo, Japan
ABSTRACT an approximate solution can be found by using sev-

eral approaches. Monte Carlo Simulation (MCS) is
Unlike the case associated with hazards that are usually in general used because of its versatility. MCS-based
considered in structural design (e.g., seismic hazard), methods for non-linear filtering technique have been
there has been a lack of research on marine envi- developed since 1990s.
ronmental hazard assessment. Akiyama et al. (2010) The term Sequential MCS (SMCS) is used in con-
established a methodology for the probabilistic hazard junction with time-dependent reliability assessment.
assessment associated with airborne chloride. Figure 1 This paper follows the procedure of SMCS applied
shows an example of hazard curve associated with to reliability analysis proposed by Yoshida (2009).
two different distances d from the coastline. In this Visual inspections of corrosion crack width and chlo-
paper, the procedure to integrate the hazard associated ride concentration distributions by coring test are used
with airborne chlorides into time-dependent structural as observational information. Using SMCS, multiple
reliability assessment of RC bridge slabs is provided. random variables related to observation information
Failure probability of RC structures in an aggres- can be updated simultaneously.
sive environment depends on many aleatoric and The effects of marine environment, inspection
epistemic uncertainties involved in the evaluation of results and the number of inspections on the updated
hazard assessment of airborne chlorides, corrosion estimates of RC bridge slab reliability are discussed.
process of steel bars, and deterioration of structural
performance (Akiyama et al. 2009). For existing struc- REFERENCES
tures it is possible to reduce epistemic uncertainties
using inspection results (Frangopol 2009). The rela- Akiyama, M, Frangopol, DM, Yoshida, I. 2009. Time-
dependent reliability of existing RC structures in a marine
tionships between observed physical quantities such
environment using hazard associated with airborne chlo-
as inspection results, and the PDFs of related random rides (submitted for possible publication).
variables are used in Bayesian updating. When nonlin- Akiyama, M, Frangopol, DM, Suzuki, M. 2010. Inte-
ear relations or non-Gaussian variables are involved, gration of the effects of airborne chlorides into
reliability-based durability design of R/C structures
in a marine environment. Structure and Infrastructure
Engineering, Taylor & Francis. (in press), 2009, DOI:
10.1080/15732470903363313.
Frangopol, DM. 2009. Life-cycle performance, manage-
ment, and optimization of structural systems under uncer-
tainty: Accomplishments and challenges. Keynote Paper
in Proceedings of the Tenth International Conference on
Structural Safety and Reliability, ICOSSAR2009, Osaka,
Japan, September 1317, 2009: in Safety, Reliability and
Risk of Structures, Infrastructures and Engineering Sys-
tems, H. Furuta, D.M. Frangopol, and M. Shinozuka, eds.,
CRC Press, Taylor & Francis Group, Boca Raton, London,
New York, Leiden, 2010, 1314, and full (23 pages) paper
on CD-ROM, Taylor and Francis Group, London, 2010,
3860.
Figure 1. Probability of exceedance of various amounts of Yoshida, I. 2009. Data assimilation and reliability estima-
air-borne chlorides in Wakkanai City at distance of 0.1 km tion of existing RC structure, COMPDYN 2009, CD281,
and 1.0 km from coastline. Rhodes, Greece.
598
Arrigoni Bridge inspection and testing program
B. Bantimba & A. Siddiqui

AI Engineers, Inc., Middletown, CT, US
ABSTRACT the 14 inch diameter steel pins in the bottom chords

and electromagnetic inspection of the steel hanger
The Arrigoni Bridge is a thirty (30) span structure cables in the main spans both of which had never been
including two prominent tied steel thru truss arches performed.
carrying Route 66 over Route 9, the Providence and The paper outlines the plan that AI and the var-
Worcester Railroad and the Connecticut River between ious non-destructive and equipment subconsultants
Middletown and Portland, Connecticut. In 2007, AI used to perform the routine visual inspection and the
Engineers, Inc. (AI) of Middletown, Connecticut was new UT pin and electromagnetic cable inspection of
tasked by the Connecticut Department of Transporta- this prominent structure. Topics discussed also include
tion to perform a regular bi-annual inspection of the equipment manifests required to visually inspect the
structure. However, in addition to this inspection, AI large trusses and an agreed upon procedure for the
was also tasked with the ultrasonic testing (UT) of maintenance and protection of traffic for this 4 lane,
heavily travelled structure.
Results from both the NBIS as well as the cable and
pin inspections are also presented.
REFERENCES
AASHTO Guide for Commonly Recognized (CoRe) Struc-
tural Elements
AASHTO Manual For Bridge Evaluation, 2008
AASHTO PONTIS Users Manual
Connecticut DOT Bridge Inspection Manual, 2.1
FHWA Bridge Inspectors Reference Manual FHWA-NHI-
03-001
FHWA Manual on Uniform Traffic Control Devices
(MUTCD)
FHWA Recording and Coding Guide for the Structure
Inventory of the Nations Bridges, FHWA PD-96-001
US DOT Publication No. FHWA-HRT-04-042
Figure 1. Arrigoni Bridge, Middletown, CT.
599
On the applicability of random field theory to transportation

network analysis
P. Bocchini & D.M. Frangopol

ABSTRACT
In the last years, a significant scientific effort has

been put in research regarding civil infrastructural
networks and, in particular, transportation networks
(e.g. Akgl & Frangopol 2003, Shinozuka et al. 2006,
Dueas-Osorio et al. 2007). In fact, lifelines have a role
of paramount importance in post-emergency response
and in the recovery activities when a natural (e.g. earth-
quake, hurricane, fire, flood, landslide) or man-made
(e.g. explosion, terrorist attack, vehicular collision)
extreme event strikes.
In sampling based performance analyses, the dam-
age can be modeled as a random field. Random field Figure 1. Percentiles of the performance indicator P as a
theory (e.g. Bocchini & Deodatis 2008) can be profi- function of the correlation length of the bridge structural
ciently used also to perform efficient parametric and damage.
sensitivity analyses that give useful information on
the relative importance of the network elements. In
these cases, the underlying assumption is that the net- further investigations, such as network maintenance
work performance indicators are more sensitive to the optimization, development planning, loss estimation
spatial correlation of the damage than to the actual and network reliability assessment.
intensity of the demand at the locations of the var-
ious lifeline components. In fact, the redundancy of
the network, its ability to redistribute traffic flows, the REFERENCES
possibility of cascading failures and the presence of
non-redundant links create a strong interdependence Akgl, F. & Frangopol, D.M. 2003. Rating and reliabil-
among the various components. Therefore, it is often ity of existing bridges in a network. Journal of Bridge
reasonable to assume that the overall performance Engineering ASCE 8(6): 383393.
Bocchini, P. & Deodatis G. 2008. Critical review and lat-
of a transportation network after an extreme event
est developments of a class of simulation algorithms
does not significantly depend on local effects of the for strongly non-Gaussian random fields. Probabilistic
hazard sources. The previously mentioned underlying Engineering Mechanics 23(4): 393407.
assumption is addressed to estimate its applicability. In Dueas-Osorio, L., Craig, J.I., Goodno, B.J. & Bostrom, A.
particular, a sensitivity analysis is conducted to evalu- 2007. Interdependent Response of Networked Systems.
ate the variation of the network performance indicator Journal of Infrastructure Systems ASCE 13(3): 185194.
as a function of the spatial correlation of the damage. Evans, S.P. 1976. Derivation and analysis of some models for
Figure 1 shows the strong dependence of the perfor- combining trip distribution and assignment. Transporta-
mance of a sample transportation network on the cor- tion research 10(1): 3757.
Shinozuka, M., Zhou, Y., Banerjee, S. & Murachi, Y. 2006.
relation length of the structural damage of its bridges.
Cost-effectiveness of seismic bridge retrofit. In Cruz,
In this paper an integrated analysis framework, Frangopol & Neves (eds.), Bridge Maintenance, Safety,
including the structural analysis, the network per- Management, Life-Cycle Performance and Cost; Proc. Int.
formance analysis (based on Evans 1976) and the conf. IABMAS 2006, Porto, Portugal, 1619 July 2006.
statistical result analysis of transportation networks, is Taylor & Francis.
presented. The same framework can be used for many
600
Design check of bridge over Tejo River in Carregado, Portugal
P. Cachim
DECivil & LABEST, University of Aveiro, Portugal
P. Fidalgo
HDP Lda, Portugal
ABSTRACT
This paper presents the independent design check per-

formed for the bridge over Tejo river in Carregado,
Portugal. The scope of the work involves geological
and geotechnical analysis of foundations for the bridge
and respective viaducts as well as the structural ans
seismic analysis for each construction. This paper only
deals with subjects that refer to the structural design
check, mainly structural conception and design.
Three different structural solutions were used for Figure 1. Aerial view of the bridge.
bridge and respective viaducts. For the northern
viaducts an in-situ concrete slab with beams was
used with column-pile foundations. The total length of
these three viaducts is 1473 m. The southern viaducts
have a total length of 9230 m and are divided into
22 viaducts. The structural solution is prefabricated
U beams with an in-situ slab. The final solution for
each viaduct is monolithic with the column-pile foun-
dations. Length of viaducts ranges from 250 to 530 m.
The bridge has 970 m length with 7 columns. Internal
spans have 130 m while the end spans have 95 m. The
structural solution is a reinforced concrete box that
support through steel struts the concrete flanges.
The design check consisted in an analysis of the
several written and drawn elements of the Project and Figure 2. Bridge during construction.
also a structural analysis of southern viaducts. Also
some particular structural analysis, mainly related with REFERENCES
the construction process, was carried out during the
duration of the works. Mathieu, h. 1984. Les techniques of verification des projects
The design review contributed decisively to the of structures. RPEE, ANO VII, no. 20/21, 7382.
success of the work, in particular regarding: the devel- EN 1991-2: 2003. Eurocode 1: actions on structures-part 2:
opment of faster, simple and economic constructive Traffic loads on bridges. CEN.
EN 1992-1-1: 2004. Eurocode 2: design of concrete
solutions; the improvement of solutions used for criti- structures-part 1-1: General rules and rules for build-
cal points where uncertainty is bigger; the optimization ings. CEN.
of the constructive process; the analysis of the behavior EN 1998-1-1: 2004. Eurocode 8: design of structures
of the structure during the constructive process. The for earthquake resistance-part 1: General rules, seismic
main conclusions of the design check mainly regard- actions and rules for buildings. CEN.
ing its importance for the success of the overall process
are highlighted in this paper, especially regarding the
development of faster construction speed, simplicity
and economy, but also regarding improvements of
solutions used in some key points.
601
Using microsimulation to estimate highway bridge traffic load
C.C. Caprani
Department of Civil & Structural Engineering, Dublin Institute of Technology, Ireland
ABSTRACT
In bridge traffic loading there is an often-made

assumption that free-flowing traffic incorporating
dynamic effects governs for spans up to about 40 m and
that thereafter congested traffic governs. This study
uses traffic microsimulation as a basis for the compar-
ison of regularly-used free and congested traffic flow
models. The free-flow model developed by OBrien and Figure 1. Illustration of the Standard Congestion Model
Caprani (2005) is used together with a synthesized ver- (SCM) for this study.
sion of the congested flow models of the literature,
illustrated in Figure 1. that the application of traffic microsimulation tends to
Traffic microsimulation offers a comprehensive reduce overall lifetime load effect values, in compari-
approach to the modelling of traffic as it models indi- son to the standard free-flow and congestion models.
vidual vehicle and driver behaviour giving a more Lastly, it was found that the governing form of traffic is
realistic picture of traffic states and their consequent sensitive to the values of DAF applied to free-flowing
load effects on bridges. The Intelligent Driver Model traffic. Indeed, even given current DAF values, the gov-
of Treiber et al (2000) is used in this study. erning form of traffic can be less than expected for
In this study, 50 days of traffic is generated for the some load effects (in particular, Load Effect 2). Until
free-flowing models whilst 240 hours of continuous further research is carried out into lifetime values of
traffic is generated for the congested models, repre- DAF, it will be difficult to adequately state governing
senting 10 weeks of data of 5 working days per week. forms of traffic for different bridge lengths. In addi-
For each of the main models traffic compositions of tion, since the governing form of traffic is shown to
0%, 50% and 90% cars are considered. In all studies, depend on the load effect considered, it seems prudent
5 sets of data were generated and processed to ascertain to consider both traffic states in any bridge assessment.
repeatability. This being the case, traffic microsimulation is shown
Three load effects for bridge lengths of 20 to 60 m to be an ideal tool for this purpose.
were calculated both before and after the application
of the traffic microsimulation model. In this way the
impact of traffic microsimulation was examined with REFERENCES
reference to traditional traffic models, whilst keep the
same underlying traffic constant. Caprani, C.C., OBrien, E.J. & McLachlan, G.J. 2008. Char-
The Composite Distribution Statistics (CDS) model acteristic traffic load effects from a mixture of loading
proposed by Caprani et al (2008) is used to extrap- events on short to medium span bridges. Structural Safety
olate the simulated load effects to determine the 30(5): 394404.
characteristic load effects in each case. OBrien, E.J. & Caprani, C.C. 2005. Headway modelling for
traffic load assessment of short-to-medium span bridges.
The microsimulation model results suggests that
The Structural Engineer 83(16): 3336.
standard congestion models are very conservative Treiber, M., Hennecke, A. and Helbing, D. 2000b. Con-
a nominal gap of 15 m may be more appropriate than gestedTraffic States in Empirical Observations and Micro-
a nominal gap of 5 m. It is also found that lifetime scopic Simulations. Physical Review E 62(2): 18051824.
load effect is sensitive to high percentages of cars and arXiv:cond-mat/0002177v2.
602
Pseudodynamic and cyclic tests on reduced-scale pier-deck sub-systems
D. Cardone, G. Perrone & S. Sofia

DiSGG University of Basilicata, Potenza, Italy
M. Dolce
Italian Dept. of Civil Protection, Rome, Italy
L. Pardi
Autostrade per lItalia S.p.A., Rome, Italy
ABSTRACT simply-supported bridges of the A16 (Napoli-Canosa)

Italian highway.
In this paper the experimental program on-going The longitudinal rebars of two slender and two squat
at the Laboratories of the University of Basilicata, pier models, have been artificially corroded through an
within the Line 6 (Seismic Vulnerability of Bridges) accelerated oxidation process, calibrated to get a given
of the SAGGI (Integrated Systems for the Global reduction of bar diameter, corresponding to a corro-
Management of Infrastructures) research project is sion period of approximately 10 years in a medium
described. deterioration ambient.
The DiSGG of the University of Basilicata was The pier shaft has been realised using micro-
involved in the Line 6 of the SAGGI project, dealing concrete, due to the reduced dimensions of the models.
with the seismic risk and vulnerability of the Italian Two concrete mixes were prepared with different
highway bridges. An extensive program of experimen- strength in compression.
tal tests, is being carried out at the Laboratory of The mechanical behaviour of micro-concrete and
DiSGG, in order to assess the seismic behaviour of the steel rebars have been evaluated by experimental
pier-deck sub-systems. tests on specimens taken during the construction of
The test apparatus consists in a very stiff steel struc- the pier models. The experimental program includes
ture, simulating the bridge deck, driven by a double- both cyclic tests at different displacement amplitudes
acting servo-hydraulic dynamic actuator which applies and pseudodynamic tests at different seismic intensity
the horizontal seismic forces/displacements of the levels.
deck. A couple of pseudo-dynamic actuators are used The main parameters of the experimental tests (i.e.
to reproduce the tributary weight of the deck. the displacement amplitudes of the cyclic tests and
Three different types of connection between deck the Peak Ground Accelerations of the pseudodynamic
and pier have been considered, i.e.: (i) rigid connection tests) have been selected based on preliminary simu-
realised by stiff Z-shaped steel sections, (ii) semi- lation analysis, with a numerical model implemented
rigid connection realised by traditional neoprene pads in Opensees.
and (iii) seismically isolated configurations realised The mechanical behaviour of the piers has been
by high damping rubber bearings (HDRB). As far as modeled with nonlinear force-based fiber elements
the bearing devices is concerned, two neoprene pads with distributed plasticity, calibrating the constitu-
replaced from a real viaduct during scheduled mainte- tive laws of steel and concrete based on the experi-
nance services have been used. The isolation devices, mental results of the characterization tests on the
instead, have been purposely designed according to materials.
specific performance objectives. Nonlinear time history analyses have been per-
The testing models consist in 1 : 3-scale RC single formed to evaluate the seismic response of the pier-
shaft piers with hollow circular cross section with 1m deck sub-systems and the cyclic behaviour of the piers
diameter and 125 mm thickness. Eight pier models, to be expected in the pseudodynamic and cyclic experi-
four with 3.2 m effective height and four with 1.7 m mental tests, respectively. Based on the outcomes of
effective height have been realized. The structural these preliminary analyses the program of the experi-
characteristics of the pier models have been derived mental tests on slender and squat piers have been
from the examination of the piers of five multi-span defined.
603
The case study and application of the substructure replacement technique

for bridges having serious scouring of foundation
D.-W. Chang, J.-L. Wang, C.-P. Chang & M.-Y. Yu

CECI Engineering Consultants Inc., Section 2 Hsinhai Road, Taipei 106, Taiwan
ABSTRACT The Substructure Replacement Technique has four

outstanding characteristics:
To resolve the dangerous situation of naked bridge
(1) This technique is capable of replacing naked
foundations, protecting or strengthening measures
and dangerous foundations with new ones, which
have to be taken or new bridges have to be rebuilt.
makes this method distinct.
Normally, protecting or strengthening measures can
(2) This technique utilizes jacks to transfer bridge
be classified into 3 groups:
loadings to temporary shoring structure. The tem-
(1) Riverbed stabilization structures, such as gabions, porary shoring structure not only sustains all
concrete blocks, weirs and submerged sills, can the loadings from superstructure but also min-
help to slow down scour tendency. imizes deviation displacement before and after
(2) Foundation protecting methods, such as gabions, construction, which is indeed a challenge.
concrete encasement and surrounding PC piles, (3) The temporary shoring structure is connected with
can provide temporary protection with little the steel columns embedded in the new con-
strengthening effects for foundation. structed RC piles, which is stable and safe. The
(3) Foundation strengthening methods, such as temporary shoring structure for every pier is the
enlarging foundations and increasing piers, can same, which means the steel beams and angles can
provide adequate strength for foundation but foun- be recycled and the project will be cost effective.
dations are still vulnerable to flood due to their (4) Only exposed foundations are demolished and
huge volume after retrofit. consolidated, and the original superstructure, pier
cap and part of the pier are retained. This enhances
Each protection or strengthening technique has its
seismic capacity of the bridge and flood con-
applicable situation.
veyance ability of the river at least expense.
Under the consideration of original bridge structure
type, river hydraulic characteristics, nice appearance, The innovative Substructure Replacement Tech-
traffic maintenance, demand of hydraulic engineering nique demolishes exposed foundations and constructs
and superstructure integrity, an innovative Substruc- new piers and foundations which meet seismic and
ture Replacement Technique has been proposed to hydraulic requirements. Because the bridge loadings
deal with the seriously exposed bridge foundations in are transferred to the temporary shoring structure,
Taiwan. By transferring the loading of superstructure the original flow of traffic can be maintained and
from original bearings to temporary shoring struc- the deviation displacement before and after construc-
ture, the Substructure Replacement Technique not tion is negligible. For naked bridge foundations, this
only ensures the safety and integrity of superstructure cost effective technique is indeed a promising solu-
and pier caps but also maintains the original traffic tion for promoting the performance of bridges at least
function. After replacing damaged foundations with expense.
new ones built below riverbeds and constructing new
columns smoothly connected with the original pier
caps, the safety of the bridge can be restored.
REFERENCE
The innovative Substructure Replacement Tech-
nique was adopted by the first time in Taiwan to cope Chang Dyi-Wei, Failures of bridges caused by Herb
with naked foundations of the Shi-chou Bridge. Total Typhoon, Discussion on the scheme for infrastructures
19 piers from P38 to P56 are strengthened using this from the disaster caused by Herb Typhoon, National
technique. Taiwan University, Oct., 2006. (in Chinese).
604
Bridge maintenance prioritization through visual inspection results
M.H. Chen
China Engineering Consultants, Inc., Taiwan
H.C. Tsai & C.Y. Wang

National Central University, Taiwan
M.C. Chen
Ching-Yun University, Taiwan
ABSTRACT
The current bridge visual inspection system in Taiwan

was introduced in this paper. According to the col-
lected visual inspected information, a further analysis
was adopted to propose a criterion for prioritization
of bridge maintenance. A bridge is composed of 20
components which are evaluated with DERU method
(see Table 1). For each damaged component, corre-
sponding maintaining methodology is recommended
and its amount is estimated. All of the healthy con-
ditions of bridge components are summarized as a
structural index, combining the other three indices of
serviceability index, vulnerable index and distinc-
tive index into a comprehensive Maintenance Priority
Index (MPI) result. Three different field cases were
discussed in this paper. A list of emergent bridges was
clearly observed as getting their MPI rating results.
Consequently, the optimum bridge maintaining strat-
egy was promising to be set by sorting the priority Figure 1. Comparative drawings of MPI distribution for
with MPI rating, and the limited resource can meet the Yilan county, Taoyuan county and entire Taiwan.
largest benefit.
The prioritization method has been applied onto REFERENCES
three different field cases. In Figure 1, all the cases
FHWA, 1995, Local Highway Bridge Replacement and
curves (Yilan county case, Taoyuan county case and
Rehabilitation Program, USA.
entire Taiwan case) were presented in the same figure. FHWA, 1995, NBI-Recording and Coding Guide for
Because the case III was based on worse initial con- the structure Inventory and Appraisal of the National
ditions with the other two, it can be seen, as a result, Bridges, USA.
most bridges come out with dramatically higher MPI Chiang, W. L., and YAU, N. J., and WANG, C. Y., 2004,
values. Important Ranking and Prioritization Index for Taiwan
Areas Bridges, Directorate General of Highways, MOTC,
R.O.C. (in Chinese).
Table 1. The criteria of DERU.
Grade 0 1 2 3 4
Degree (D) None Good Failure Slightly Failure Failure Seriously

Extent (E) Unable Inspection <10% 10%30% 30%60% >60%
Relevancy (R) Unable Determine Slight Small Middle Large
Urgency (U) Unable Determine Routine Maintain Within 3 years Within 1 year Emergency
605
Time-dependent reliability analysis of systems with repairable

or non-repairable components
T. Cheng & M.D. Pandey

Department of Civil and Environmental Engineering, University of Waterloo, Waterloo, Ontario, Canada
ABSTRACT time horizon. In this sense, repair is an effective way

to maintain the functionality of the system, which is
The paper presents an exposition of time-dependent well known to engineers. For this reason, critical civil
system reliability analysis, and it illustrates several engineering systems, such as bridges, pipelines and
distinctions between the analysis of the system with transportation systems, are designed and operated with
repairable and non-repairable components. repairable components. This paper presents an analyti-
In case of a system with non-repairable compo- cal consistent approach to evaluate life-cycle cost. It is
nents, the computation of the cumulative distribution important to emphasize that in such cases an implicit
(CDF) of the time to failure is conceptually straight use of non-repairable system reliability analysis will
forward. However, in case of a system with repairable lead to incorrect results.
components, the concepts of failure rate and sys- The method presented in this paper is generic, and it
tem unavailability are of relevance for life-cycle cost can be easily extended to other types systems subjected
analysis. The computation of failure and repair rates to more involved inspection and maintenance policies
as well as unavailability is complicated, because it (Pandey et al. 2008).
involves convolutions of distributions of time to failure
and time to repair. Nevertheless, given the advance-
ments in computing, these quantities can now be easily REFERENCES
computed.
Birolini A. 1999. Reliability Engineering: Theory and Prac-
The paper also formulates the evaluation of life tice. Berlin: Springer.
cycle cost for both types of system. The analysis shows Pandey, M.D., & Cheng, T., & Xie, W.C. 2008. Stochastic
that non-repairable systems are costly over lifetime, renewal process model for infrastructure asset manage-
because their failure will result in the loss of ser- ment. Inaugural International Conference of the Engi-
vices provided by the structure for the rest of the neering Mechanics Institute.
606
An empirical life-cycle assessment of the relative cost-effectiveness

of alternative materials for reinforcing bridge decks
A. Cope & S. Labi

Purdue University, W. Lafayette, Indiana, USA
C. Fang
Applied Research Associates, Inc., Champaign, Illinois, USA
ABSTRACT when its unit price is approximately 7 times more than

that of traditional carbon steel.
This paper compares the cost-effectiveness of alter- In conclusion, the research showed that using SS
native bridge deck reinforcement materials. First, the generally leads to significantly higher initial costs but
life-cycle activity profiles associated with each alter- drastically reduced costs over the bridge life cycle, par-
native are established on the basis of empirical data. ticularly when user costs are considered in the analysis.
The profiles highlight the differences in the frequency In environments that are more vulnerable to the effects
of life cycle interventions (deck replacements and deck of corrosion, the relative superiority (in terms of cost-
rehabilitations) which translate into differences in the effectiveness) of stainless steel is expected to be even
user cost of the alternative reinforcement materials higher. In the current national environment that is char-
a profile with more interventions has more frequent acterized by uncertainty of sustained funding and high
workzones and subsequently, higher user cost. Thus user expectations, technologies that yield longer last-
the user cost of each alternative was calculated. Also, ing and superior performing infrastructure are useful
the agency cost of initial construction and subse- for the operations of transportation agencies and to the
quent rehabilitation were estimated. Finally, the cost- wider society in general.
effectiveness of each material alternative was analyzed
over the entire life cycle using the equivalent uniform
REFERENCES
annual cost (EUAC) which converts all cash flows to
a yearly equivalent amount over a specified analysis Yunovich, M., Thompson, N. G., Balvanyos, T., Lave, L.,
period, for each alternative. The alternative with the Highway Bridges, Appendix D, Corrosion Cost and Pre-
lower EUAC is deemed the more economically effi- ventive Strategies in the United States, FHWA-RD-01-
cient. To test the robustness of the evaluation outcome, 156, McLean, VA., 2002.
the key input variables were incrementally adjusted Li, S. Wang, M., Li, S. Model for cover cracking due to cor-
and the impact of the relative cost-effectiveness was rosion expansion and uniform stresses at infinity, Appl.
observed. The key variables included the durations Math. Modeling 32 (7), pp. 14361444, 2008.
Berra, M., Castellani, A., Coronelli, D., Zanni, S., Zhang, G.
of initial construction project, deck replacement, and
Steelconcrete bond deterioration due to corrosion: finite-
deck rehabilitation; traffic volume; workzone speed; element analysis for different confinement levels, Mag. of
deck length and deck width, relative unit price of the Concrete Res. Vol. 55 (3), 237247, 2003.
reinforcement materials; discount rate; detour length FHWA. Corrosion Evaluation of Epoxy-Coated Metallic
and speeds where applicable; vehicle occupancy; and Clad and Solid Metallic Reinforcing Bars in Concrete.
minimum hourly wage, average fuel economy; and fuel U.S. DOT, FHWA-RD-98-153, 1998.
price. Cross, W., Duke, E., Kellar, J., Han, K., and Johnston, D.
The results suggest that solid stainless steel (SS) is Stainless Steel Clad Rebar in Bridge Decks. South Dakota
the more economically efficient reinforcement option. Department of Transportation, 2001.
Hartt, W., Rodney, P., Lynogorski, D., Liroux, V., Virmani,
However, for bridges with very low traffic volumes,
Y. P. Corrosion Resistant Alloys for Reinforced Concrete.
the attractiveness of SS is far lower than that for FHWA, 2007.
high-volume bridges. Also, it was determined that as Hawk. H. Bridge Life-Cycle Cost Analysis, NCHRP Report
user costs increase, the relative attractiveness of solid 483 Transp. Res. Board, Washington DC, 2003.
stainless steel also increases by a certain rate. The sen- Xi, Y., Abu-Hejleh, N., Asiz, A., & Suwito, A. Performance
sitivity analysis for reinforcement price showed that Evaluation of Various Corrosion Protection Systems of
solid stainless steel is generally more attractive unless Bridges in Colorado, Colorado DOT, Denver, CO, 2004.
607
A procedure to derive probabilistic fatigue strength data for riveted joints
J.A.F.O. Correia, A.M.P. de Jesus, A.L.L. da Silva & J.F.N. da Silva

School of Sciences and Technology, University of Trs-os-Montes and Alto Douro, Vila Real, Portugal
IDMEC-FEUP, UCVE, Porto, Portugal
ABSTRACT
The availability of probabilistic fatigue strength data

for riveted connections is mandatory to carry out relia-
bility analysis of ancient riveted bridges. A procedure
to derive probabilistic S-N fields (P-S-N fields) for
riveted connections is proposed. Strain-life data as
well as fatigue crack growth data from plain material
are used to compute the total fatigue life of a sim-
ple riveted connection (Fig. 1), integrating both local
and fracture mechanics approaches. The basic fatigue Figure 1. Riveted joint (dimensions in mm).
data is inputted in the probabilistic form as well as
some parameters of the model with higher uncertainty
(e.g. crack initiation size). The probabilistic strain-life
model as proposed by Castillo and Fernndez-Canteli
(2009) was used to compute the crack initiation. In
order to model the crack propagation, the Pariss law
(Paris & Erdogan, 1963) was used with fixed expo-
nent m and assuming a log-Normal distribution for
the C coefficient. A three dimensional finite element
model of the riveted joint is proposed in order to assess
the local stresses/strains at the critical location. This
model is able to account the clamping effects of the
rivet on local stresses/strains. The clamping stresses
on rivet and friction are assumed random variables, Figure 2. Predicted P-S-N field for the riveted connection.
following triangular probability density distributions.
Details about the experimental data (plain material and
REFERENCES
riveted connection) and numerical modeling can be
found in Silva (2009). Castillo, E. & Fernndez-Canteli, A. 2009. A Unified Statisti-
The probabilistic inputs are accounted in the fatigue cal Methodology for Modeling Fatigue Damage, Springer.
modeling procedure using the Monte Carlo sampling Paris, P.C. & Erdogan, F. 1963. A critical analysis of crack
technique. Figure 2 illustrates the P-S-N field com- propagation laws. Transactions of The ASME. Series E:
puted for the riveted connection and compares it with Journal of Basic Engineering 85: 528534.
the experimental data illustrating the very satisfac- Silva, J.F.N. 2009. Comparison of the fatigue behavior
tory performance of the model. All experimental data between riveted and bolted connections, MSc. Thesis,
University of Trs-os-Montes and Alto Douro, Portugal
falls within a 98% probability of failure band and is (in Portuguese).
symmetrically distributed around the 50% percentile.
The P-S-N field was derived from 10000 fatigue life
calculations, for each stress range applied to the con-
nection, using randomly selected inputs, respecting the
respective statistical distributions.
608
Uncertainty evaluation of reinforced concrete structures behavior
P.J.S. Cruz, I. Valente, J.C. Matos & J. Batista

ISISE Institute for Sustainability and Innovation in Structural Engineering, University of Minho, Guimares,
Portugal
ABSTRACT
When evaluating the behavior of reinforced concrete

structures, as several bridges, we should take into
consideration that the majority of parameters, like
used materials and geometry, vary along the struc-
ture. This article presents a methodology, that takes
this fact into consideration, and which purpose is to
characterize, in a consistent way, if any reinforced con-
crete bridge presents a behavior within the expected
one, or not. The methodology is divided on following
steps: 1) Develop a calibrated deterministic numerical
model; 2) Determine random distribution function for
each input parameter; 3) Develop a non linear prob-
abilistic analysis; 4) Calculate a liability index which Figure 3. Simply supported beam results.
relates, in a consistent way, the proximity of numerical
and experimental data.
Two sets of laboratory tested beams, with dif-
ferent support conditions, simply supported in one
batch (Figure 1) and mixed supported on the other
(Figure 2), are firstly analyzed by this methodology
(Matos et al., 2008). Those beams were executed at
same time, presenting different reinforcement typolo-
gies and concrete covering. Main conclusions of this
Figure 4. Mixed supported beam results.
first analysis are the applicability of purposed method-

ology (Figure 3 and 4) and the importance of gathering
Figure 1. Simply supported beam. any data from analyzed reinforced concrete structures.
The behavior of reinforced concrete bridges can be so
evaluated through this methodology, presenting it, a
great utility for their safety evaluation by detecting
any abnormal behavior.
REFERENCE
Matos, J.C., Valente, I. e Cruz, P. 2010. Avaliao de
incertezas no comportamento at rotura de vigas de beto
armado, ASCP09 1o Congresso de Segurana e Con-
servao de Pontes, ISBN 978-989-20-1559-0, Lisboa,
Figure 2. Mixed supported beam. pp. 512.
609
Toward a more rational serviceability considerations for high performance

steel bridges
S. Darjani, M.A. Saadeghvaziri & S. Saigal

Department of Civil and Environmental Engineering, NJIT, Newark, New Jersey, USA
N. Aboobaker
New Jersey Department of Transportation, Trenton, New Jersey, USA
A. Khan
Johnson, Mirmiran & Thompson, Inc., Trenton, New Jersey, USA
ABSTRACT There are a few alternate design methods which

have been developed to better address serviceability
There have been significant advances in develop- and durability issues. However, these methods have
ment of high performance materials over the past two not been adopted in part due to practical limitations in
decades. High Performance Steel (HPS) is an exam- their application but mostly due to lack of consensuses.
ple that provides up to 18% cost savings and up to Prior studies have suggested that speed parameter is
28% weight savings when compared with traditional the most important parameter in determining bridge
steel bridge design materials. HPS offers higher yield dynamic response under moving truck load. In this
strength, enhanced weldability, and improved tough- study, bridge dynamic responses (deflection & accel-
ness which results in smaller cross sections, lighter and eration) versus speed parameter are investigated for
much more economical designs. On the other hand, several cases and a good approximation of bridge
smaller cross section leads to more flexible bridges response in terms of speed parameter is presented.
that do not satisfy the existing serviceability deflection This paper includes initial results of an extensive
criteria. analytical study to develop more rational serviceabil-
AASHTO Standard Specifications limit live load ity and durability criteria.The analytical study employs
service deflection to L/800 for general bridges and 2-D and 3-D Finite Element (FE) models to evalu-
to L/1000 for bridges that are used by pedestrians. ate dynamic response of bridges under moving truck
These limits were employed to avoid undesirable load. Results are compared to special cases where
structural and psychological effects due to their defor- exact solutions exist. The results are in agreement
mations. However, results of prior studies indicate with the exact solutions. Currently, parameter study is
that deflection and L/D limits do not necessarily being performed. Among parameters considered are
address these objectives. Existing limits do not pre- truck speed, number of spans, damping ratio, num-
vent damages in structures because they check global ber of trucks, and spatial effects. The paper will also
deflection, while the damages are a consequence of present vibration criteria of other available countries
local deformations such as connection rotations and specifications such as Canada and Europe.
twisting of floor-beams relative to support members.
Furthermore, human susceptibility is more influenced
by derivatives of deflection rather than the deflection REFERENCES
itself. Thus, there is a need for development of a more
rational serviceability criterion which is the objective AASHTO, 2007. AASHTO LRFD Bridge Design Specifica-
tions, 4th Ed., American Association of State Highway
of this study.
and Transportation Officials, Washington, D.C. Interim
Control of undesirable psychological effects on Revisions, 2008.
passengers and pedestrians is apparently one of the Roeder, C. W., Barth, K. E., Bergman, A., and Christopher,
primary reasons for AASHTO deflection limits. How- R. A., 2001. Improved live-load deflection criteria for
ever, prior research work indicates that it is not just steel bridges. National Cooperative Highway Research
live load defection but vertical acceleration and bridge Program, NCHRP Interim Rep. 20-07/133, National
dynamic characteristics (e.g., frequency) that control Research Council, Washington, D.C.
vibration and human perception.
610
Refined load rating analysis of in-service bridges in North Carolina, USA
S. Das
URS Corporation, Morrisville, North Carolina, USA
ABSTRACT removal of posting limits on specific bridges. How-

ever, there are some instances where the actual bridge
In order to conduct an assessment of the actual load condition supported by the field load test data justi-
carrying capacity of the states in-service bridges, fies lowering the rating factors thereby, recommending
North Carolina Department of Transportations Bridge either a decrease of existing posting limits or even
Management Unit has identified several bridge struc- posting of certain bridges that are currently not posted.
tures to perform refined load rating using a combina-
tion of 3-D Finite Element Analysis and Diagnostic
Load Testing. The short-listed bridges carry primary REFERENCES
routes and are mostly constructed prior to the 1950s.
The majority of these bridges require posting for North AASHTO, Manual for Condition Evaluation of Bridges, 2nd
Edition, 1994, with interim revisions thru 2003.
Carolina legal loads. Bridge types included in this
URS Corporation (URS), Report on Bridge Load Rating
project are single or multi-span, skewed or square, Evaluation through Diagnostic Load Testing and Finite
reinforced concrete deck girders, concrete slabs, steel Element Analysis of Bridge No. 350032 in Gaston County,
girders, and three sided reinforced concrete culverts. NC, September 2008.
The refined load rating analyses have been completed URS Corporation (URS), Report on Bridge Load Rating
on seven bridges, each having its unique geometric, Evaluation through Diagnostic Load Testing and Finite
material and structural characteristics. Element Analysis of Bridge No. 120078 in Cabarrus
The general procedure for evaluating the current County, NC, September 2008.
load ratings by determining the actual live load carry- URS Corporation (URS), Report on Bridge Load Rating
Evaluation through Diagnostic Load Testing and Finite
ing capacity of the bridge initiates with the develop-
Element Analysis of Bridge No. 250060 in Cumberland
ment of a 3-D finite element (baseline) model of the County, NC, September 2008.
bridge superstructure, followed by field load testing of URS Corporation (URS), Report on Bridge Load Rating
the bridge, then utilizing the field measured strain and Evaluation through Diagnostic Load Testing and Finite
displacement data to validate and refine the baseline Element Analysis of Bridge No. 560008 in Madison
model and finally, using this refined model to com- County, NC, October 2008.
pute the live load ratings of the bridge. The diagnostic URS Corporation (URS), Report on Bridge Load Rating
load testing of the bridge enables to capture the real- Evaluation through Diagnostic Load Testing and Finite
time live load response characteristics and its spatial Element Analysis of Bridge No. 770042 in Robeson
County, NC, October 2008.
distribution.
URS Corporation (URS), Report on Bridge Load Rating
Live load ratings for the North Carolina bridges Evaluation through Diagnostic Load Testing and Finite
computed according to this refined methodology pro- Element Analysis of Bridge No. 690001 in Pasquotank
vide for a realistic estimate of the reserve load carrying County, NC, January 2009.
capacity of the structures. In most cases, the results of URS Corporation (URS), Report on Bridge Load Rating
the refined rating procedure indicate that rating factors Evaluation through Diagnostic Load Testing and Finite
may increase by up to 75% when compared with results Element Analysis of Bridge No. 800055 in Rutherford
of the conventional analysis procedures thereby, rec- County, NC, March 2009.
ommending either a substantial increase or complete
611
Designing the rehabilitation of the Henley Street Bridge in Knoxville,

Tennessee, USA
S. Das
URS Corporation, Morrisville, North Carolina, USA
ABSTRACT expansion joints located only at the abutments. How-

ever, the continuity of the superstructure resulted in
The existing Henley Street Bridge is a six-span, 423 m increased force and deformation demands due to inter-
long open spandrel reinforced concrete arch bridge action of multi-span arches subjected to combined load
flanked by 50 m long, three-span approach girder effects at various sections of the bridge structure. A
structures at each end. The arch span lengths range combination of innovative design techniques are used
from 56 m to 97 m with an average rise to span ratio to mitigate these adverse load effects.
of 0.30. The bridge construction work is scheduled to com-
The existing bridge deck contains 26 expansion mence in 2010. The estimated cost of the rehabilitation
joints that are believed to be the primary sources for work is $30 million (US currency). When completed,
water intrusion causing concrete deterioration in the the structure will be the longest jointless multi-span
elements of the arch structure during its 79-year life. open spandrel deck arch bridge in the world.
In the present condition of the bridge, the arch ribs,
and several spandrel columns do not rate for the design
vehicle HS-20. Further, the arch ribs at the spring lines REFERENCES
and spandrel columns located near the span quarter
points and arch crowns exhibit seismic demands in AASHTO. 1996. Standard Specifications for Highway
Bridges, 16th Edition 1996.
excess of their respective structural capacities. Based
Lwin, M.M. 2008. Sustainability Considerations in Bridge
on a detailed condition assessment of the existing Design, Construction and Maintenance US DOT/FHWA,
structure, and the results of analysis, a decision was Aspire, Winter 2008.
made to salvage the arch ribs, the arch piers, the pier McCullough, C.M. & Thayer, E.S. 1948. Elastic Arch
foundations, the approach curtain walls and portions Bridges, 2nd Edition, 1948.
of the abutments. The existing superstructure deck and Newmark, N.M. 1938. Interaction between Rib and Super-
the spandrel columns would be replaced as a part of structure in Concrete Arch Bridges, Transactions, Ameri-
the rehabilitation work. can Society of Civil Engineers, 103: 6280, 1938.
Special emphasis was given to arrive at a design Tennessee Department of Transportation. 2005. Bridge Foun-
dation Report Henley Street Bridge (US 441) over
solution that would address some of the pertinent
Tennessee River, Knox County, State Project Number
elements of a sustainable design process. The key 47017-1213-94, November 2005.
areas that were considered include enhancing mobil- URS Corporation. 2000. Bridge Inspection Report Bridge
ity, durability and safety, utilizing high performance on US441/SR33 over the Tennessee River and SR58 in
materials, applying context sensitive solutions, and Knoxville, Tennessee, State Project Number 47017-1213-
complying with environmental and preservation laws 94, Federal Project Number BH-NH-33(17), May 2000.
and regulations. URS Corporation (URS), Seismic Evaluation Report - Bridge
Built in 193032, the bridge has been determined on US441/SR33 over the Tennessee River and SR58 in
eligible for listing in the National Register of Historic Knoxville, Tennessee, State Project Number 47017-1213-
94, Federal Project Number BH-NH-33(17), June 2002.
Places. Therefore, the bridge improvements have been
URS Corporation (URS), Structure Evaluation Report
designed in accordance with the National Historic Bridge on US441/SR33 over the Tennessee River and
PreservationAct and the Department ofTransportation SR58 in Knoxville, Tennessee, State Project Number
Act of 1966. 47017-1213-94, Federal Project Number BH-NH-33(17),
Further, in order to arrive at a durable bridge April 2004.
structure that would be capable of handling one addi-
tional lane of traffic, a 26 m wide (out-to-out), 523 m
long continuous superstructure deck is designed with
612
Stress intensity factors evaluation for riveted beams applying

FEA with VCCT
A.M.P. de Jesus & J.A.F.O. Correia

University of Trs-os-Montes and Alto Douro, Vila Real, Portugal
IDMEC-FEUP, UCVE, Porto, Portugal
ABSTRACT
Structural integrity assessments of old steel riveted

bridges are more and more frequent. Most of these
structures were built at the end of the 19th century
or the beginning of the 20th century with angles and
plates joined by rivets and made of puddle iron or
wrought steel. Fatigue is one major concern for these
structures since they show a long operational period
with increasing traffic intensity, many times without
the required rehabilitation procedures. Despite the S-N
approach is widely used to assess the fatigue dam-
age for riveted steel constructions, Fracture Mechanics
appears as an alternative to perform residual life calcu- Figure 2. Stress intensity evolution for the cracked beam
lations. However, the use of the Fracture Mechanics is under the action of a bending moment.
very often limited to the application of simplified for-
mulae for stress intensity factors evaluation, available beam. The beam was subjected to both axial loads
in standard handbooks that may lead to inconsistent (N) and bending moment (M). A 3D finite element
fatigue life predictions. model of the riveted beam was proposed to compute
A methodology for the evaluation of stress intensity the stress intensity factors for a crack propagating at
factors of cracked riveted members, based on detailed the plate web, using the Virtual Crack Closure Tech-
3D finite element models, is proposed. Despite being nique (VCCT) (Krueger 2004). The clamping stresses
generally enough to be applied to other riveted of rivets are accounted in the model through a temper-
geometries of interest, the proposed methodology is ature variation. Figure 2 illustrates the stress intensity
demonstrated for a riveted T beam, also investigated factor variation for the riveted beam under a bending
by Moreno & Valiente (2004) using an analytical moment, M .The stress intensity factor was normalized
approach. Figure 1 illustrates the geometry of the using the maximum bending direct stress ( = Mb/2I ).
After an initial increase in the stress intensity factor,
a reduction is observed until zero. This reduction is
accompanied with an increase in tensile stresses in the
angles near to the crack.
REFERENCES
Moreno, J., Valiente, A. 2004. Stress intensity factors in
riveted steel beams. Engineering Failure Analysis 11:
777787.
Krueger, R. 2004. Virtual crack closure technique: History,
approach, and applications. Applied Mechanics Reviews
57(2): 109143.
Figure 1. Geometry of the cracked riveted T beam

(b = 0.5 m).
613
Deterioration and maintenance of RC bridge decks under uncertainty:

Condition and reliability indicators
A. Dec & D.M. Frangopol

ABSTRACT In this paper, condition and reliability indices

are directly related by using a corrosion model that
In this paper, a condition-based approach describing affects both. In fact, spalling of concrete surface is
lifetime deterioration of RC bridge decks is proposed. a direct consequence of active corrosion of the steel
Time-dependent performance indicators such as con- reinforcement.
dition and reliability indices are evaluated over time in The applicability of this approach has been proved
order to develop a rational bridge maintenance strat- by studying an existing RC bridge deck located near
egy. A probabilistic analysis considering the condition Wausau, Wisconsin, USA. The results obtained have
index a non-linear function of time is introduced. The highlighted that maintenance planning can be devel-
condition index represents the percentage of dete- oped based on visual inspections data and reliability
riorated bridge deck area (concrete spalling due to can be predicted.
environmental attack). The reliability index can be
obtained by considering the flexural failure mode
described by its limit state function. REFERENCES
The proposed approach is condition-based. This
means that condition index governs the decision pro- Frangopol, D.M. & Curley, J.P. 1987. Effects of damage and
redundancy on structural reliability. Journal of Structural
cess in terms of maintenance assessment. Moreover,
Engineering. 113(7): 15331549.
through the application of this approach, visual inspec- Frangopol, D.M., Kong, J.S. & Gharaibeh, E.S. 2001.
tions are able to predict the residual lifetime of the deck Reliability-based life-cycle management of highway
depending on the inspected deck location (top or bot- bridges.Journal of Computing in Civil Engineering. 15(1):
tom of the deck, right or left lane of traffic). Figure 1 2734.
shows the mean of the condition index over time under Frangopol, D.M., Kallen, M-J. & van Noortwijk, J.M. 2004.
maintenance. Initially, the condition index is perfect Probabilistic models for life-cycle performance of deteri-
(i.e., zero) until cracks propagate to the concrete sur- orating structures: review and future directions. Structural
face; when service life is reached, the condition index Engineering and Mechanics. 130(12): 18651874.
Gaal, G.C.M. 2004. Prediction of deterioration of concrete
starts to deteriorate due to crack propagation.
bridges. PhD thesis, Delft: Delft University Press.
Kim, S. & Frangopol, D.M. 2009. Optimal planning of
structural performance monitoring based on reliability
importance assessment. Prob. Engin. Mech. 25(1): 8698.
Neves, L.A.C. & Frangopol, D.M. 2005. Condition, safety
and cost profiles for deteriorating structures with empha-
sis on bridges. Reliability Engineering and System Safety.
89(2): 185198.
van Beek, A., Gaal, G.C.M., van Noortwijk, J.M. & Bakker,
J.D. 2003. Validation model for service life prediction of
concrete structures. 2nd International RILEM Workshop
on Life Prediction and Aging Management of Concrete
Structures 257267, Paris.
van Noortwijk, J.M. & Frangopol, D.M. 2004. Two prob-
abilistic life-cycle maintenance models for deteriorating
civil infrastructures. Probabilistic Engineering Mechan-
ics. 19(4): 345359.
Figure 1. Profile of mean of the condition index over
80 years with maintenance.
614
Multi directional hysteretic damper with adaptive post-elastic stiffness

for seismic protection of bridges in near fault zones
M. Dicleli & A. Salem-Milani

Middle East Technical University, Ankara, Turkey
ABSTRACT slight flexibility of supporting structure will cause

small bending in the cores which are supposed to
This paper presents a new hysteretic damper suitable be subjected to pure torsion, and thus cause some
for seismic protection of highway bridges in near- non-uniformity in strain distribution in case of elasto-
fault (NF) zones. The damper is a Multi-directional plastic material model with zero plastic hardening.
Torsional Hysteretic Damper (MTHD). The MTHD is However, small plastic hardening in steel will ensure
made of eight steel cylinders named yielding cores, uniform distribution of plastic strains over the yielding
symmetrically arranged, which are designed to yield part of the yielding cores. The cores are thus judged to
in torsion. The torsion is created by arms attached to be capable of stable cyclic performance provided that
the cores. The torsional arms are guided by a rail sys- suitable grade of steel is used.
tem at top of the device which is designed to convert To study the effect of post-elastic hardening fea-
planar multidirectional displacements of the deck (to ture of MTHD on the performance of a typical bridge
which the rail system is attached) into rotation in the under NF ground motion, nonlinear time history
arms and twisting in the cores. All eight yielding cores (NLTH) analyses are conducted. NLTH analyses of
are supported against bending by a single stiff central seismic isolated bridges with MTHD and with hys-
column. Easily adjustable arm length allows for easily teretic dampers having elasto-plastic behavior, using
adjustable reaction force and maximum allowable dis- NF ground motions revealed the adaptive behavior of
placement of MTHD. This unique working mechanism the device which is a result of this gradual harden-
of the device makes it capable of large displacement ing feature, such that at lower displacements (DBE),
capacities. More importantly, the damper possesses a force levels are close to a regular system while at high-
hyperbolically varying post-elastic stiffness as a result est levels of displacements (MCE) when displacement
of its special working mechanism which creates a geo- control becomes critical in preventing the deck from
metric hardening effect. The variable stiffness is found falling off of the supports, the device hardens to make
to be effective in limiting the lateral displacement of the substructure yield and limit decks displacement at
the seismic-isolated bridge decks in near-fault zones any direction. The results of sensitivity analyses, per-
and is the focus of this paper. formed to study the effect of the MTHD on seismic
Brief illustrated description of the device is given performance of NF bridges in relation to the ground
and the associated analytical formulations of force- motion magnitude and characteristic strength of the
displacement response of MTHD are briefly intro- energy dissipation device, are presented.
duced. Since stable function of metallic dampers rests The paper is devoted to analytical issues only. Tests
on the cyclic performance of energy dissipaters at are under way in the Engineering Sciences Department
their repeated plastic-strain excursions, finite element of Middle East Technical University to verify MTHDs
analysis of yielding cores (energy dissipater unit of working mechanism, results of analytical formulations
MTHD) was performed to study the strain distribu- and cyclic performance of the device which will be
tion over the energy dissipater. The analysis show that presented in subsequent publications.
615
Distribution of live load effects in integral bridge abutments and piles
M. Dicleli & S. Erhan

ABSTRACT and piles different than that calculated using the live
load distribution equations developed for the girders of
Structural analysis of highway bridges using compli- jointed bridges. Therefore, using AASHTO live load
cated 3-D finite element models (FEMs) to determine distribution equations may results in either conserva-
live load effects in bridge components is possible due tive or unconservative estimates of the live load effects
to the readily available computational tools in design in the piles and abutments of integral bridges. Thus, in
offices. However, throughout the design process, using this study, live load distribution equations for integral
such complicated methods is tedious, time consuming bridge substructures are developed. For this purpose,
and expensive. Therefore, most design engineers pre- numerous 3-D and corresponding 2-D structural mod-
fer using simplified 2-D structural models of bridges els of typical integral bridges are built and analyzed
and live load distribution equations available in cur- under AASHTO live load. In the analyses, the effect
rent bridge design codes such as AASHTO (American of various superstructure and substructure properties
Association of State Highway Transportation Offi- such as span length, girder spacing, girder stiffness,
cials) LRFD (Load and Resistance Factor Design) abutment height, pile size, pile spacing and foundation
Bridge Design Specifications (2007) to determine live soil stiffness are considered. The results from the 2-D
load effects in bridge components. In AASHTO LRFD and 3-D analyses are then used to calculate the live
Bridge Design Specifications, live load distribution load distribution factors for the abutments and piles
equations are available only for the girders of jointed of integral bridges as a function of the above men-
bridges. AASHTO does not have any provisions for the tioned properties. live load distribution equations are
calculation of live load effects in integral bridge com- then developed to estimate the live load moments and
ponents including the girders, abutments and piles. shear in the abutments and piles of integral bridges
Consequently, these live load distribution equations using these live load distribution factors and nonlinear
are also used for designing the girders of integral regression analysis methods to address the above men-
bridges. In addition, most design engineers generally tioned uncertainties and to provide useful tools to the
calculate the live load effects in the abutments and piles bridge engineering community at large for the design
of integral bridges by using the AASHTO live load dis- of integral bridge abutments and piles under live load
tribution equations developed for the girders of jointed effects. It is observed that the developed live load dis-
bridges. This approach is based on the assumption that tribution equations yield a reasonably good estimate
the same rotations about a transverse axis perpendicu- of live load moment and shear in the abutments and
lar to the longitudinal direction of the bridge occur both piles of integral bridges.
in the abutments and the girders under live load due
to the monolithic construction of the superstructure-
abutment joint in integral bridges. However, it is REFERENCE
anticipated that the concentrated rigidity of a particular
girder combined with those of the adjacent girders con- AASHTO 2007. LRFD Bridge Design Specifications, 4th
Edition. Washington, D. C., USA.
nected to the abutment having a smeared rigidity, may
produce a live load distribution within the abutment
616
Static and dynamic model validation and damage detection

using wireless sensor network
S. Dorvash, S. Pakzad, R. Yao & K. Okaly

ABSTRACT Ljung, G. M. and Box, G. E. P. 1978. On a Measure of Lack

of Fit in Time Series Models, Biometrika, Vol. 65, No.2,
Structural health monitoring (SHM) is brought to 297303.
attentions due to aging of civil infrastructures and Overschee, P. V. Moor, B. D. 1994. N4SID: Sub-
space Algorithms for the Identification of Com-
their undeniable effect in the public safety. Tradition-
bined Deterministic-Stochastic Systems. Automatica
ally, SHM has been developed based on the wired ISSN 0005-1098 CODEN ATCAA9, vol. 30, no. 1,
network of sensors and centrally depository of data. pp. 7593.
While, the high costs and the installation difficul- Pakzad, S.N. 2010. Development and deployment of large
ties associated with the wired sensor network have scale wireless sensor network on a long-span bridge.
limited the large-scale application of SHM, technolog- Smart Structures and Systems, An International Journal,
ical advances in wireless sensor network (WSN) have in press.
made the SHM more affordable and potentially scal- Pakzad, S.N., and Fenves, G.L. 2009. Statistical analysis
able. In some recent SHM projects WSN is selected of vibration modes of a suspension bridge using spa-
tially dense wireless sensor network. ASCE Journal of
as a promising data acquisition system which implies
Structural Engineering, 135(7):863872.
the progressive trend of using WSN in SHM. Pakzad, S.N., Fenves, G.L., Kim, S., and Culler, D.E. 2008.
This paper presents a set of static and dynamic Design and Implementation of Scalable Wireless Sen-
tests and damage detection on a 3-dimensional steel sor Network for Structural Monitoring. ASCE Journal of
truss utilizing WSN. The Imote2 platform, developed Infrastructure Engineering, 14(1):89101.
by Intel, together with SHM-A sensor board, is used Rice, J. and Spencer, B. F. 2008. Structural Health Moni-
for acceleration measurements. An analytical model of toring sensor development for Imote2 Platform. Sensors
the truss structure is developed based on the available and Smart Structures Technologies for Civil, Mechanical,
specification of the truss and the properties derived and Aerospace Systems. San Diego, CA, USA, Vol. 6932,
693234.
from static tests. Extracting the modal properties of
Rice, J.A. and Spencer, B.F. 2009. Flexible Smart Sensor
the structure from dynamic tests, the analytical model Framework for Autonomous Full-scale Structural Health
is updated. Subspace state space algorithm is used to Monitoring,. NSEL Report Series, No. 18, University
identify the dynamic properties of the truss. Finally, a of Illinois at Urbana-Champaign. (http://hdl.handle.net/
local damage is simulated on the structure and a set 2142/13635).
of data from new measurements is passed through the Sohn, H., Farrar, C. R. 2001. Damage Diagnosis Using Time
statistical damage detection algorithms to indicate the Series Analysis of Vibration Signals, Smart Materials and
changes in the dynamic characteristics of the system. Structures, Vol.10, No 3, 446452.
U.S. department of transportation, 2008. Status of the
Nations Highways, Bridges and Transit: Conditions &
Performance. Report to Congress, United State.
REFERENCES
Wang, Y., Lynch, J. P. and Law, K. H. 2005. Validation of an
Farrar, C. R., Duffey, T. A., Doebling, S. W., Nix, D. A. 1999. integrated network system for real-time wireless monitor-
A Statistical Pattern Recognition Paradigm for Vibration- ing of civil structures. Proc. 5th Intl Ws. Structural Health
based Structural Health Monitoring, 2nd International Monitoring, Stanford, CA, September 1214.
Workshop on Structural Health Monitoring, Standford, Whelan, M. J., Janoyan, K. D., 2009. Design of a Robust,
CA. High-rate Wireless Sensor Network for Static and
Fugate, M. L., Sohn, H. and Farrar, C. R. 2001. Vibration- Dynamic Structural Monitoring. Journal of Intelligent
based Damage Detection Using Statistical Process Con- Material Systems and Structures, Vol. 20, No. 7, 849863.
trol, Mechanical Systems and Signal Processing. Vol. 15,
No. 4, 707721.
617
Modern permanent and portable restraint systems for bridges
T. Edl & A. Barnas

DELTA BLOC International GmbH, Vienna, Austria
ABSTRACT Research projects and modern developments of

industry brought up that it is possible to provide high
What does really happen when a 38 ton truck crashes containment levels for bridges at acceptable loads
into a bridge restraint system? Due to some hazardous on the bridge structure. Nowadays bridge designers
accidents on bridges in the past when heavy goods are able to choose from various kinds of premium
vehicles broke through bridge safety barriers causing restraint systems complying with national require-
structural damage on bridge structures, the necessity ments for bridge design and the European standards
for a more detailed understanding of the impact pro- for restraint systems.
cedure was obvious. Therefore some research projects Designing bridge structures by using the failure
under great efforts of accredited test houses, national loads of barrier anchors is not state of the art. For
road authorities, bridge designers and industry were having a load case that comes close to reality also the
carried out to show what really happens when a 38 impact loads mentioned in some national standards are
ton truck crashes into a bridge barrier system. Now not sufficient for calculating impacts on safety barri-
the industry offers highly sophisticated barrier sys- ers. Now it is possible to operate with forces measured
tems to provide very high containment levels causing in crash tests. Numerical simulations on this basis
less stress for the bridge structure for bridge design- brought up that premium restraint systems decrease
ers it is possible to optimize their static load cases and the impact load on bridges to such an amount that an
calculations. upgrading of the containment level on existing bridges
is also possible.
REFERENCES
EN 1317-1, Road restraint systems Part 1: Terminology and
general criteria for test methods, April 1998
EN 1317-2, Road restraint systems Part 2: Performance
classes, impact test acceptance criteria and test methods
for safety barriers, April 1998.
Hellmich K., Stella J., Stangl E., Piringer S., Heimel H.,
Plomer J. 2002, Restraint systems on bridges acceptance
tests with force measurements, Final report of the research
contract of theAustrian Ministry of Traffic, Innovation and
Technology, Road research contract Nr. 3554.
RVS 15.47, 2005: Brckenausrstung, Rckhaltesysteme auf
Brcken
ENV 1991-2, 2004, Eurocode 1 Actions on structures, Part
2: General actions Traffic loads on bridges.
Figure 1. H2/TB51 test with 13 ton bus and H4b/TB81 test

with 38 ton truck.
618
Effect of soil bridge interaction on the distribution of live

load effects among integral bridge components

ABSTRACT The analyses results revealed that soil-bridge inter-

action has a significant effect on the magnitudes of
An integral abutment bridge (IAB) is one in which the live load moments in the components of IABs.
the continuous superstructure, the abutments and the Including the effect of backfill behind the abutments
single row of steel H piles supporting the abutments in the structural model is generally found to result in
are built monolithically to form a rigid frame structure. larger superstructure support and abutment moments
In bridge design, most bridge engineers prefer using and smaller superstructure span and pile moments.
simplified two dimensional (2-D) structural models The difference between the live load moments for
and live load distribution factors available in current the cases with and without soil-bridge interaction
design codes to determine live load effects in bridge effects is found to be a function of foundation soil
components. Although the monolithic construction of stiffness. However, soil-bridge interaction is found to
IABs forces the substructures to interact with the back- have only a negligible effect on live load shear in
fill and foundation soil under thermal and gravitational the superstructure. Furthermore, it is found that the
load effects (Dicleli & Albhaisi 2003), the current state analyses performed using the equivalent pile length
of design practice in North America and Europe nor- concept inconsistently yield either conservative or
mally neglects soil-bridge interaction effects in live unconservative estimates of the internal forces in the
load analyses of IABs. That is, the backfill behind the components of IABs except for the superstructure
abutments is not considered in the 2-D structural mod- shear where the results of the equivalent pile length
els of IABs for live load analysis and the piles are model coincide with those of the models including
usually modeled as simple equivalent cantilevers fixed soil-bridge interaction effects. Based on the findings of
at some distance below the ground surface. Although this research study, it may be recommended to include
many research studies have been conducted on the the abutment-backfill and soil-pile interaction behav-
effect of backfill and foundation soil on the perfor- ior in the structural model of short to medium length
mance of IBs under thermal effects (Khoadir and IABs for the purpose of live load analyses. The linear
Hassiotis 2005), research studies on the performance soil-bridge interaction modeling techniques presented
of IBs under live loads are scarce. in this paper may be used for this purpose.
In this study, the effect of soil-bridge interaction on
the magnitude of internal forces in IAB components
(deck, abutment and piles) due to live loads is studied. REFERENCES
For this purpose, structural models of typical IABs are
built by including and excluding the effect of backfill Dicleli M. & Albhaisi S. M. 2003. Maximum length of inte-
gral abutment bridges supported on steel h-piles driven in
and foundation soil. The analyses of the models are
sand, Engineering Structures, 25(12): 14911504.
then conducted under AASHTO live load. In the anal- Khodair, Y. A. & Hassiotis, S. 2005. Analysis of soil
yses, the effects of the backfill and foundation soil on pile interaction in integral abutments, Computers and
the magnitude of the internal forces in IAB compo- Geotechnics, 32(3): 201209.
nents are studied for various structural, geometric and
geotechnical parameters such as bridge size, abutment
height and thickness, pile size and orientation, number
of spans and foundation soil stiffness.
619
Live load distribution in integral bridge girders

ABSTRACT analyzed. In the analyses, the effects of various super-

structure properties such as span length, number of
Integral abutment bridges (IABs) possess many eco- design lanes, prestressed concrete girder size and spac-
nomical and functional advantages. Consequently, ing as well as slab thickness are considered. The results
they have become very popular in many parts of USA, from the analyses of two and three dimensional FEMs
Canada and Europe. However, standard design meth- are then used to calculate the live load distribution
ods for IABs have not been fully established yet. factors (LLDFs) for the girders of IABs as a function
Thus, many practicing engineers use the provisions of the above mentioned parameters. LLDFs for the
for regular jointed bridges in current bridge design girders are also calculated using the AASHTO formu-
specifications such as AASHTO (American Associa- lae. Comparison of the analyses results revealed that
tion State Highway Transportation Officials, 2007) to LLDFs for girder moments and exterior girder shear of
design IABs. This also includes using such provisions IABs are generally smaller than those calculated using
for the design of IAB girders under live load effects. AASHTO formulae especially for short spans. How-
Most bridge engineers use simplified two-dimensional ever, AASHTO LLDFs for interior girder shear are
structural models and live load distribution factors found to be in good agreement with those obtained for
(LLDFs) readily available in bridge design specifi- IABs. Consequently, direct live load distribution for-
cations such as AASHTO (2007) to determine live mulae and correction factors to the current AASHTO
load effects in bridge girders. The live load distribu- live load distribution equations are developed to esti-
tion equations (LLDEs) were basically developed for mate the girder live load moments and exterior girder
jointed bridges where the superstructure is separated live load shear for IABs with prestressed concrete
from the abutments via expansion joints. However, in girders. Comparison of the LLDFs obtained from the
the case of IABs, the monolithic construction of the analyses of FEMs and those calculated using the devel-
superstructure-abutment joint forces the superstruc- oped equations revealed that the developed formulae
ture and the abutments to act together under live load yield a reasonably good estimate of the live the load
effects. The continuity of the superstructure-abutment moment in short to medium span prestressed concrete
joint in IABs is found to improve the distribution of live IAB girders.
load moment among the girders especially for short
spans (Dicleli & Erhan, 2009). Accordingly, using the
LLDFs in AASHTO LRFD Bridge Design Specifica- REFERENCES
tions (2007) for the design of IAB girders may result in
incorrect estimates of live load effects. As a result, in AASHTO 2007. LRFD Bridge Design Specifications, 4th
Edition. Washington, D. C., USA.
this study, LLDEs for the commonly used prestressed
Dicleli, M. & Erhan, S. 2009. Live Load Distribution Formu-
concrete girders of single-span IABs are developed lae for Pre-stressed Concrete Integral Abutment Bridge
to offer a practical tool to design engineers for the Girders. ASCE Journal of Bridge Engineering 14(6):
calculation of live load effects in IAB girders. For 472486.
this purpose, two and three dimensional finite ele-
ment models (FEMs) of several IABs are built and
620
Comparison of fatigue behavior under concentrated loads of orthotropic

decks and crane runways
M. Euler & U. Kuhlmann

University of Stuttgart, Institute of Structural Design, Stuttgart, Germany
ABSTRACT
The introduction of repeatedly moving single loads

(wheel loads) in a plated steel structure is a frequent
design situation. As a rule the wheel loads are trans-
ferred via distributing means into the load carrying
structure such as an asphalt surfacing in case of the
steel deck of a road bridge or a rail in case of a crane
runway girder (Fig. 1).
In case of modern plated steel structures welds are
usually located in the load application region leading
to a significant increase of the actual stresses com-
pared with the computed nominal stresses according
to engineers theory.
The increase in stress has got several reasons:
(i) The shapes of the weld toes and roots form geomet- Figure 1. Load application into a steel deck (a) and a crane
ric notches amplifying the nominal stresses. (ii) The runway girder (b), schematic load application section (c).
welding leads to material homogeneities; moreover,
residual stresses are induced by the welding procedure.
(iii) The concentrated load causes local stress peaks
(Fig. 2). Altogether, the wheel loaded region under-
goes a complex multiaxial state of stress with several
stress rising features.
Considering the application of a stationary static
single load the stress-rising effects might be negligible
because the stress peaks can be eliminated by harmless
local plastifying. In contrast, the wheel loads recipro-
cate normally, and the cyclic occurrence of the stress
peaks might lead to an accelerated material failure that Figure 2. State of stress in the load application region.
means fatigue.
The state of stress in the wheel load application In this paper a summary of known fatigue phenom-
region of an orthotropic steel deck and a crane run- ena caused by wheel loads is presented. Finally, the
way shows similarities. The occurrence of local stress current design approaches of the fatigue assessment
components in addition to the global nominal stresses for orthotropic decks and crane runways are compared
is characteristic of both. Since the local stress compo- focusing on following issues:
nents interact with the flexural stresses due to global
bending a design situation with combined stresses has local stresses and their distribution
to be assessed. As the different stress components act influence of wearing
cyclically and out of phase the wheel load application fatigue loading
turns into a complex multiaxial fatigue problem with fatigue categories
non-proportional loading. theory of failure
621
The strain development in concrete under cyclic loading
M. Foglar
Czech Technical University in Prague, Prague, Czech Republic
ABSTRACT
Based on the experiments of J.O. Holmen, paramet-

ric description of the development of secant modu-
lus of elasticity of concrete under cyclic loading is
proposed.
Constants a and b are introduced, a for the decrease
of the secant modulus of elasticity in the first phase
of its development under cyclic loading, b for the
remaining proportion of the original secant modulus
of elasticity at the beginning of the third phase of its
development.
A mathematical function for describing the strain
development in a concrete specimen under cyclic load-
ing is sought for. This function should be able to give
a value of modulus of elasticity in every particular Figure 1. Example of the fatigue damage function for
moment of cyclic loading, thus respecting the three various load levels.
phases in strain development under cyclic loading. The
reduced value of modulus of elasticity can be then the rapid decrease of modulus of elasticity in the third
used for calculating deflections increased by damage phase of cyclic loading:
accumulation caused by cyclic loading.
The instant value of modulus of elasticity after ni
load cycles is:
where all the constants and variables are explained in

where Eni modulus of elasticity of concrete after ni load the full text of the paper.
cycles; Fi = fatigue damage function after ni load The fatigue damage function should be used for
cycles; and En0 = modulus of elasticity of concrete at assessing deflections of structures subjected to cyclic
the start of the cyclic loading. loading either in in-hand calculations or inserted
The fatigue damage function decreases the original into FEM software producing a complete useful life
modulus of elasticity of concrete at the start of cyclic analysis of a particular structure.
loading to quantify the deteriorative effect of ni load
cycles of cyclic loading. It comprises of a power and
an exponential part. The exponential part is dependent REFERENCES
on Smax (the maximum stress level during the cyclic
Holmen, J.O. 1979. Fatigue of concrete by constant and
loading) and represents rapid decrease of modulus of variable amplitude loading. NTH-Trondheim, Division of
elasticity in the first phase of cyclic loading, together Concrete Structures Bulletin, 791, Trondheim.
with the stable progressive decrease of modulus of Barr, B.I.G. and Lee, M.K. 2002. An overview of the fatigue
elasticity in the second phase of cyclic loading. The behavior of plain and fiber-reinforced concrete. Cement &
exponential part is independent of Smax and represents Concrete Composites, 24: 299305.
622
Cracking of reinforced concrete deck on skewed bridges
G. Fu
J. Feng
New Institute, Gouda, The Netherlands
J. Dimaria & Y. Zhuang

Wayne State University, Detroit, MI, USA
ABSTRACT using optimal combinations of ingredients in concrete

and to minimize the constraint between the deck and
A large number of highway bridges are skewed, due to the supporting superstructure.
the requirement of crossing other waterway or road-
way with various angles. For example, in Michigan
they represent 35% of all bridges in the state. It has REFERENCES
been noticed that many of recently constructed rein-
forced concrete decks on skewed bridges have shown AASHTO Standard Specifications for Highway Bridges
significant cracking, especially in the corner areas. 17th Ed. 2002, Washington, DC
They are perceived to be relevant to the skew geom- AASHTO LRFD Bridge Design Specifications, 3rd Ed,
etry. This research effort was to investigate the cause 2004, Washington, DC.
H. Cohen, G. Fu, W. Dekelbab, and F. Moses Predict-
of the observed cracking and to develop measures to
ing Truck Weight Spectra under Weight Limit Changes
mitigate or eliminate it. and Its Application to Steel Bridge Fatigue Assessment,
Accordingly, this study had a focus on corner crack- American Society of Civil Engineers Journal of Bridge
ing in reinforced concrete decks of skew highway Engineering, Sept. 2003a, Vol.8, No.5, p.312
bridges. A survey of state transportation agencies in Fu, G., J. Feng, J. Dimaria, Y. Zhuang Bridge Deck Corner
the U.S. was conducted on this subject. It found that Cracking on Skewed Structures, Final Report RC-1490
deck corner cracking in skew bridges is commonly to Michigan Department of Transportation, Department
observed in the entire country. Deck inspection for of Civil and Environmental Engineering, Wayne State
bridges in Michigan was also performed in this study. University, September 2007
Fu, G., J. Feng, W. Dekelbab, F. Moses, H. Cohen, D. Mertz,
Cracking intensity in these decks was viewed as an
and P. Thompson Effect of Truck Weight on Bridge Net-
effect of several possible causal factors, which was work Costs, Transportation Research Board, NCHRP
collected from 40 bridge decks, including 20 straight Report 495, 2003b
and 20 skewed structures. Analysis of the inspection G. Fu, J. Feng, W. Dekelbab, H. Cohen, F. Moses, and D. Mertz
results indicates no clearly agreeable causal relations. Demands of Commercial Vehicle Weights on Highway
Two skew decks were instrumented using temperature Bridge Infrastructure, (submitted to) ASCE Journal of
and strain sensors for the concrete and the ambient Bridge Engineering, 2007
environment. Concrete decks temperature and strain Fu, G., S. Alampalli, and F.P. Pezze Lightly Reinforced
response was collected to thermal-, shrinkage-, and Concrete Bridge Deck Slabs on Steel Stringers: A Sum-
mary of Field Experience, Final Report on R142 to
truck-wheel-loads. Test results and thereby calibrated
FHWA, FHWA/NY/RR-94/161, Engineering Research
finite element analysis results show that the main cause and Development Bureau, New York State Department
of skew deck corner cracking is cement concretes of Transportation, June 1994
thermal and shrinkage load. Based on current Michi- Krauss, P.D. and Rogalla, E.A. Transverse Cracking in
gan practice of skew deck design and construction, Newly Constructed Bridge Decks NCHRP Report 380,
additional reinforcement in the corner areas is there- Washington, DC., 1996
fore recommended to reduce concrete stresses. Further TNO DIANA Users Manual,Application Modules Release
research is also recommended to develop solutions 8.1, 2003
623
The Filetto Bridge on the Santerno River near Bologna (Italy): Seismic
retrofit and reinforcement design
G. Gasparini, S. Silvestri, T. Trombetti & C. Ceccoli

University of Bologna, Department DISTART, Bologna, Italy
ABSTRACT
The Filetto Bridge was built about in 1930, to over-

pass the Santerno river and to connect the village of
Fontanelice to the village of Casalfiumanese, two little
towns near Bologna, a city located in the northern part
of Italy. The bridge has a total length of 110 m, subdi-
vided into 4 spans, each one 27 m long. The restraints
are composed by two abutments at the two ends of
the 4-span bridge and by 3 intermediate piers laying Figure 1. Historical photographs: (a) the gravel roadway
in the riverbed; the deck is 4,5 meters wide (only one leading to the bridge path; (b) overall image.
roadway 3,0 m wide). The section of the piers is com-
posed by 4 square columns 70 cm70 cm each, located
at the angles of a total section of 240 cm240 cm in
plan: the columns are connected each other by several
arched beams. The abutments are a kind of concrete
caissons, laying on the slopes at the two sides.
The bridge was constructed with the post-stressed
concrete technique (Figure 1).
The aim of the research work here presented is
to investigate the seismic behaviour of this struc-
ture according to the rules of the new Italian Seismic
Design Code, in order to focus the basic problems and Figure 2. (a), (b) elastomeric bearings.
to study some possible solutions that can improve the
seismic behaviour of the bridge.
Preliminary analysis shows that the existing struc-
ture is not able to support the new seismic excitation
mainly because the benefit of the post-stress is partially
lost due to the traction stresses in the deck, and the
present supporting devices cannot support the drifts
between the pile and the deck.
The results obtained from several static and
dynamic analysis show the necessity of a reinforce-
ment design; the fundamental works which provide
the structure with the best performances are: Figure 3. (a), highly resistant carbon fiber strips; (b) painted
fiber.
complete demolition of the superior r.c. slab of
the whole deck and realization of a new slab with reinforcement of the r.c. beams of the decks with
lightweight aggregated concrete; the insertion of highly resistant carbon fiber strips,
substitution of the actual restraints with elas- in order to restore the loss of post-tension effect
tomeric/rubber bearings (isolators), see Figure 2; (Figure 3).
realization of structural joints on the abutments to
avoid the effects of thermal expansion on the slab;
624
Masonry arch bridges towards a hierarchical assessment framework
N. Gibbons & P.J. Fanning

School of Architecture, Landscape & Civil Engineering, University College Dublin, Dublin, Ireland
ABSTRACT in part, the resulting scatter of assessment ratings

achieved for a single bridge. What is often more dif-
Masronry arch bridges are estimated to account for ficult to rationalize is the anomaly of an apparently
more than 40% of European bridges. Often more than higher level, or more rigorous, assessment algorithm
100 years old these bridges have performed well in resulting in a lower assessment rating.
service and are arguably the most durable and sustain- UCD is working with the National Roads Author-
able bridge type. However, the gradual deterioration ity in Ireland to rationalize a hierarchical framework
of materials with time, coupled with the increase in of assessment algorithms whereby increasing assess-
loading from modern road and rail vehicles, make ment effort is rewarded by demonstrable convergence
re-assessment, maintenance, repair and strengthening toward the ultimate capacity of the bridge. The aim
inevitable in order to ensure that safety, performance is to develop an algorithm hierarchy for masonry
and serviceability are sustained at an acceptable level. arch bridges. As part of this work 12 masonry arch
While numbers of masonry arch bridges in the US are bridges, characteristic of the range of bridges on the
lower these issues are equally relevant. Irish road network, have been assessed using three
Bridge managers and owners use a variety of assessment methods. This paper presents a discussion
assessment algorithms to identify safe load capacities. of the assessment approaches, the assessment ratings
The underlying theory behind different assessment achieved and the trends in assessment ratings of the dif-
approaches varies. The required bridge data also varies ferent assessment approaches vis--vis each other and
from method to method as do the dimensional models also measured responses, to passing weighed vehicles,
used. This diversity of assessment methods explains, for two of the bridges
625
Target proof load factors for highway bridge assessment in Central

and Eastern European Countries (CEEC)
J.D. Gmez & J.R. Casas

Universitat Politcnica de Catalunya, Barcelona, Spain
ABSTRACT Table 3. Proposed proof load factors for non-documented

bridges and Dutch traffic conditions.
The paper proposes a simple method to obtain the tar-
get proof load that guarantees a required safety level
Span length
on the bridge against the actual traffic. The proof load
(m) 2.3 3.6 5.0
factors are obtained by means of a reliability-based
calibration with 3 different safety levels (2.3, 3.6, 5.0). 10 1.18 1.61 2.25
Two traffic scenarios have been considered, one repre- 15 1.23 1.67 2.33
sentative of Western Europe (The Netherlands) and 20 1.33 1.80 2.51
other representative of New Member States (NMS) 25 1.41 1.91 2.64
and Central and Eastern Europe (Czech Republic, 30 1.45 1.95 2.70
Poland, Slovakia and Slovenia). The final result is 35 1.49 1.99 2.75
a simple procedure based on charts and tables that
a bridge evaluator without specific knowledge of
reliability theory can easily apply. Table 4. Proposed proof load factors for non-documented
As an example, in table 1 is presented a summary of bridges in NMS and CEEC countries.
the results obtained for a safety level ( = 2.3), as func-

tion of actual resistance and span-length, for Holland, Span length
for the case that bridge information is available. (m) 2.3 3.6 5.0
For the rest of the countries, similar values are
obtained as a function of safety level and span-length 10 0.83 1.13 1.57
(ARCHES-D16 2009). Therefore, the values in table 2 15 0.89 1.20 1.65
are proposed for proof load factors to be used in NMS 20 1.01 1.36 1.85
and CEEC countries in the absence of more accurate 25 1.08 1.44 1.97
30 1.11 1.46 2.00
Table 1. Proof load factors for Dutch traffic and = 2.3. 35 1.12 1.48 2.01
Span-length (m)
values obtained for the specific bridge for a Reliability
R/Rn 10 15 20 25 30 35 index = 2.3.
The values presented in tables 3 and 4 were obtained
1.0 0,18 0,38 0,49
0,9 0,61 0,70 0,84 1,07 1,16 1,22
for the considered countries. In the cases that docu-
0,8 0,86 0,98 1,14 1,26 1,31 1,35 mentation is not available,
0,7 1,04 1,12 1,26 1,33 1,37 1,41
0,6 1,11 1,18 1,28 1,37 1,41 1,44
0,5 1,14 1,20 1,30 1,39 1,44 1,46 REFERENCES
ARCHES-D16 2009. Recommendations on the use of soft,
Table 2. Proof load factors for NMS and = 2.3. diagnostic and proof load testing. Deliverable D16.
ARCHES Project. VI EU Framework Program. Brussels
Span-length (m) (available on line at http://arches.fehrl.org)
Gmez, J.D. & Casas, J.R. 2008. Assessment of bridge
R/Rn 10 15 20 25 30 35 capacity through proof load testing. Proceedings of 4th
International Conference on Bridge Maintenance, Safety
1.0 0.31 and Management, IABMAS08. Seoul, Korea. Rotterdam:
0.9 0.15 0.28 0.45 0.55 0.59 0.61 Balkema
0.8 0.51 0.58 0.69 0.78 0.82 0.84 Casas, J.R &. Gmez, J.D. 2010. Bridge capacity assessment
0.7 0.63 0.69 0.82 0.94 0.96 0.98 by combined proof-loading and WIM data. Proceedings
0.6 0.72 0.78 0.92 1.00 1.04 1.05 of 5th International Conference on Bridge Maintenance,
0.5 0.78 0.84 0.96 1.04 1.07 1.09 Safety and Management, IABMAS 10. Philadelphia ,
USA. Rotterdam: Balkema
626
Synthetic fiber ropes to replace steel wire in pedestrian suspension bridges
S. Gupte & R. Betti

Columbia University, New York, United States of America
T. Zoli
HNTB Corporation, New York, United States of America
ABSTRACT
Rope making has seen a transition from the traditional

hemp ropes used during the Inca Empire to the steel
wire ropes in the 1830s and now to synthetic fiber
ropes. Synthetic fiber has displayed immense poten-
tial in terms of its tensile strength, which has made
its application in various industries possible. Research
and development has led to the manufacture of high
modulus and high tenacity products, which have better Figure 1. Last surviving Figure 2. Synthetic Rope
strength-to-weight ratios than steel ropes. Steel wire hemp rope Inca Bridge. used for mooring.
ropes which are used for pedestrian suspension bridges
prove to be highly redundant, considering the loads
they face during their lifetime. They are also more
prone to weathering. The cost of repair or replace-
ment of corroded wire is higher as the frequency of
replacement goes up based on various climatic condi-
tions. Synthetic ropes provide a feasible option but its
application to this field is yet to be explored. Based on
the strength requirements for a pedestrian rope bridge,
appropriate fiber has to be selected for the bridge.
Companies which cater to the rope industry conduct
numerous tests to constantly upgrade their products
and to overcome any drawbacks the material might
have. This study makes an attempt to substitute steel
wire in rope bridges to synthetic fiber ropes made from
high modulus plastic. By studying the test results avail- Figure 3. Trail Bridge in Nepal.
able from the companies for every product, a decision
about the material with apt properties is made. This standards will be used as guideline for determining
will not only make the bridge light but also make the the applicability of the synthetic ropes. The capacity
construction process less time consuming and tedious. of the ropes is well illustrated by the fact that they are
A suitable structural form has been developed which used in heavier industries like the off shore mooring;
takes advantage of the suspension and stress ribbon where adverse environments too are bound to affect
structural types for pedestrian bridges. These bridges the life of the rope.
will be built on a community basis along remote trails
in secluded mountainous regions across the world.
Local manpower and material will be used as far as REFERENCES
possible in order to provide economic advantage to
the people. Foster, Gale. Advantages of Fiber Rope over Wire Rope.
Similar programs conducted by developmental Cordage Institute, MA, USA.
organizations have been studied. The Swiss Develop- Chou, Rafael; Daughters, Michael & Nulle, Randy. Bene-
ment Corporation (SDC), Helvetas and Government fits of Synthetic Rope as Wire Replacement. COFE 2005,
Duncan BC, Canada, April 2005.
of Nepal are jointly involved in building rope bridges McCorkle, Eric. HMPE Ropes in the Trawl Industry. Fish
under the Trail Bridge Program in Nepal. These short Expo, Seattle, WA, November 2003.
span bridges can be easily built from synthetic ropes. Tuladhar, Artha. Trail bridge building in the Himalayas:
The SDC also developed standards for building such enhanced access, improved livelihoods. Zrich, Switzer-
bridges in the mountainous terrain of Nepal. These land: Practical Action Pub. 2007.
627
Seismic evaluation of Sogutlucesme Viaduct
A.F. Gurdil, A. Yagcioglu & A.G. Ozcebe

Temelsu International Engineering Services Inc., Ankara, Turkey
ABSTRACT
Marmaray is one of the most challenging infrastruc-

ture projects in Turkey. The purpose of the project is
to provide safe, comfortable and cost-effective train
transportation for passengers and cargo between the
Asian and European sides of Istanbul in Turkey. The
project consists of three contracts namely; Bosporus
Crossing Contract comprising an immersed tunnel, the Photograph 1. Aerial view of the Viaduct.
Commuter Rail Contract which includes constructing
a third interstate railroad and upgrading the existing Necessities of the rehabilitation works are also briefly
two lanes commuter rail system for 100 years of ser- summarized.
vice life. The third contract is related with the Rolling
Stock.
The major structure within the scope of second con- REFERENCES
tract is Sogutlucesme Station Viaduct. This viaduct
has a particular importance for being the first above- AASHTO-SSHB. 2002, Standard Specifications for Highway
Bridges. The American Association of State Highway and
ground Railroad Passenger Station constructed on a
Transportation Officials, USA.
viaduct in Turkey, and also has special worth in both ACI 318-02. 2002, Building Code Requirements of Structural
aesthetic aspects and socio-cultural life of the precinct. Concrete and Commentary. American Concrete Institute,
It comprises 6 parallel bridges with 19 axes on hori- USA.
zontal curve as distinct structural systems; 4 railway Aydinoglu, M.N. 2005. Evaluation of seismic performances
platform having lengths of about 500 meters and 2 pas- of existing or retrofitted bridges by non-linear methods.
senger platforms with lengths of 400 meters. Platforms Bowles, J.E. 1997, Foundation Design and Analysis.
consist of prestressed precast reinforced concrete box McGraw-Hill.
girders. Piers are constructed as single columns of CALTRANS 1999, Bridge Design Specifications Manual
and Design Memos. State of California, Department of
two octagonal sections varying in shape along the
Transportations, USA.
height. Footings are formed as spread or piled footings DBYYHY 2007, Specifications for the Structures to be Build
depending on the subsoil conditions. in Seismic Areas, Turkish Republic Ministry of Public
The presented paper summarizes detailed investi- Works and Settlement.
gation and design works including condition survey, Erdogan, T.Y. 2003, Concrete. Ankara: METU Foundation
structural capacity and seismic evaluation studies for Press.
the viaduct for its conformity with the state of the art ERQ 2006, Marmaray CR1 Contract Employers Require-
design codes and new seismic design criteria revised ments. General Directorate of Construction of Railways,
after 1999 major earthquakes. The seismic capaci- Harbors and Airports, Turkey.
FEMA356. 2000, Prestandard and Commentary for the Seis-
ties of all structural components have been evaluated
mic Rehabilitation of the Buildings. Federal Emergency
by determining their Capacity/Demand ratios for two Management Agency, USA.
different seismic excitation scenarios having different Sucuoglu, H. 2007. Comparing of performance based design
hazard levels for return periods of 70 and 475 years, methods in Turkish Earthquake Code. Turkish Engineer-
and comparing those with the acceptance criteria. ing News: 444445.
628
Sustainable design for steel-concrete bridges
G. Hauf, P. Maier & U. Kuhlmann

Institute of Structural Design, University of Stuttgart, Germany
ABSTRACT
As bridges are of vital importance in the worldwide

infrastructure network and significant in economy, the
request for sustainable structures is urgent. Sustain-
ability of bridges means lifecycle design, but lifecycle
design over a span of more than 100 years. So in
contrast to sustainability issues for buildings the dura-
bility and the flexibility to adapt to changed conditions
and increased traffic volume play a crucial role for
the assessment of the overall lifecycle performance
(LCP) of bridges. This contribution tries to draw a
general concept assessing environmental, economic
and social or functional quality as the 3 faces of a Figure 1. Composite bridge Dambachtal.
sustainability qualification, however in view of the
special needs of bridges. First example calculations
for medium span bridges in the frame of an expertise
ordered by the Hessian Road and Traffic Administra- demolition of the bridge have to be considered as well.
tion to assess the cost-effectiveness of steel-concrete As these costs develop over the lifetime of the bridge,
composite bridges, particularly with regard to the a fully integrated lifecycle model has to be derived. In
construction method, see Kuhlmann et al. (2006), the frame of a European RFCS research project SBRI
have demonstrated the advantages of steel composite (Kuhlmann, U. et al. 2009) with 9 partners among oth-
bridges (Fig. 1). ers a tool for evaluating new steel/composite bridges
Three different evaluation methods for cost- from sustainable perspective will be developed. This
efficiency are presented and the conclusion is drawn project also aims at minimizing environmental and
that for bridges it is not sufficient to compare only cost impact of modern steel / composite bridges
construction costs without considering construction using a holistic approach which combines Lifecy-
time, neither construction method nor the type of cle Assessment (LCA), Lifecycle Cost (LCC) and
construction. improvement of social and functional quality by a
As the construction of a bridge causes traffic hin- Lifecycle Performance (LCP) analysis.
drances it is intended to minimize these in order to Looking into the future a certain responsibility
keep socio-economical costs as low as possible. The regarding sustainability has to taken into account. The
effect of these costs can be taken into account follow- criterion of sustainability has become a key issue.
ing EWS, one of the evaluation methods. Similarly the There is a need for sustainable long-living structures,
aspect of reduced construction time can be considered also taking into consideration future demands such as
by the approach of compensation costs according to increasing traffic volume.
ARS wherein additional costs are accepted for a shorter The calculations and comparisons conducted in the
construction time. expertise want to boost the willingness to consider var-
Regarding bridge costs over an entire lifetime con- ious aspects of a structure within its lifespan and to take
struction costs represent only a small part. Costs for new paths assessing and evaluating bridge structures
maintenance, restoration, modernization as well as concerning sustainability and lifecycle performance.
629
Estimating natural frequencies, damping ratios and FEM models

of suspension bridges from wind response measurement
A.L. Hong & R. Betti

Department of Civil Engineering and Engineering Mechanics, Columbia University, New York, NY, USA
F. Ubertini
Department of Civil and Environmental Engineering, University of Perugia, Perugia, Italy
ABSTRACT
In this paper, a novel approach to identify dynamic

characteristics of structures under stochastic exci-
tation is proposed. Following a general framework
of the Stochastic Subspace Identification (SSI) tech-
nique (Van Overschee & De Moor 1996), an enhanced
algorithm, which is named Enhanced Canonical Cor-
relation Analysis (ECCA), to properly determine the
order of a system in state variable estimate has been
developed.
The SSI technique is closely related to multivariate
statistical analysis. The implementation process of the
SSI can be represented in two steps: the first step is
to estimate the state vectors and the second step is to
extract the dynamic properties from the estimated state
vectors. The first step includes the determination of the
systems order and can be carried out through various
algorithms. Currently, three different algorithms are
considered for state vector estimates, which are asso-
ciated with classical multivariate techniques: Partial
Least Square (PLS), Canonical Correlation Analy-
sis (CCA), and Multiple Linear Regression (MLR) Figure 1. Field data: the normalized singular value distri-
techniques (Arun & Kung 1986). butions considered for the PLS, CCA, MLR, and ECCA
The efficiency of the proposed approach has been techniques.
proven by comparing it with the existing approaches
in applications to numerical data as well as field mea-
surements of a suspension bridge, New Carquinez weakly excited modes have been identified without
Bridge (NCB) in California. In such comparison, the using a so-called stabilization diagram. Therefore,
focus is on the investigation of capabilities of the four using the ECCA technique in stochastic realization
techniques (the PLS, CCA, MLR, and ECCA) to dis- could significantly reduce the need for a procedure for
criminate structural modes from noise modes in terms the selection of structural modes.
of singular values of their corresponding matrices.
Figure 1 presents the normalized Singular Value Dis-
tributions (SVD) considered in the four techniques. In REFERENCES
stochastic realization, the dimension of the state vec-
tors, the order of a system, has been determined only Arun, K.S. & Kung, S.Y. 1986. Generalized principal compo-
nents analysis and its application in approximate stochas-
based on such SVD. tic realization. In Uday B. Desai (ed.), Modeling andAppli-
As a result, it has been demonstrated that the ECCA cation of Stochastic Processes. Boston: Kluwer Academic
is quite effective in discriminating structural modes Publishers.
from noise ones considering the phase of the state vec- Van Overschee, P. & De Moor, B. 1996. Subspace Iden-
tor estimates. In fact, in the implementation of the tification for Linear Systems-Theory, Implementation,
ECCA, the structural modes of the NCB including Applications. Boston: Kluwer Academic Publishers.
630
Using orthogonal pairs of rollers on concave beds (OPRCB) in seismic

design and retrofit of highway bridges
M. Hosseini
Structural Engineering Research Center, International Institute of Earthquake Engineering and Seismology (IIEES),
Tehran, Iran
A. Soroor
Earthquake Engineering Department, School of Engineering, Science and Research Branch of the Islamic Azad
University (IAU), Tehran, Iran
S. Khavari
Civil Engineering Department, Sama Organization (Affiliated to the Islamic Azad University Roudehen Branch),
Roudehen, Iran
ABSTRACT
In this paper the use of Orthogonal Pairs of Rollers

on Concave Beds (OPRCB), as a simply manufac-
tured and installed and low cost isolating system has
been studied for seismic isolation of bridges. Rollers
installed in two orthogonal directions make possible
the movement of the superstructure in all horizontal
directions, and concave beds give the system restoring
and re-centering capabilities (Figure 1).
The high vulnerability of the bridge structure,
shown in Figure 2, has been discussed in a previous
study (Hosseini et al. 2008).
For investigating the efficiency of OPRCB isolators
the sample bridge have been considered once with its
Figure 3. Acceleration time history of the isolated bridge
deck and the hysteretic response of the bridge isolators to
Imperial Valley earthquake scaled to 0.7 g.
conventional support systems, and once installed on
OPRCB isolators, and its seismic responses have been
calculated using several accelerograms. The responses
related to Imperial Valley earthquake, as a sample, are
shown in Figure 3.
Based on the numerical results it can be con-
cluded that using OPRCB isolators in seismic design
and retrofit of highway bridges is very effective in
Figure 1. The main parts of the OPRCB isolators (Hosseini reduction of seismic demands.
and Soroor 2009).
REFERENCES
Hosseini, M., Banaeizadeh,Y., and Khavari, S. R. 2008 (July).
Vulnerability Assessment of an Existing Highway Bridge
by Push-Over and 3-D Nonlinear Time History Analyses
for its Seismic Retrofit Design, The 6th US National Seis-
mic Conference on Bridges and Highways, Charleston,
SC, USA.
Hosseini, M. and Soroor, A. 2009, A Study on the Use of
Orthogonal Pairs of Rods on Concave Beds (OPRCB) as
a Base Isolation Device for Building Systems, Proceed-
Figure 2. Computer model of the bridge for seismic evalu- ings of the 8th Intl Congress on Civil Engineering, Shiraz
ation. University, Shiraz, Iran, 1417 May, Paper S5000.
631
Two cable-stayed bridges designed for easy access and low maintenance
R. Hsu
Parsons Brinckerhoff, Inc.
1 INRODUCTION opened to traffic. (Figure 2) The 500 m or 1,640-foot

long main span is the longest span in Thailand. It is
Learning from access difficulties to aging bridges for also longer than any cable-stayed bridge spans in the
inspection and repair, bridge owners are paying more United States.
attention to access and details in design of new bridges.
Cable-stayed bridge is a new bridge type, which
became popular during 1980s. The American cable- 3 SPECIAL DESIGN FEATURES
stayed bridges built in this period of time are showing
aging and require major rehabilitations. When the William Natcher Bridge crossing the Ohio River
design of the William H. Natcher Bridge started in between Kentucky and Indiana and the Kan-
1991, the Parsons Brinckerhoff (PB) design team has chanaphisek Bridge crossing the Chao Phraya River
paid great attention to the access and maintenance that in Bangkok, Thailand are designed to minimizing
resulted in an easy access and low maintenance bridge. the future maintenance cost. This new concept was
Subsequent to the Natcher Bridge, PB improved the developed in 1991. Low maintenance achieved by
Kanchanaphisek Bridge design and made it easier for eliminating components, which require inspection,
access and lower in maintenance cost than the Natcher repair and replacement, or by designing components
Bridge. which could be removed easily for replacement. Low
Designing a long lasting and owner-friendly bridge maintenance cost is achieved also by providing easy
was the goal set by the PB design team at the beginning and ready access to all major components, such as
of both projects. This means providing easy access for cable anchors.
inspection and maintenance and minimizing mainte- Both Natcher Bridge and Kanchanaphisek Bridge
nance items.These goal and means are achieved during have no tie-downs, which is common to cable-stayed
conceptual design stage of both bridges. bridges before year 2000. Kanchanaphisek Bridge has
no bearings as well. Natcher Bridge uses elastomeric
pads for bearings and bumpers because they require
2 BRIDGE DESCRIPTION low maintenance. All bearings and bumpers could be
accessed directly and replaced easily.
2.1 William H. Natcher Bridge Both Natcher Bridge and Kanchanaphisek Bridge
have a large chamber on top of their towers to house the
William H. Natcher Bridge has a 1200-foot main span.
cable anchors. Steel cable anchorage frames become
It is flanked by two 500-foot back spans. The bridge
working platform not only for future maintenance
carries two lanes of traffic in each direction between
work but also for original construction. The most spe-
Kentucky and Indiana. The bridge provides a 70-foot
cial design is an opening in the chamber bottom slab
under clearance and a 1000-foot wide navigation chan-
and a ring on the roof ceiling. This allows lifting heavy
nel for barge traffic. (Figure 1) The bridge has been
replacement cable anchors directly from a truck parked
opened to traffic since 2002. It has been the longest
on the deck up to their final positions. It also allows
span crossing the Ohio River.
delivery of heavy repair equipment.
These two bridges are low maintenance examples
2.2 Kanchanaphisek Bridge to future cable-stayed bridge design.
On November 15, 2007, the Kanchanaphisek Bridge
crossing the Chao Phraya River in south Bangkok
632
Development of a cooperative management framework for bridge

maintenance using IFC data model
M.G. Huang, H.-J. Kim, K.-H. Kim & S.-H. Lee

School of Civil and Environmental Engineering, Yonsei University, Seoul, South Korea
ABSTRACT
The bridge maintenance management benefits from

various software applications which assist managers in
solving a multitude of engineering problems. However,
these software applications lead to large volumes of
unstructured or loosely structured data with poor inter-
operability and reusability. For instance, Bridge Man-
agement System (BMS), which is a wide-accepted
system for both project-level and network-level main-
tenance management of existing bridges, may be not
interoperable with other tools, such as asset manage-
ment tools and health monitoring systems. In order to
ensure data consistency and interoperability between
software applications, bridge information modeling
technique is recently emerged along with the rapid
development of building information modeling (BIM)
and Industry Foundation Classes (IFC) data model
(Lee and Jeong, 2006).
An Industry Foundation Classes (IFC) based data
model, i.e. an extension to IFC, is proposed for
representing the maintenance information of bridge
structures. The objective of the research is to rec-
ommend a common data structure and a cooperative
computer environment which enhance the informa- Figure 1. Coceptual framework model for BIM and BMS
tion exchangeability and consistency between various integration.
computer applications in bridge maintenance manage-
ment.
A generic process model is firstly investigated the methodology similar to that used by Building-
for representing the activities existing in the bridge SMART. The integration framework model brings new
maintenance process, such as performance require- functions such as 3D visualization, real-time struc-
ment identification, condition assessment, mainte- tural analysis, automatic quantification and costing to
nance planning, and maintenance operation man- bridge maintenance management. Additionally, inter-
agement; the process model is used to identify the operability and data consistency between computer
information items for the development of data model applications are guaranteed.
and is schematically described. Then a data model
extension is developed for IFC, which is neutral and
open data model developed within the International REFERENCE
Alliance for Interoperability (IAI). Then, a concep- Lee, S.-H. & Jeong, Y.-S. 2006. A system integration frame-
tual framework model is developed to integrate bridge work through development of ISO 10303-based product
information model and BMS (refer to Fig. 1). The model for steel bridges. Automation in Construction 15:
IFC data model extension is developed following 212228.
633
Development of live load model using weigh-in-motion data
E.-S. Hwang
Kyung Hee Universisty, Yongin, Korea
The live load model used in the bridge design rep-

resents the effects of traveling vehicle on the bridge.
The current live load model in Korea Bridge Design
Code needs to be updated to consider the fast growing
truck traffic and weight.Also the reliability-based code
requires statistical data of the load model. This paper
deals with the development of live load model using
data from BWIM (Bridge Weigh-In-Motion) system
and portable WIM (Weigh-In-Motion) system.
Truck weight data are collected on various sites in
the area. They include highways, national roads and
provincial roads. Procedures for determining the max-
imum load effects are determined. From the collected
data, maximum weights are estimated for each truck
type and sites using the extreme analysis. Only upper
10% of data are used in the estimation and assumed
as having extreme type distribution. Various factors
such as truck types, total weights, headway distances,
and correlation of truck types and weights are con-
sidered for each lane. The maximum load effects are
evaluated for different span lengths of the bridge. Mul-
tiple presence of trucks in one lane (series of trucks)
and two or more lanes(side-by-side trucks) are consid-
ered. The probability of multiple presences of trucks is
determined from the video recording and other stud-
ies. The results are compared with live load models Figure 1. Proposed live load model.
from various design codes.
Two different design live load models are proposed.
The first model is composed of two or three axle
loads with uniformly distributed lane load. The second the probability of side-by-side truck probability. More
model is composed of five axle loads with uniformly reliable and long-term data for truck weights and prob-
distributed lane load. Both models are under review ability of multiple presences should be collected for
by designers and practitioners before the final model is better estimation of actual effects of truck live load-
proposed for the new LRFD Korea Bridge Deign Code. ing. Also live load model for very long span bridges
Multi-lane loading factors are also proposed based on needs additional investigations.
634
A review of metallic bridge failure statistics
B.M. Imam & M.K. Chryssanthopoulos

Faculty of Engineering and Physical Sciences, University of Surrey, Guildford, GU2 7XH, UK
ABSTRACT
Studying the failures of the past can be useful in mit-

igating the incidence and potential of future failures.
A first step towards the understanding and quantifica-
tion of the risk of failure of bridges can be provided
by acquiring knowledge on the failure mechanisms
of existing structures and the root of causes of col-
lapse. Studies aimed independently for different types
of bridges, in terms of their material and form, may
lead to the identification of predominant failure causes
and modes for each bridge type. Clearly, trends picked
up through statistical analysis can aid in identifying
and understanding the potential of the most signifi-
cant hazards affecting bridge structures and help in Figure 1. Failure causes leading to collapse of metallic
planning against their consequences. bridges.
In this paper, damage/failure/collapse cases of
metallic bridges worldwide from the early 19th cen- Table 1. Cause of bridge collapses by date of failure.
tury up to the present are synthesised and classified. A
database of 164 failures of metallic bridges extracted Collapse cause Pre-1900 19001940 19411990 1991
from the literature, the web and news reports was
compiled. Of the 164 reported cases, 87 (51%) were Limited 10 (53%) 5 (36%) 4 (13%) 0 (0%)
classified as collapse, 73 (47%) as no collapse and knowledge
4 (3%) as unknown. The evaluation of the database Design errors 4 (21%) 4 (29%) 7 (23%) 5 (25%)
is carried out in terms of identifying the factors con- Natural hazards 1 (5%) 3 (21%) 8 (26%) 6 (30%)
tributing to the failure or collapse, the modes of failure Accidents 1 (5%) 1 (7%) 6 (19%) 4 (20%)
or collapse, structural form of the bridges, construc- Human errors 2 (11%) 1 (7%) 4 (13%) 2 (10%)
tion date and age of the failed bridges and the number
of casualties. The paper concludes with a discussion
of the significance of consequence analysis in terms
of risk assessment giving some general guidelines on occurred due to natural hazards and accidents over the
how consequences of bridge failure can be quantified. years. Remarkably, design and human errors remain
The distribution of failure causes for the collapse constant in percentage terms during the investigated
database is shown in Figure 1. The most important 150-year period, given the emphasis on quality assur-
factors (almost equally) contributing to collapse are ance and other human error mitigating procedures
design errors, limited knowledge, and natural hazards introduced in the past thirty years.
(21%). Clearly, there is no single dominant cause of Scour and buckling failures were found to be the
collapse for metallic bridges, though the top three most common collapse modes for metallic bridges
amount to almost two thirds of the whole. Further whereas fatigue was found to be the most common
analysis of the database (Table 1) shows that limited failure mode in the case of non-collapsed (damaged)
knowledge played a major role in the recorded col- metallic bridges. The statistics have also shown that
lapses up to the mid-twentieth century but decreasing truss bridges are more susceptible to total collapse
considerably in recent years. On the other hand, Table failures and girder bridges to failures that do not
1 shows an overall increase in the bridge collapses that necessarily result in total collapse.
635
Shear transfer mechanism in slab-on-girder bridges
A. Jamali & M.A. Saadeghvaziri

Department of Civil Engineering, New Jersey Institute of Technology, Newark, New Jersey, USA
ABSTRACT
Headed shear connectors are commonly used to trans-

fer longitudinal shear forces across the concretesteel
interface in bridges. While the connectors are mainly
meant to create composite action, they restrain the free
shrinkage of concrete decks and as a result cause trans-
verse cracking.A good understanding of the composite
action and interaction between stud and its surround-
ing concrete is very beneficial for developing a remedy
for the problem of deck cracking, which is the most sig-
nificant part of bridge maintenance cost in the US. The Figure 1. Distribution of shear force at deck-girder inter-
aim of this work, therefore, is to develop more insights face.
into the shear transfer mechanism in welded shear
studs to quantify its effect on bridge deck cracking.
Preliminary results indicate that under gravity load
shear studs uniformly contribute to transfer of shear
stresses, although the end studs (farther from cen-
ter) have slightly more contribution. However, under
shrinkage load the end studs transfer majority of the
load (Fig. 1). Furthermore, it is determined that a wider
shear stud spacing can be used without significant
increase in deflections.
Furthermore an innovative concept to possibly elim-
inate shrinkage deck cracking is briefly discussed. It
may be possible to achieve this goal if a mechanism is
developed such that it delays the composite action to
the time that a significant portion of the free shrink-
age strain has occurred. This can be achieved by using
threaded studs (Fig. 2), which are not fastened for
the period concrete undergoes a significant part of it
shrinkage which can be limited to the time that the
bridge is still not open to traffic. In this period the studs
are freely slipping with the concrete and no tensile Figure 2. Distribution of shear force at deck-girder inter-
stresses are developed in the deck. Upon completing face.
the bridge construction, all the studs are tightened so
that full composite action can be achieved. Practical- Saadeghvaziri M.A & Hadidi R. 2002. Cause and Control
ity of the purposed method is a matter which requires of Transverse Cracking in Concrete Bridge Decks. Report
further investigation. No. FHWA-NJ-2002-19, Federal HighwayAdministration.
Washington, DC, 2002.
REFERENCE
Badie S.S. et al. 2002. Large Shear Studs for Compos-
ite Action in Steel Bridge Girders. Journal of Bridge
Engineering 7(3): 195203.
636
Estimation of optimum isolator parameters for effective mitigation

of seismic risk for bridges

ABSTRACT sensitivity analyses, the parameters that affect the opti-

mum values of Qd and kd are identified. Next, for each
Seismic isolation of bridges is a design methodology one of the identified parameters, nonlinear time his-
that is based on limiting the magnitude of the seismic tory (NLTH) analyses of typical SIBs are conducted
forces transferred to the substructures. The perfor- to determine the optimum values of Qd and kd for
mance of seismic isolated bridges (SIBs) is measured an assumed range of values of the parameter under
by the maximum isolator force and displacement (MIF consideration. Next, the available data is plotted as a
and MID). The MIF represents the magnitude of the function of some dimensionless parameters proposed
seismic force transferred to the substructures. Thus, by Makris& Black (2004a, 2004b). Nonlinear regres-
it has a remarkable effect on the design of the sub- sion analyses of the plotted data are then conducted
structures. The MID is generally used to determine to obtain closed form equations for the optimum val-
the isolator size as well as the width and type of the ues of Qd and kd , to minimize the MID and MIF.
expansion joints. In some cases, the widths of the sub- The obtained closed form equations are then verified
structures may be governed by the MID. Accordingly, using a suite of ground motion data different than that
for a given ground motion, smaller isolator force and used for the development of the same equations. It is
displacement will produce a more economical bridge observed that the optimal Qd is proportional to peak
design under seismic effects. ground acceleration of the ground motion and bridge
Seismic isolators used in bridge applications may be mass and inversely proportional to the peak ground
classified into two groups as rubber-based and sliding- acceleration to peak ground velocity (Ap /Vp ) ratio of
based (Naeim & Kelly 1999). The force-displacement the ground motion. It is also found that the optimal
hysteresis of these isolators is generally idealized kd is proportional to peak ground acceleration of the
as bilinear for design purposes. The characteristic ground motion and bridge mass, inversely proportional
strength, Qd and the post-elastic stiffness, kd , are the to Qd and is a polynomial function of the Ap /Vp ratio
main isolator parameters that affect the behavior of a of the ground motion.
SIB for a given ground motion with specific frequency
characteristics and intensity (Dicleli & Buddaram
REFERENCES
2006). Thus, the optimal selection of these isola-
tor parameters based on minimizing the MID and Dicleli, M.and Buddaram S. 2006. Effect of isolator and
MIF will result in an economical design of the SIB. ground motion characteristics on the performance of
Accordingly, the objective of this research study is to seismic-isolated bridges. Earthquake Eng. and Structural
formulate closed form equations as functions of the Dyn., 35(2), 233250.
isolator, bridge and ground motion properties to calcu- Makris, N. and C. J. Black, 2004a. Dimensional analysis of
late the optimum Qd and kd of the isolator to minimize rigid-plastic and elastoplastic structures under pulse type
the MID and MIF. Using the developed equations it excitations. Journal of Engineering Mechanics, ASCE,
130(9): 10061018.
will be possible to select the proper isolator properties Makris, N. and C. J. Black, 2004b. Dimensional analysis of
that will result in an economical SIB design. bilinear oscillators under pulse-type excitations, Journal
To achieve the above stated objective, first, sensi- of Engineering Mechanics, ASCE, 130(9):10191031.
tivity analyses are conducted to investigate the effects Naeim, F. and Kelly, J. M., 1999. Design of Seismic Isolated
of several bridge, isolator and ground motion parame- Structures; From Theory to Practice, Wiley , Chichester,
ters, on the optimum values of Qd and kd . From these U.K.
637
Development of a new cycle counting method for cyclic thermal strains

in integral bridge piles

ABSTRACT
The daily and seasonal temperature changes result in

imposition of cyclic horizontal displacements on the
continuous bridge deck of integral bridges and thus on
the abutments, backfill soil, steel H-piles, and cycle
control joints at the ends of the approach slabs. Due
to these seasonal temperature changes the abutments
are pushed against the approach fill and then pulled
away, causing lateral deflections at the tops of the
piles that support the bridge as observed from Fig. 1
(French et al. 2004). The magnitude of these cyclic
displacements is a function of the level of tempera-
ture variation, type of the superstructure material and
the length of the bridge. As the length of the integral
bridges gets longer, the temperature-induced cyclic
displacements and forces in steel H-piles components Figure 1. Pile displacement due to thermal gradient.
may become larger as well. This may result in the
reduction of their service life due to low-cycle fatigue
effects (Dicleli & Albhaisi 2003, Arsoy et al. 2004). amplitude representative of a number of small ampli-
In this study, the field measurements obtained for tude cycles (primary and secondary) existing in a typ-
integral bridges are used to determine the amplitude ical temperature induced displacement/strain history
and the number of temperature induced cycles on in steel H-piles of integral bridges.
steel H-piles in integral bridges. Using the obtained It is found that, the secondary strain cycles have a
measurements, the number of large strain cycles per negligible effect on low cycle fatigue life of steel H
year due to seasonal temperature changes and the piles in integral bridges.
number and relative amplitude (relative to the ampli-
tude of large displacement/strain cycles, i.e. = small
strain cycle amplitude / large strain cycles ampli- REFERENCES
tude) of small strain cycles per year due to daily or
weekly temperature changes are determined. Addi- Arsoy S., Duncan J.M., Barker R.M. 2004. Behavior of a se-
tionally, the number of small cycles (secondary cycles) miintegral bridge abutment under static and temperature-
induced cyclic loading. Journal of Bridge Engineering,
between the maximum and minimum cycle above
Vol. 9, No. 2.
and/or under the large strain is counted. Using the Dicleli M., Albhaisi S.M. 2003. Maximum lengths of inte-
available data on the number and amplitude of temper- gral abutment bridges based on the strength of abutments
ature induced displacement-strain cycles, a new cycle and the performance of steel H-piles under cyclic thermal
counting method is developed to determine the number load-ing. BU-CEC-03-01, Department of Civil Engineer-
and amplitude of large and small displacement/strain ing and Construction, Bradley University, Peoria, IL.
cycles (small strain cycles are composed of primary French C., Huang J., Shield C. 2004. Behavior of concrete
and secondary strain cycles). Then, a new equation integral abutment bridges. Final Report.
is obtained to determine a displacement/strain cycle
638
Properties of self-consolidating light-weight concrete in massive structures
M. Kaszynska
West Pomeranian University of Technology, Szczecin, Poland
ABSTRACT
Structural light-weight high performance concrete

has been used in bridges for several years as
the self-consolidating concrete. The latest trend is
the utilization of SelfConsolidating Lightweight
High-Performance Concrete in bridge engineering.
This concrete combines the favorable properties of
lightweight high-performance concrete (LWC) with
those of self-consolidating concrete (SCC) and offers
an excellent solution for such applications.
This paper presents test results of hydration heat, Figure 1. Compressive strength and density of concrete
workability properties and mechanical properties after 28 days of hardening.
(compressive strength, tensile strength and modulus of
elasticity) of light-weight self-consolidating concrete,
hardening under adiabatic conditions (to simulate the
true conditions of concrete hardening inside of a mas-
sive structure) and isothermal conditions. The influ-
ence of curing conditions on mechanical properties of
LWSCC were considered.
The tests were performed for concrete mixtures
made using Portland cement CEM I 42,5R (Euro-
pean Standard EN 197-1:2000), fly ash, silica fume,
Sika Viscocrete 3 superplasticizer and two type of
lightweight aggregate: Pollytag size of 48 mm and
04 mm, Liapor size of 02 mm and 48 mm, natu- Figure 2. Relationship between the compressive strength
ral sand 02 mm and natural coarse aggregate size of concretes hardened under isothermal and adiabatic condi-
28 mm. A constant amount of paste was assumed tions.
with a variable proportion of light-weight aggregate to
natural aggregate. The natural aggregate was replaced (laboratory) and concretes hardening under conditions
by the same volume of the light-weight aggregate. simulating the curing in massive structures, for con-
Compressive strength was tested on cube specimens crete with Pollytag aggregates and Liapor aggregates is
150 150 150 mm. Splitting tensile test was carried presented in Figure 2.The relationship is approximated
out on cube specimens 100 100 100 mm. Mod- using a logarithmic function.
ulus of elasticity was tested on standard cylinders The developed concrete mixes have good self-
150 300 mm. consolidating properties, density lower than 2000 kg/m3 .
The relationship between the compressive strength The required properties of concrete strongly depend on
and density of concrete after 28 days of hardening, the fabrication procedure, including initial preparation
obtained from tests is shown in Figure 1. of aggregates, dosage method and mixing of com-
The relationship between the compressive strength ponents. The light-weight self-consolidating concrete
of concretes hardened under isothermal conditions requires a careful selection of the aggregates.
639
Seismic vulnerability assessment and comparing various retrofit designs

for an existing highway bridge
S.R. Khavari
Civil Engineering Department, Sama Organization (Affiliated to the Islamic Azad University Roudehen Branch),
Roudehen, Iran
M. Hosseini
Structural Engineering Research Center, International Institute of Earthquake Engineering and Seismology (IIEES),
Tehran, Iran
ABSTRACT
In this paper, having the high vulnerability of the

bridge proven in a previous study of the authors
(Hosseini et al. 2008), three retrofit designs, including:
1) increasing the strength and stiffness of the bridge by
adding the fixity of deck connections to abutments in
lateral direction and using diagonal bracing elements
at one end of the deck beside the bridge abutments
in longitudinal direction, 2) adding some diagonal
braces between columns in transverse direction and at
one end of the deck beside abutments in longitudinal
direction, and 3) using dampers in diagonal members
between piers in lateral direction and also at one end Figure 2. The maximum values of displacement response
of the deck in longitudinal direction were investigated histories (cm) of the retrofitted bridge with the three designs.
for selecting the most appropriate retrofit technique

(Figure 1).
To find out if these retrofit techniques are adequate,
in addition to another set of POA (Hosseini et al.
2008), the results of which can not been presented here
because of lack of space, a set of Nonlinear Time His-
tory Analyses (NLTHA) was performed by using some
appropriate accelerograms. To get a better insight into
the differences between the three retrofit design and
also the effect of damping value of the dampers used
in the third retrofit design the maximum displacement
responses of the retrofitted structures, in both longi-
tudinal and lateral directions, and also those related
to different values of damping coefficient of the used
dampers in the third retrofit design were obtained.
Figure 2 shows a sample of these results.
Based on the numerical results the third design leads
to lower response peak values. However, the final deci-
sion should be made based on technical and financial
considerations.
REFERENCE
Hosseini, M., Banaeizadeh, Y., and Khavari, S. R. 2008. Vul-
nerability Assessment of an Existing Highway Bridge by
Push-Over and 3-D Nonlinear Time History Analyses for
its Seismic Retrofit Design, The 6th US National Seismic
Conference on Bridges and Highways, Charleston, SC,
Figure 1. The three proposed retrofit designs. USA.
640
Experimental evaluation of spliced PSC-steel-PSC hybrid girder connected

with advanced joints consisting of perfobond ribs
S.-H. Kim, J.-H. Yoon & S.-J. Park

School of Civil & Environmental Engineering, Yonsei University, Seoul, Korea
J.-H. Won & C.G. Lee

GS Engineering & Construction, Seoul, Korea
S.-J. Kim
Sampyo Engineering & Construction, Seoul, Korea
ABSTRACT
A spliced I-girder bridge is to connect prefabricated

segments which are made in the factory and trans-
ported to a site. This girder usually uses PSC segments
since it can be easily connected by tendons. However,
this study focuses on the steel-PSC hybrid system.
Thus, it is necessary to study the joint which can be
applied to both spliced concept and hybrid system.
This study proposes the joint consisting of multiple
perfobond ribs in connecting spliced steel and PSC
segments. The test beam consists of a steel component
in the center, two PSC components at both ends, and
two joints connecting the steel and PSC components.
Figure 1. Photograph of test set-up.
However, joints are embedded into the PSC compo-
nents by casting concrete to the PSC and joint at the
same time. Thus, two spliced PSC segment and one
steel segment is fabricated in a factory. Then, these
segments are transported to a test site and assembled
into one girder. The spliced steel segment can be con-
nected to the joint part of PSC segment by bolting,
since the joint also has a connecting steel section.
For the spliced PSC-steel-PSC hybrid I-girder with
a length of 40m, a full scale test is conducted under
three-point loading as shown in figure 1.
The first crack is visually detected at the interface
between the end of lower steel plate in joints and PSC
components. As the load increases, typical flexural Figure 2. Load-displacement curve at the center.
cracks propagate to the top surface of the girder. In
with joints consisting of perfobond ribs is found to
the final state, tensile failure at the bottom occur in the
be an efficient system in order to guarantee sound per-
PSC segment, whereas the joint part do not failed. The
formance up to the ultimate capacities of both the PSC
load-displacement curve at the center of the steel-PSC
and the steel girder.
hybrid test beam is plotted in figure 2. The displace-
ment responses are linear until the first crack forms.
After the first crack occurs, nonlinear behavior is initi- REFERENCES
ated due to the crack propagation. The ultimate load is
1,812 kN, corresponding to displacements of 601 mm. Oguejiofor, E.C. and Hosain, M.U., 1997. Numerical analy-
sis of push-out specimens with perfobond rib connectors.
Experimental results show that the ultimate strength
Computers & Structures, 62(4): 617624.
of the spliced hybrid girder is determined by the Won, J.-H., Lee, C.-G., Yoon, J.-H., and Kim, S.-H., 2009.
strength of the PSC part. In addition, the spliced hybrid Experimental study on behavior of joint in hybrid steel-
girder shows sufficient ductility ratio, even though per- concrete beams. Proc. 5th International Symposium on
fobond ribs are considered as stiff shear connectors. Steel Structures, Korean Society of Steel Construction,
Therefore, the spliced hybrid PSC-steel-PSC girder Seoul, Korea: 455458 (in CD).
641
Construction of the hybrid cable-stayed bridge
D.G. Kim, H.C. Kwon & K.J. Lee

Dealim Industrial Co. Ltd, Seoul, Korea
ABSTRACT is 103 m and constructed by auto climbing form. The

main span of 327 m across the lake is constructed using
Cheong Poong Bridge is hybrid cable-stayed bridge, the free cantilever method and the side span of 57.5 m
which has main span of 327meters crossing over the is constructed by the full staging method.
lake Chung Ju, Korea. The Construction of the foundations and pylons
The bridge has spans of 57.5 m + 327 m + 57.5 m, started in March 2006 and superstructure construc-
which makes the uplift reaction by the unbalance of tion began in April 2007. It is now under construction
the span composition. Therefore, Hybrid system that and expected to open to traffic at year of 2010. The
is the combination of composite section (reinforced Cheong Poong cable-stayed bridge would play a role
concrete deck and I-shaped steel girder) for the cen- of symbolic landmark for the tourism development in
ter span and concrete section for side spans has been Chungcheong province, Korea as well as of transporta-
adopted, accepting the decreasing effect of weight bal- tion by replacing the existing concrete girder bridge.
ance. And the bridge is located at inland unusually in
Korea for cable-stayed bridge, floating crane and large
barge which are needed for erection steel segment can REFERENCE
not access inland. So Construction method of girder
by girder and panel by panel erection is applied and D-H. Yoo, J-S. Ko and C. Moon., Design of Cheong-poong
(Steel-concrete Hybrid cable-stayed) bridge, 33rd IABSE
weight of erection unit can be controlled in less than
Symposium, September 2009.
200 KN, consequently applying the light weight of der-
rick crane. Concrete pylon height with typical H-shape
Figure 1. A view of Cheong Poong Bridge.
642
Low cycle fatigue strength of cruciform welded joints

considering plate thickness
K. Kinoshita & K. Ueda

Department of Civil Engineering, Gifu University, Japan
ABSTRACT
Welded joints of steel bridge piers may have thick

plate thickness which could reduce fatigue strength
of welded joints. Hence, it is necessary to grasp the
influence of the plate thickness on fatigue strength of
welded joints. In this study, the influences of the plate
thickness ranging from 25 mm to 40 mm on low cycle
fatigue strength were investigated by bending fatigue
tests under large plastic strain on cruciform welded
joints.
The test set up is shown in Figure 1. Span was
450 mm, and three-point-bending tests under displace-
ment controlled conditions were done. To conduct the
alternating loading condition, we upset the specimen
for every cycle. Applied plastic strain range at weld toe
was over 10%, which was based on previous studies Figure 1. Test set up.
(for example, Sakano et al. (1997)). However, mea-
suring the plastic strain range at weld toe was difficult
because of the limit of strain measurement. So, we car-
ried out finite element method (FEM) analyses before
fatigue tests in order to obtain the relationship between
local plastic strain range at weld toe and plastic strain
range at the location of measuring strain gauge, which
was located in 5mm apart from the weld toe.
As a result, we can apply the local plastic strain at
the weld toe over 10% when the plastic strain at strain
5mm apart from the weld toe is 5%. Moreover, it can
be expected that fatigue strength of specimens with
40 mm thickness is lower than that of specimens with
25mm thickness.
From fatigue test results, the cracks on the specimen
with both 25mm and 40 mm thickness initiated from Figure 2. Low cycle fatigue test results.
the weld toe, and these cracks propagated widely. This
result indicates that the fatigue failure was governed
by crack initiating from the weld toe. The influences the cruciform welded joints. This result is reasonable
on low cycle fatigue strength of plate thickness were because the local strain increases with increase in the
examined using the relationship between plastic strain plate thickness as mentioned in the results of FEM
range and the number of cycle as shown in Figure 2. analyses.
Figure 2 shows the low cycle fatigue strength at the
number of cycles when 5 mm crack was detected on
specimens. Figure 2 shows that scatter of both results REFERENCES
is small. Furthermore, by increase of the plate thick-
Sakano et al. (1997) Low cycle fatigue behavior of steel
ness, the fatigue strength decreased approximately by
beam-column joint, Journal of structural mechanics and
40%. Consequently, the increase in the plate thick- earthquake engineering 563(I-39):4960.
ness may affect the low cycle fatigue strength of
643
An analysis on the torsion of hybrid bridges with corrugated steel webs

considering tensile strength of concrete
H.J. Ko & H.E. Lee

School of Civil, Environmental and Architectural Engineering, Korea Univ., Seoul, Korea
J. Moon
Civil & Environmental Engineering, University of Washington, Seattle, State of Washington, USA
ABSTRACT In this study, to analyze the torsional behavior of

the composite box-girder with corrugated steel webs
Even though composite box-girders with corrugated by proposed method, the analysis program having an
steel webs are widely used in civil engineering prac- iterative algorithm is developed. For discretized points
tice, the torsional behavior of composite box-girders of maximum surface strain, assumed values of effec-
with corrugated steel webs is not fully understood yet. tive thickness and crack angle are updated iteratively
Mo et al. (2000) studied the torsional behaviors of by all equations established in each step. Final result
the composite box-girder with corrugated steel webs. can be acquired pictorially by plotting all points for
They developed the analytical model for torsional each maximum surface strain discretized.
behavior of the cross section of the composite box- The developed program is successfully verified by
girder with corrugated steel webs. Then, Mo et al. comparing with the experimental result, and it is found
(2006) proposed the torsional design procedure for that the analytical result shows good accuracy for the
the composite box-girder with corrugated steel webs torque-rate of twist relationship by considering the
based on the analytical model proposed by Mo et al. effect of tension behavior of the concrete.
However, the tension behavior of concrete is not con-
sidered in their analytical model so that the initial
stiffness is generally underestimated. In addition, their REFERENCES
analytical models only applicable to the rectangular
section. Moon, J. et al. 2009a. Shar strength and design of trape-
In this study, analytical model proposed by Mo et al. zoidally corrugated steel webs. J Constr Steel Res.
is modified for more general case. 65:1198205.
At first, the relationship between stresses of rein- Moon, J. et al. 2009b. Lateral-torsional buckling of I-
forcement and concrete was established by the space girder with corrugated webs under uniform bending.
Thin-Walled Struct. 47:2130.
truss analogy to express the effective thickness of the
Yi, J. et al. 2008. Interactive shear buckling of trapezoidally
slab and crack angle in terms of each stress. The princi- corrugated webs. Eng Struct. 30:165966.
pal tensile stress was not neglected in this relationship Mo, Y.L. et al. 2000. Torsional Behavior of Prestressed Con-
to fully consider the property of concrete. Then all crete Box-Girder Bridges with Corrugated Steel Webs.
strains of reinforcement and concrete were expressed ACI Structural Journal, 97(6):849859.
as equations of the effective thickness of the slab and Mo, Y.L. et al. 2006. Torsional Design of Hybrid Con-
crack angle using compatibility conditions. crete Box Girders. J. Struct. Engrg. ASCE, 11(3):
For the constitutive law, this model is considered 329339.
the softening effect for the compressive behavior and Hsu, T. T. C. 1988. Softened Truss Model Theory for Shear
and Torsion. ACI Structural Journal, 85(6):624635.
the tension stiffening effect for the tensile behavior in
Belarbi, A. & Hsu, T. T. C. 1995. Constitutive Laws of
concrete slab. Softened Concrete in Biaxial Tension-Compression. ACI
Through these equilibrium equations, compatibility Structural Journal, 92(5):562573.
conditions and the constitutive law, the relationship Collins, M. P. 1972. Torque-Twist Characteristics of Rein-
between external torque, internal stresses, internal forced Concrete Beams, Inelasticity and Non-Linearity in
strains and twist angle can be established. Structural Concrete, University of Waterloo Press.
644
Application of a new metal spraying system for steel bridges. Part 5.

Analysis of the corrosion prevention mechanism of the system
T. Kondo
Institute of Technologists, Saitama Prefecture, Japan
S. Okuno
Nitto Engineering Co. Ltd., Osaka Prefecture, Japan
H. Matsuno & M. Ohshiba
Dai Nppon Toryo Co. Ltd., Osaka Prefecture, Japan
ABSTRACT this paper, the evaluation results that include surface

visual observations, electrochemical potential mea-
An outline of a new metal spraying system that has surements of the exposed metal surface, scanning
been applied to steel surface at room temperature, electron microscope observations of the specimen sec-
the corrosion inhibition effect of the system, and tion, and electron probe microanalysis of the specimen
the corrosion prevention mechanisms of the zinc- surface and section are reported. On the basis of the
aluminum pseudo alloy-sprayed steel structures had above results, the corrosion prevention mechanism
been introduced in IABMAS04 and IABMAS06. with the metal spraying system is discussed in com-
Furthermore, the reference service life prediction parison with other sorts of sprayed metal. Main fruits
based on the actual condition of steel bridges applied of this research are the following.
the metal spraying system had been reported in IAB-
(1) It is judged that zinc-aluminum pseudo alloy-
MAS08.
sprayed film that consisted of a mixture of both
Japan often experiences high humidity and hot tem-
metals maintains least noble potential on the
peratures, especially during the summer. Therefore,
basis of the measurement of the electrochemical
metallic corrosion prevention techniques are impor-
potential.
tant requirements for the steel construction projects.
(2) It is considered that zinc-aluminum- sprayed film
With this in mind, a new metal spraying system was
has sufficient effect of anodic protection working
developed.
according to corrosion occurring in the X-cut part.
Conventional galvanized steel has been used as a
(3) It is recognized that aluminum or zinc-aluminum-
corrosion prevention method for many years. How-
sprayed film, even if nonpaint coating, has corro-
ever, such technique mars the beauty of the scenery,
sion surrounding interception effect that restrains
and it does not have sufficient durability in a serious
chlorine penetration. Meanwhile, zinc-sprayed
corrosive environment. If a paint coating is applied
film without paint coating is not expected to main-
to a galvanized surface, it will not have sufficiently
tain the corrosion surrounding interception effect.
hardened film adhesion.
Addition of the paint coating to the zinc film is
Further, conventional metal spraying systems
expected to maintain the above effect. However, it
require stringent measures to prepare a clean surface,
is important that inhibition of the chlorine pene-
and they must operate at a high temperature more than
tration with combination of sealing materials and
the melting point of the sprayed metal. Therefore, con-
paint is discussed.
ventional metal spraying systems require a high- grade
(4) It is considered that aluminum in the sprayed film
blasting process for steel surface preparation, and this
is not easy to be oxidized based on oxygen analy-
technique cannot be used on covered surfaces such as
sis in the sprayed film. Zinc in the zinc-aluminum
concrete or mortar, which have poor heat resistance.
pseudo sprayed film has less oxygen than zinc sin-
The new metal spraying system, a useful metal-
gle sprayed film. Therefore, it is presumed that
lic corrosion prevention method suitable for some
zinc oxidation is prevented by mixed aluminum.
steel bridges, has been experimentally evaluated. The
And, it is judged that aluminum contributes to
sprayed complex zinc and aluminum film was pro-
restrain oxidation consumption of the sprayed film
nounced to offer corrosion resistance in excess of steel
that maintains corrosion inhibition effect.
galvanized with 550 g/m2 of zinc. Adhesion of the
sprayed metal is very stable with a surface preparation
method that coats the primer, which is a substratum REFERENCE
surface preparation material. ResearchAssociation of Metal Spraying System for Structures
Some kind of metal-sprayed specimen had been (2006) Design, Construction & Repair Manual for Steel
exposed the harsh seashore for 8 years and a half. In bridges.
645
Assessment of existing steel structures Recommendations

for estimation of the remaining fatigue life
B. Khn
Leader Department for Bridge Construction, Verheyen-Ingenieure GmbH & Co. KG, Bad Kreuznach, Germany
ABSTRACT are publisher as Joint-JRC-ECCS-Scientific Technical

Reports, that may be downloaded from the e-bookshop
In many countries and regions, the traffic infra- of the Commission, see http://eurocodes.jrc.ec.
structure projects suffer from low funding. There bud- europa.eu.
get is tight for new infrastructure building and, thus, These Technical Reports are being prepared by
the importance of inspection, maintenance and assess- experts from CEN/TC 250/SC3, the CEN-Technical
ment of the existing traffic infrastructure increases. Committee responsible for the preparation and further
A new fatigue assessment guideline for the estimation development of Eurocode 3, and from ECCS Techni-
of the remaining fatigue life of steel bridges has been cal Committees related to the subject in question. They
written by technical committee 6 from ECCS [Khn et aim at:
al. 2008]. It will be a useful tool for the complementa-
tion of bridge management systems, used commonly Giving the scientific background and further expla-
for condition assessment. nations to the Eurocode 3 rules as published,
Design specifications and rules are harmonised Presenting the state of the art and preparing the
throughout Europe. They are under constant devel- field for the maintenance, further harmonisation
opment, but there is still a lack of forwarding and and further development of the Eurocode 3 rules.
concentrating experiences as well as developing rules
for the fatigue assessment on existing steel struc- A high priority project of CEN/TC 250 is to include
tures. This paper presents a guideline with a proposed in the Eurocodes technical rules for the assessment
fatigue assessment procedure for existing steel struc- and retrofitting of existing structures, which becomes
tures embedded in information about old materials and more and more important in the context of sustain-
non-destructive testing methods for the evaluation of able development in the construction field. To this
details. Particular attention is paid on remedial mea- end the presented JRC-ECCS-Joint Report has been
sures which are proposed for weak details and damages published.
caused by fatigue. The developed fatigue assessment
procedure can be applied to existing steel structures
under cyclic loading in general, but the guideline con- REFERENCE
centrates on the existing traffic infrastructure made
from old steel, because of the public importance. Khn, B. Luki, M., Nussbaumer, A., Gnther, H.-P.,
The proposed procedure summarizes, regroups and Helmerich, R., Herion, S., Kolstein, M.H., Walbridge,
arranges the knowledge in the field of assessment on S., Androic, B., Dijkstra, O., Bucak, . 2008. Assess-
existing steel to be applied by practicing engineers.The ment of Existing Steel Structures: Recommendations for
Estimation of Remaining Fatigue Life. JRC Scientific
procedure is a milestone in knowledge transfer from a
and Technical reports. Joint Report Prepared under the
state of scientific knowledge to state-of-the-art. JRC ECCS cooperation agreement for the evolution
According to the agreement between the Joint of Eurocode 3 (programme of CEN / TC 250), Editors:
Research Centre (JRC) of the European Commis- G. Sedlacek, F. Bijlaard, M. Gradin, A. Pinto and S.
sion in Ispra and the European Convention for Dimova, First Edition, EUR 23252 EN, ISSN 1018-5593,
Construc-tional Steelwork (ECCS) a series of reports JRC 43401
646
Computational methodologies for the blast vulnerability assessment

of steel bridge girders
L.D. Leininger & A.M. Coughlin

Hinman Consulting Engineers, San Francisco, California, USA
ABSTRACT This comparison identifies the relationship between

the scaled range, Z, and the girder response for both the
A challenge for bridge security is to efficiently develop fast-running SDOF model and the high-fidelity finite
a thorough risk assessment accounting for numerous element model. The results indicate that the flexural
bridge assets within an integrated transportation net- response of a long-span, stiff, girder exposed to a man-
work, and prioritize those assets for the allocation portable threat may not govern the solution at scaled
of blast retrofit/mitigation funding. Typically, a risk ranges once thought to be compatible with such an
assessment includes quantification of the vulnerabil- assumption. Future analysis will examine the trade-
ity of tens to hundreds of individual bridge spans offs between the time it takes to perform a high-fidelity
and numerous associated explosive threats. For these model and the advantage to the additional refinement
assessments, it is desirable to have a robust, fast- as well as the ideal scaled range for getting the most
running, computational tool that can give an accurate accurate SDOF results.
scoping value for the blast vulnerability of the primary
structural members. These scoping calculations popu-
late risk ranking strategies like those recommended in REFERENCES
AASHTO (2002) and FWHA/AASHTO (2003). Once
scoping assessments have been made and risk priori- AASHTO. Guide to HighwayVulnerability Assessments. May
2002.
tization has been set, only the highest risk spans are
Biggs, J.M. 1964. Introduction to Structural Dynamics,
reevaluated with higher-fidelity models. McGraw Hill: New York.
This paper presents study of a bridge girder FWHA/AASHTO). Recommendations for Bridge and Tunnel
response using both a fast-running SDOF (Single Security. 2003.
Degree of Freedom) dynamic modeling technique Nichols, A. Users manual for ALE3D: An arbitrary
as defined in dynamics textbooks (Biggs 1964) and Lagrange/Eulerian 3D code system. LLNL technical
recommended in military handbooks (UFC 3-340- report: UCRL-MA-152204 Rev 6. 2007.
01 2008 & UFC 3-340-02 2002) and a non-linear, UFC 3-340-01, TM 5-855-1, AFPAM 32-1147(I), NAV-
dynamic, 3-dimensional fully-coupled, finite element FAC P-1080, DAHSCWEMAN-97 Design and Analysis
of Hardened Structures to Conventional Weapon Effects
model inALE3D (Nichols 2007). The blast response of
(FOUO). 01 June 2002.
a representative steel bridge girder shown in Figure 1 UFC 3-340-02. Structures to Resist the Effects of Accidental
is computed and discussed for the two methodologies. Explosions. 5 December 2008. (supersedes TM5-1300).
Figure 1. Sample bascule bridge girder used in the study.
647
Design live load for long span bridges
M. Lutomirska & A.S. Nowak

ABSTRACT (CDF) of uniformly distributed load for considered

span lengths using all results, daily maximum and
The available live load models were developed for weekly maximum combinations. The mean value of
short and medium span bridges. The present paper uniformly distributed load oscillates between 7.30 and
deals with the development of live load for long span 10.95 kN/m (0.50 and 0.75 k/ft). The bias factors (ratio
structures. The maximum loading for a short span of mean to nominal) were calculated for the heav-
bridge is caused by the heaviest possible trucks that iest 75-year combination of vehicles. The 75-year
can travel over it. In contrast to short and medium uniformly distributed loads were derived from extrap-
spans, long span live load must be considered as traffic olated distributions. It was noticed that the bias factor
mix, where one heavily overloaded truck does not have values for some sites do not exceed 1.25, which is
significant influence. The roadway is not necessarily similar as for short and medium spans, as shown in
entirely covered by the heaviest trucks. As the load the NCHRP Report 368 (1999). It is recommended
length increases, the load per unit length will decrease to use HL-93 also for those long spans. Brides in
and its distribution will be closer to uniform, because localizations with high ADDT and high percentage of
cars and other light vehicles are injected into the traffic overloaded trucks require development of site specific
stream. For long span bridges critical scenario is traffic models. For some sites with very heavy traffic, the bias
jam situation with vehicles moving at crawling speed factor reaches value 2.0. Therefore, for those bridges
and very small headway distance. the uniformly distributed load should be higher. The
The number of trucks and participation of heavy coefficient of variation is calculated from the slope of
vehicles in the traffic patter have a trend of rapid transformed CDF from weekly maximum values.
growth. Higher percentage of trucks runs overweight, The developed live load model is recommended to
particularly because it is to their economic advan- be taken into consideration in the AASHTO LRFD
tage. The design live load based on actual weigh-in- Bridge Design Specifications.
motion data reflects current traffic patterns, quantities
of trucks and their weights. The available data base
includes weigh-in-motion (WIM) truck surveys and REFERENCES
videos of the traffic jam situations taken on selected
bridges. The weigh-in-motion database was obtained AASHTO LRFD Bridge Design Specifications. (2007).
American Association of State Highway and Transporta-
for a variety of sites from the project NCHRP 12-76.
tion Officials, Washington, D.C.
The available WIM data served as a basis for simu- Nowak, A.S., 1999. Calibration of LRFD Bridge Design
lation of a traffic jam situation. Values of uniformly Code. NCHRP Report 368, Transportation Research
distributed load were derived for a wide spectrum Board, Washington, D.C., USA
of span lengths between 180 m and 1500 m (600 ft Sivakumar, B., Ghosn, M., Moses, F., 2008. Protocols for
and 5000 ft). Trucks were kept in actual order, as Collecting and Using Traffic Data in Bridge Design. Final
recorded in the WIM surveys. Results of the simula- Report, NCHRP Project 12-76.
tions were plotted as a cumulative distribution function
648
Management of interstate 95 in Pennsylvania
T.P. Macioce
Pennsylvania Department of Transportation, Harrisburg, Pennsylvania, United States of America
H.M. Berman
Pennsylvania Department of Transportation, King of Prussia, Pennsylvania, United States of America
D. Constable
Federal Highway Administration, Harrisburg, Pennsylvania, United States of America
ABSTRACT Maintenance and preservation of the bridges are key

to uninterrupted operations. The objective is continu-
The Interstate highway system is the primary trans- ally performing routine maintenance activities, e.g.,
portation network in the United States. On the eastern water blasting of beams, structural plating or patch-
coast, Interstate 95 links Maine to Florida. The seg- ing, to extend the service life of the structures until
ment through Pennsylvania is 82 km in length and corridor reconstruction occurs. The need to perform
consists of numerous structural transportation assets emergency repairs occasionally occurs, however, and
including 209 bridges, 13 dynamic message sign must be dealt with while maintaining the confidence
structures and 28 traffic incident cameras. Effective, of the traveling public.
proactive administration of bridge assets on this corri- Planned improvement projects in the corridor con-
dor is essential to continued safe traffic operations on sist of projects ready for construction, projects that are
the Interstate. mid-term and long range projects. For the I-95 corri-
The key aspects of Pennsylvanias I-95 corridor dor, in the near term, approximately $350 million in
management approach include bridge administration, construction will be bid for a variety of structural and
inventory of the bridge assets, maintenance and preser- interchange improvement projects. Over the next ten
vation of the bridges, and the planning for improve- years approximately $2 billion in construction projects
ments of the corridor which have become the funda- will occur. The long range plan of projects occurring
mental ingredients for the successful support of the in the next 20 years and beyond contain forecasts for
vital Northeast region infrastructure component. an additional $4 billion in construction.
Bridge administration is a broad umbrella effort The over-arching goals of PennDOTs asset man-
covering many functional activities, including bridge agement strategies with an emphasis on a bridge
evaluation by inspection and load rating analysis, program including Interstate 95 are:
performing on-demand structural repairs, preserva- Ensuring that bridges are safe for the efficient
tion, and routine maintenance to realize the maximum
movement of people and goods.
useful life of a bridge possible before its eventual Reducing the backlog of bridge deficiencies
replacement. Using good practice in design, construction, and
The bridge asset inventory is obtained and updated
maintenance to sustain the continual improvement
as changes are identified through the bridge safety
of our bridges in a cost effective manner and achieve
inspection program at the element level. The capture of
a bridge life of 100 years.
element level data provides detailed information on the
deficiencies and modes of distress contributing to the The approach and strategies employed in Pennsyl-
overall condition ratings, appropriate repair strategies, vania to manage the I-95 corridor will ensure the goals
quantities or extent of distress and associated costs. of safety, mobility, quality of life for the Philadelphia
This allows for improved asset management decisions. metropolitan region.
649
Monitoring of the Manhattan Bridge and interferometric radar systems
L. Mayer
IDS Ingegneria Dei Sistemi S.p.A., Pisa, Italy
B. Yanev
New York City DOT, New York, NY
L.D. Olson
Olson Engineering, Inc., Wheat Ridge, Colorado
A. Smyth
Columbia University, New York, NY
ABSTRACT
Presented herein is a monitoring and measurement

study of the vertical and torsional displacements of
the midspan of the Manhattan Bridge using Interfero-
metric Radar and Global Positioning Systems (GPS).
The Manhattan Bridge is a particularly interesting case
study due to its immense size, unique loadings, high
traffic volume, age, and recent multimillion dollar
rehabilitation and stiffening program. The main span
of the Manhattan Bridge is 1470 ft with an overall Figure 2. Vertical displacement time series data of the trans-
verse beams measured by the IBIS-S system during train
anchorage-to-anchorage length of 2920 ft. The cur- passage.
rent configuration includes 7 vehicular traffic lanes and
four subway tracks. The total amplitude of displacement, peak to peak
The Interferometric radar system is an non- was on the order of 300 mm when loaded by a single
contacting, innovative microwave radar sensor train. Both measurement systems (GPS and Interfero-
(IBIS-S) used to simultaneously measure the displace- metric Radar) compared well with one another and are
ment and vibration responses of multiple locations of promising technologies in the evolving area of bridge
a structure from distances up to 0.5 kilometer. The deflection and vibration measurements. The GPS sys-
IBIS-S system has a maximum accuracy of 0.01 mm tem lends itself to long-term monitoring while the
(0.0004 inch) and a maximum sampling frequency of IBIS-S system can be rapidly deployed for short-term
200 Hz (Nyquist frequency of 100 Hz). displacement and vibration monitoring. Both systems
GPS systems use triangulation from satellite signals have advantages over traditional bridge instrumenta-
to accurately locate the absolute position of a receiver tion (e.g. strain gauges and accelerometers) in terms
and are routinely used in a variety of applications. of the ease of deployment and parameters measured
The systems were employed to measure the midspan (displacement).
deflections of the bridge under normal automobile
and train traffic loadings. The GPS data characterizes
the maximum deflections as well as deflection time REFERENCES
histories simultaneous at 80 points along the midspan.
Kogan, Mikhail G., Won-Young Kim, Yehuda Bock, and
Andrew W. Smyth, Load response in a large suspen-
sion bridge during the NYC marathon revealed by GPS
and accelerometers, Seismol ogical Research Letters,
79(1):1219, February 2008.
Taylor J.D. (Ed.), Ultra-wideband radar technology. CRC
Press, 2001.
Henderson F.M. & Lewis A.J. (Eds.), Manual of Remote Sen-
ing. Principles and Applications of Imaging Radar, 3rd
Ed., Wiley & Sons, 1998.
Skolnik M.I. (Ed.), Radar Handbook. McGraw-Hill, 1990.
Figure 1. Manhattan (foreground) and Williamsburg (back- Marple S.L. Jr., Digital spectral analysis with applications,
ground) Bridges. Prentice-Hall, 1987.
650
A rational partial composite bridge beam transition
P.S. McCarten
Opus International Consultants Limited, Napier, New Zealand
ABSTRACT
The New Zealand Transport Agency Bridge Manual

requires composite steel girder-concrete slab bridge
beams to be assessed to the NZS 3404 Steel Design
requirements. To account for shear connector fracture
this standard limits the shear connection capacity for
partially composite beams to 50% of that required for
a fully composite beam. This criterion is unduly severe
often requiring load restrictions to be applied and this
typically does not correlate with bridge performance.
Simplified plastic analysis methods for determining
the flexural capacity of a steel girder-concrete slab
bridge beam require the shear connector strength
determined for the typical static failure modes, being
concrete bearing and shear connector shear strength,
either in the gross section or fixing to the steel girder
flange. Shear connector fracture is also a critical limit
state through the partial composite beam transition that
needs to be appropriately assessed.
The paper summarises desk top research review-
ing trends in the flexural capacity versus shear connec- Figure 1. Transition Interaction Curves (Burnet and
tion capacity interaction curves for composite beams Oehlers, Fully Smoothed and Discrete Smoothed) for two
with the neutral axis in either the concrete slab or the beam cases.
steel beam. This review has shown the following:
Reliability of the shear connector capacity can be
used to model the shear connector fracture limit state Calibration of the model is limited and further test-
through the transition ing is suggested to confirm the appropriateness of the
A discrete smoothed empirical transition model is evaluation procedure.
rational, repeatable and gives consistent reliability
to fracture failure, refer to Figure 1
Correlation to the shear connection limits specified REFERENCES
in international standards is met
For NZS 3404 Steel Design an approximate 20% New Zealand Transport Agency, 2004. Bridge Manual
increase in load carrying capacity can be achieved New Zealand Standards, 1997. NZS 3404: Steel Design
Burnet M.J. and Oehlers D.J., 2001 Fracture of mechanical
with this typically negating the need to load restrict shear connectors in composite beams. Journal Mechanics
bridge beams of Structures and Machines, 29(1), 141
The method has been specifically developed for
application to an extreme positioned overload vehicle
load case where fatigue effects can be ignored.
651
Structural health monitoring for bridge management
P.S. McCarten
Opus International Consultants Limited, Napier, New Zealand
ABSTRACT comprehensive suite of SHM tools for Bridge Man-

agers to effectively manage their bridge stock. In this
Structural Health Monitoring (SHM) is a new tech- paper an example SHM programme is outlined for
nology in bridge and structure engineering and its a typical high risk durability site to show how the
integration into bridge and structures management SHM could be integrated into the bridge manage-
has not yet been determined. The technology involves ment reporting. From this example it is observed the
tracking of any aspect of a bridges health by using greatest benefits for a proactive preventive mainte-
reliably measured data and analytical simulations in nance bridge management regime are associated with
conjunction with heuristic experience so that the cur- the basic level interrogation of the SHM data. As it
rent and expected future performance of the bridge for is recognized that SHM should not be considered as
at least the most critical limit events can be described equating to a structure evaluation assessment it is con-
in a proactive manner. SHM has clear advantages over sidered the extent of information gains from SHM
traditional Non Destructive Testing and Evaluation needs to be reviewed to ensure SHM is cost effective.
because once the monitoring is installed data flow is Recent research has shown that risk based con-
continuous. dition measures correlate well with bridge funding
Historically bridge management was focused on investment and it has been determined that if basic
defect management with inspection programmes sup- interpretive data from an SHM programme can be
porting the development of maintenance programmes. integrated into the performance measures then more
Performance measures for bridges are being developed robust data will be available and SHM will be more
as a management tool allowing the effectiveness of readily justified.The paper considers these matters pre-
work programmes to be reported. The performance senting options for integrating SHM data into the full
measures developed to date utilize qualitative con- range of likely performance indicators. With the devel-
dition information, rather than defect information, opment of robust performance indicators informed
and will continue to evolve as the industry better and transparent bridge management decision making
understands and defines the holistic asset management will be achieved, and, enhanced bridge management
inputs required to maximise the effectiveness of lim- reporting made possible.
ited budgets and be able to present the outcomes to the
wider community. Performance Indicators for bridge
Essentiality, Functionality, Serviceability, Safety and REFERENCES
Stability are being considered. Structural indicators
being considered include Condition, Reliability, Risk McCarten P.S., 2004. A systems method for defining best
and Work Record. The vision is for structural perfor- bridge maintenance management practice. Proceedings
mance to be based on two prime factors, a risk indicator 2nd International Association of Bridge Maintenance and
presenting structure strength and a condition indicator. Safety Management Conference, Kyoto, Japan: Balkema
McCarten P.S., 2009 Justifying Structural Health Monitor-
It is recognized that SHM is an evolving field and
ing. 7th Austroads Bridge Conference, Auckland, New
has matured significantly over the last decade with a Zealand
focus on developing reliable means of acquiring, man- Aktan A.E., Catbas F.N. Grimmelsman K.A. & Pervizpour
aging, integrating and interpreting reliable structural M., 2003 Development of a Model Health Monitoring
data for maximum information gain and at lowest cost. Guide for Major Bridges. Report to FHWA Research and
The paper also recognizes the importance of having a Development DTFH61-01-P-00347, Drexel University
652
Destructive test of a steel slab-on-girder bridge
J. McConnell, M. Chajes, H. Shenton III, K. Michaud & C. Russo

J. Ross
Larson Design Group
ABSTRACT
This paper presents the testing procedures and results

from a unique destructive bridge test of a steel I-girder
bridge. The subject bridge was in good condition and
only decommissioned due to traffic pattern improve-
ments. The goal of the test was to load the structure
well beyond its elastic limit in an attempt to observe
transverse moment redistribution and investigate the
ultimate capacity of the structure. Instead, the test
revealed the tremendous reserve capacity of the struc-
ture. The bridge was able to withstand a loading
mimicking a design truck with a load equal to 17 times Figure 1. Cross-section strains with increasing load.
an HS-20 vehicle without the data indicating any yield-
ing of the girders or any other significant damage being
observed.
A second usefulnesses of the test data is quanti-
fying the load distribution throughout the structure.
As an initial step to quantifying the load distribution
throughout the structure, the strains through the cross-
section of the girders are analyzed. Figure 1 shows
such strains at four (nearly even) intervals of increas-
ing load at a given girder cross-section, near the center
of the footprint of the applied loading. Each cycle is
the loading equivalent of one HS-20 vehicle (or 72
kips). It is hypothesized from this data that there was
some loss of composite action between cycles 4 and 8
due to the close proximity of the heavily concentrated
loadings to the gauged site. It is then possible that fric-
Figure 2. Comparison of actual and FEA strains.
tion between the slab and the girder resulted in the
increasing height of the neutral axis between cycles 8
two lines are similar after load cycle 8. Additional pre-
and 17. Future work will seek to more fully understand
liminary data analysis suggests that this FEA model is
the specific phenomena and behavior here.
a promising tool for obtaining more detailed informa-
The third application of the field testing data is to
tion about the bridge and potentially extrapolating the
validate a finite element model of the structure. Com-
results to determine the bridges behavior at its ultimate
parisons between the field data and the corresponding
load. Future work will seek to quantify the changes in
FEA results show that (see example in Fig. 2), at low
load distribution or stiffness that were observed dur-
load levels, the finite element analysis is accurate in
ing the testing and incorporate these into the FEA to
predicting the bottom flange strains. However, between
improve its usefulness.
load cycles 4 and 7(for the example shown), there is
very little increase in strain; otherwise the FEA and
field data are in good agreement as the slopes of the
653
Bridge inspection and repair technologies on the expressways in Japan
H. Miyauchi
Central Nippon Expressway Co., Ltd
ABSTRACT for painting corrosion, which is developed by for-

mer Japan Highway Public Corporation. This painting
To ensure expressway driving comfort, we NEXCO- diagnosis system applies image-processing technol-
Central put extensive effort into inspection, repair ogy with the computer, and it is a system evaluating a
and reinforcement work. It is concerned, however, deterioration phenomenon of the steel bridge painting
that the damages of the structures by many factors for the objectivity and fixed-quantity.
will more increase in the future. Therefore, we need
more and more effective maintenance. In this sense,
detailed inspection becomes very important, however, 2 REPAIR
we can not enough inspection because of difficulty of
approaching to bridge, and lack of inspection engi- The repair design is carried out in order to achieve the
neer and so on. In order to effective maintenance, purpose of repair. It consists of planning and consider-
NEXCO-Central is actively developing inspection ing the cause of crack, the degradation mechanism and
systems. degradation prediction, range and scale of the repair,
In this paper, I would like to introduce the inspection environmental condition, safety construction period,
systems, which is carried out by NEXCO-Central, and economic efficiency, degree of importance of the
the typical example of damage of concrete bridges and structure, load influence to the environment, repara-
steel bridges, and also the technologies of inspection bility, and subsequently, appropriate repair methods
and repair. and repair materials, repair time, etc are selected.
1 INSPECTION
3 CONCLUSION
The most commonly employed method for concrete
structure inspection is visual inspection from the road The aging structures are increasing and as it turned
surface. If any defect is detected by this survey, then, out more and more expense for inspection and main-
close range visual inspection and hammering test are tenance is needed in near future. Therefore, we must
conducted by engineers. This method, however, is carry out effective inspection and maintenance under
costly, takes a long period of time, and requires traffic the appropriate evaluation. Then, development of
regulations. Furthermore, the inspection results could inspection and repair technologies is indispensable,
vary depending on the engineer, so there has been and upbringing and improving the skill of the engineer
a strong demand to develop an effective inspection is very important, too.
method. This paper represents several methods being
developed at NEXCO-Central by utilizing ICT tech-
nologies: one is the infrared concrete defects detection REFERENCES
method which can be easily used on sight by engi-
neers, and the others are the digital crack measurement [1] NEXCO-Central, March, 2008 Annual report 2008
method, and inspection method for PC grouting. [2] NEXCO-Central, April, 2006 Manual for maintenance
On the other hand, there are several types of dete- and inspection
rioration of steel structures, mainly fatigue and cor- [3] JSCE, Dec 2007, Standard Specifications for Concrete
rosion. This paper introduces Paint View method Structures 2007, Maintenance
654
Improving transparency of bridge condition for decision-making

and analysis in bridge management
J.T. Newlin
Civil and Environmental Engineering, Bucknell University, Lewisburg, PA USA
W.L. Myers
Penn State Institutes of Energy and Environment, University Park, PA USA
G.P. Patil
Center for Statistical Ecology and Environmental Statistics, Penn State University, University Park, PA USA
S.W. Joshi
Department of Computer Science, Slippery Rock University, Slippery Rock, PA USA
ABSTRACT The hotspot detection tool identified a region in

southwestern Pennsylvania where there is a zone of
Transparency of collected information on bridge counties with an elevated rate of bridges that have not
infrastructure condition leads to more informed and been evaluated for bridge scour risk. The detection
more effective decisions regarding where and how tools also identified a region in southeastern Pennsyl-
to appropriate resources so that the maximum bene- vania where there is a zone of counties with an elevated
fit is gained for the bridge network. The condition of rate of bridges that are considered to be scour critical.
bridges can be made more transparent to the general Comparing the scour critical hotspots to the hotspots
public and bridge managers through the use of spatial for poor overall bridge condition (based on the suf-
and spatiotemporal hotspot detection with surveillance ficiency rating) shows some agreement between the
geoinformatics. A hotspot is an area that is consid- hotspot for scour critical bridges and the hotspot for
ered unusual, such as an anomaly, aberration, outbreak, bridges with a sufficiency rating below 50 percent.
elevated cluster, or critical resource area. Based on
response properties of geospatial data, spatial and spa- REFERENCES
tiotemporal hotspots can be detected. The goal of
hotspot detection is to identify, delineate and assess Federal Highway Administration (FHWA). 1995. Record-
ing and Coding Guide for the Structure Inventory
the significance of hotspots. A scan statistic seeks to
and Appraisal of the Nations Bridges, FHWA-PD-96-
identify hotspots of cells that have an elevated response 001. Washington, D.C.: Office of Engineering, Bridge
compared with the rest of the region. Hotspot detec- Division.
tion can provide surveillance information that can be Kulldorff, M. 1997. A spatial scan statistic. Communications
valuable for mitigation, containment, prevention, and in Statistics: Theory and Methods 26: 148196.
other management strategies. Modarres, R. & Patil, G.P. 2007. Hotspot detection with
With the detection of hotspots, it is possible to bivariate data. Journal of Statistical Planning and Infer-
quickly and graphically identify if there is a region with ence 137: 36433654.
a concentration of bridges having poor ratings. Pre- Patil, G.P., Joshi, S.W. & Koli, R.E. 2009. Digital governance
and hotspot geoinformatics of detection and prioritization
liminary analysis demonstrates how hotspot detection
with environmental policy case studies (invited paper),
can be used to address bridge management questions Joint Statistical Meetings, Washington, D.C.
related to the scour critical rating and to the overall Patil, G.P., Joshi, S.W. & Koli, R.E. 2010. PULSE, Progres-
condition of bridges according to the sufficiency rat- sive upper level set scan statistic for geospatial hotspot
ing in the 2008 National Bridge Inventory for the state detection, Environmental and Ecological Statistics 17: in
of Pennsylvania. The following management questions press.
are explored: Patil, G.P. & Taillie, C. 2004. Upper level set scan statistic for
detecting arbitrarily shaped hotspots. Environmental and
1 Are there hotspots where the bridge scour risk has Ecological Statistics 11: 183197.
United States Government Accountability Office (US GAO).
yet to be determined or where bridge scour risk is 2008. Highway Bridge Program: Clearer Goals and Per-
high? formance Measures Needed for a More Focused and
2 How well do the hotspots for poor overall bridge Sustainable Program, Report to Congressional Commit-
condition (based on the sufficiency rating) relate to tees GAO-08-1043. Washington, D.C.: US Government
the hotspots for high scour risk? Accountability Office.
655
Countermeasures for rain- and wind-induced vibrations on the Meiko Nishi

Bridge
A. Noguchi
Planning and Design Team, Nagoya Branch, Central Nippon Expressway Company Limited
H. Miyauchi
Engineering Inspection Department, Central Nippon Expressway Company Limited
ABSTRACT ii) Viscous shear dampers in particular exhibit very

exceptional damping performance with respect to
The Meiko Nishi Bridge was also the first bridge rain- and wind-induced vibrations.
in Japan on which rain- and wind-induced vibration iii) Comparatively large displacement occurred on
observations were conducted. the round high-damping rubber dampers. How-
A high-damping rubber damper was developed as a ever, based on the results of the damper perfor-
method of dealing with rain- and wind-induced vibra- mance test, these dampers were judged to possess
tions. This high-damping rubber damper improves satisfactory safety with respect to fatigue.
attenuation performance for cable vibrations by con-
A cable excitation test was conducted on the actual
necting the cable and the girder through rubber that
bridge to check the damping performance of the
possesses high energy-absorbing performance.
cables on which round high-damping rubber dampers
However, the high-damping rubber damper cable
and viscous shear dampers had been installed. Based
had failed. The cause was the fatigue that had accumu-
on the results of measurement, the following were
lated in the high-damping rubber damper. The cause of
determined.
the failure was the offset deformation. Offset deforma-
tion is the relative displacement that occurs between i) The logarithmic decrement of the cables on which
the girder and cable as a result of sunlight and changes high-damping rubber dampers and viscous shear
in temperature. Accordingly, it was decided to develop dampers were installed was 0.02 or greater, and
a high-damping rubber damper that was not affected the Scruton number was 60 or greater. Both the
by offset deformation. design values and the measured values were such
Existing high-damping rubber dampers are rectan- that rain- and wind-induced vibrations would be
gular in shape. As it was recognized that the shape unlikely to occur.
of the damper affects damping and fatigue endurance ii) The logarithmic decrement determined from the
performance, the shape was changed to a round shape measured values was only about 60% of the design
to develop a high-damping rubber damper that would value.
not be affected by offset deformation. iii) On the cables that were measured in this test,
In order to check the performance of the dampers, the viscous shear dampers exhibited better per-
the cable displacement was measured continuously formance than the high-damping rubber dampers.
from April through November 2008. Measurements
Judging from the results of long-term measurement
were conducted for the round high-damping rubber
and excitation tests conducted under actual conditions,
dampers and viscous shear dampers.
the measures that have been taken on the Meiko Nishi
Based on the results of past inspections and long-
Bridge to prevent rain- and wind-induced vibrations
term measurement, the following were determined.
will remain effective as long as periodic inspections
i) Viscous shear dampers and round high-damping are not neglected.
rubber dampers exhibit outstanding damping per-
formance with respect to rain- and wind-induced
vibrations.
656
Proposal for modification of load and resistance factors in the AASHTO

LRFD bridge design code
A.S. Nowak & P. Paczkowski

ABSTRACT The resistance factors that are currently in the

AASHTO Code are:
In the new generation of design codes load and resis-
1.0 for prestressed concrete girders in flexure,
tance factors are determined in the reliability based
0.9 for reinforced concrete girders in flexure,
calibration procedure. A general statement for assur-
1.0 for composite and non-composite steel girders
ing acceptable safety is that the factored resistance
in flexure,
of the components shall exceed the factored load
effects. Mathematically it is expressed by basic design The proposed Strength I load combination is
equation:
where Qi is the load effect, Rn is the nominal resistance, and the corresponding reduced resistance factors are:
i is load modification factor, i and are respectively 0.9 for prestressed concrete girders in flexure,
load and resistance factors. 0.8 for reinforced concrete girders in flexure,
Theoretically the load and resistance factors can 0.9 for composite and non-composite steel girders
take any value.The structural performance can be mea- in flexure,
sured in terms of the reliability index, . Therefore, the
selection criterion for design requirements is close- The reliability analyses were performed and it was
ness to the target , or acceptable risk of failure. The shown that new load and resistance factors will result
load factors are expected to be larger than 1, and the in slightly higher reliability indices. The proposed
resistance factors are expected to be smaller than 1. changes will allow for future rational modifications
AASHTO LRFD (2007) load and resistance factors of resistance factors reflecting the improved quality of
were determined in the calibration process (Nowak materials and fabrication.
1999). The statistical parameters of load and resistance
were determined using the data available in 1980s
REFERENCES
(Ellingwood et al. 1980 and Nowak 1999). Live load
was based on the truck survey in Ontario performed AASHTO, Bridge Design Specifications, American Asso-
in 1975. Mechanical properties of materials such as ciation of State Highway and Transportation Officials,
concrete and steel were obtained from literature. Washington, D.C., 2007.
Over the last two decades there were significant Ellingwood, B., Galambos, J.V., MacGregor, J.G., and
improvements in numerical and analytical methods Cornell, C.A., (1980). Development of a Probability
resulting in enhanced accuracy in prediction of the Based Load Criterion for American National Standard
actual resistance. In addition, it was observed that the A58. NBS Special Publication 577, Washington, DC,
National Bureau of Standards.
quality of materials improved over the years (Nowak Nowak, A.S., Calibration of LRFD Bridge Design Code,
and Szerszen 2000). Therefore, resistance factors can NCHRP Report 368, Transportation Research Council,
be increased. However, in cases when the resistance Washington, D.C., 1999.
factors are already set to 1, such an increase could Nowak, A.S. and Szerszen, M.M., Calibration of Design
result in factors that are larger than 1 which can be Code for Buildings (ACI 318) Part 1&2: Statistical Models
philosophically unacceptable. for Resistance, ACI Structural Journal, Vol. 100, No. 3,
This study presents a proposal for modification of 2003, pp. 377382 & pp. 383391.
the load and resistance factors in the AASHTO LRFD
(2007). The current Strength I load combination is:
657
Recommendations for dynamic allowance in bridge assessment
E.J. OBrien & A. Gonzlez

A. nidaric
ABSTRACT to a bridge dynamic load scenario defined by weigh-

motion based probability distributions which are used
Correct evaluation of the behavior of highway bridges to obtain the characteristic value for static load effect.
under heavy traffic loading is extremely important both Similarly, but independently, the distributions are
for the enhancement of design techniques, and also for applied to a dynamic model to obtain the characteris-
the assessment of existing infrastructure. It is widely tic total load effect. Comparison between the total and
accepted that shortfalls exist in the determination of the static results yields the site-specific allowance for
the traffic load which the bridge may be required to dynamic interaction. A method of field measurement
support during its expected lifetime due to inadequate is proposed and tested based on these simulations.
consideration of amongst other factors, the dynamic Some specific issues concerning the dynamic
interaction between the bridge structure and the heavy allowance associated to: (a) pre-existing bridge vibra-
vehicles crossing it. The latter is addressed in deliv- tions; (b) maximum total effects developing in sections
erable D10 within the 6th EU framework ARCHES different from midspan, (c) the existence of a bump
project (20062009), which is summarized in this prior to the bridge, or (d) critical loading cases such
paper. As part of these investigations, lifetime static as cranes, are also discussed in the paper. Finally,
load effect values are combined with realistic dynamic ARCHES gives general recommendations on dynamic
amplification factors to obtain an overall total life- allowance purely based on the bridge length and road
time load effect. This process has two distinct parts: class, which represent a significant reduction in rela-
(1) The calculation of bridge static load effect due tion to Eurocode values for one-lane bridges and ISO
to site-specific traffic flow along with the resultant road class A. These values can be further reduced if a
assessment of bridge lifetime static load effect, and the better knowledge of the bridge response was acquired
selection of those loading events that are deemed criti- through numerical simulations and field tests. In fact,
cal (statically); and (2) The assessment of the levels of the analysis of bridge measurements collected dur-
dynamic interaction occurring between a bridge and ing the ARCHES project led to dynamic allowances
its associated vehicular traffic. close to 1.0 for the heaviest loading scenarios in dif-
The procedure to obtain a site-specific dynamic ferent sites. Site-specific parameters such as a support
amplification factor using numerical simulations and condition closer to fixed-fixed than simply supported
available experimental data is described. It is shown conditions, a critical traffic scenario inducing less
that certain bridges are not susceptible to high levels dynamics, or a particular profile with an ideal distri-
of vehicle-bridge interaction when loaded by a crit- bution of irregularities within a given road class could
ically heavy vehicle or a critical combination of clearly lead to smaller dynamic allowances.
vehicles. Discussions on a range of numerical stud-
ies into the site-specific level of total load effect
(dynamic + static) and corresponding allowance for REFERENCE
dynamics of typical medium span highway bridges are
provided. ARCHES Programme. Assessment and rehabilitation of cen-
A procedure to experimentally obtain a site-specific tral European highway structures,WP2: structural assess-
dynamic allowance is justified numerically and then ment and monitoring. EU 6th framework, 20062009.
tested experimentally. Initially the approach is applied http://arches.fehrl.org/. Accessed October 2, 2009.
658
Influence of steel reinforcement corrosion on the stiffness of simply

supported concrete beams
F.J. OFlaherty, P.S. Mangat & P. Lambert

Centre for Infrastructure Management, , Sheffield, UK
E.H. Browne
Halcrow Group Limited, Birmingham, UK
ABSTRACT the stiffness of the beam would have to decrease 40%

before it exceeds serviceability limits. Therefore the
The in-service performance of reinforced concrete main steel corrosion would have to exceed 14% before
beams can be severely affected through corrosion of deflection criteria becomes an issue.
the steel reinforcement when it becomes subjected to The main conclusions from the results reported in
harsh corrosive environments containing chlorides and this paper are as follows and apply within the limit of
carbon dioxide. In such instances, corrosion is likely the parameters covered by the test data in the paper.
to occur in the steel reinforcement, with the expansive
reinforced concrete beams show a loss in stiffness
nature of the corrosion products likely to induce crack-
with increasing corrosion of the main and shear steel
ing and spalling of the concrete. A loss of structural
reinforcement;
integrity (stiffness) will occur and this can severely
an equation linking the decrease in stiffness to main
influence the serviceability of the member.
steel corrosion was determined as follows:
The purpose of this paper is to investigate the
% decrease in stiffness = 2.83 degree of main
relationship between degree of corrosion and loss of
steel corrosion
stiffness in corrosion damaged under-reinforced con-
main steel corrosion would have to exceed 14%
crete beams. Beams (100mm x 150mm cross section)
before deflection criteria are exceeded.
were subjected to accelerated corrosion in the labora-
tory and subsequently tested in flexure to failure. The
paper reports on the results of these tests and relates the REFERENCES
degree of corrosion in the main steel to the percentage
loss in stiffness in the concrete beams. Al-Sulaimani, G. et al. 1990. Influence of Corrosion and
A total of twenty reinforced concrete beams were Cracking on the Bond Behavior and Strength of Rein-
tested in flexure.The results show that stiffness reduces forced Concrete Members. ACI Structural Journal 87 (2):
with increasing main steel corrosion. Referring to 220231.
Figure A, a linear correlation is evident with R2 = 0.6. British Standard Institution 1997. Structural use of concrete:
The decrease in stiffness is equal to 2.83 times the Code of practice for, design and construction BS 8110-1.
Davies, H. 1996. Repair and maintenance of corrosion
degree of corrosion of the main steel. This means that
damaged concrete: Materials strategies, Concrete Repair,
Rehabilitation and Protection. London: E &FN Spon.
Mangat, P.S., Elgarf, M.S. 1999. Flexural strength of con-
crete beams with corroding reinforcement, ACI Structural
Journal 96 (1): 149158 JanFeb 1999
Mangat, P.S., OFlaherty, F.J. 1999. Long-term performance
of high stiffness repairs in highway structures, Magazine
of Concrete Research 51 (5): 325339.
Mangat, P.S., OFlaherty, F.J. 2000. Influence of elastic mod-
ulus on stress redistribution and cracking in repair patches.
Cement and Concrete Research 30: 125136.
Figure A. Relationship between % loss in stiffness and main

reinforcement corrosion.
659
Influence of shear reinforcement corrosion on the performance

of under-reinforced concrete beams
F.J. OFlaherty, P.S. Mangat & P. Lambert

Centre for Infrastructure Management, , Sheffield, S1 1WB, UK
E.H. Browne
Halcrow Group Limited, Birmingham, B16 8PE, UK
ABSTRACT
Reinforced concrete beams are normally designed as

under-reinforced to provide ductile behaviour at fail-
ure i.e. the tensile moment of resistance, Mt(0) , is less
than the moment of resistance of the compressive
zone, Mc . Since it is well established that the steel in
reinforced concrete beams is prone to corrosion, the
residual flexural strength is normally the main con-
cern of asset managers. However, concrete cover to the
shear reinforcement is less than that to the main steel
and therefore, may suffer higher levels of corrosion
due to chloride penetration and carbonation.
The paper investigates the influence of shear rein- Figure 1. Relationship between Mt(ShearCorr) /Mc and the
percentage of shear reinforcement corrosion.
forcement corrosion on the performance of rein-
forced concrete beams. Two groups of eight beams
tests (>0.91). Shear failure was evident only at higher
(100mm 150mm cross section), with two vary-
degrees of shear reinforcement corrosion for Type 1A
ing degrees of under-reinforcement (Mt(0) /Mc ratios)
(>18.7%, shown highlighted in Figure A).
at zero corrosion (control), were tested in flexure.
The main conclusions from the results reported in
The results show that despite exhibiting varying lev-
this paper are as follows:
els of shear reinforcement corrosion (the main steel
remained uncorroded throughout), flexure was still The predominant failure mode for under-reinforced
the dominant mode of failure. However, all beams did concrete beams exhibiting low degrees of shear rein-
exhibit a decrease in flexural strength with increasing forcement corrosion is flexural (<18.7% in this
shear reinforcement corrosion levels indicating that investigation)
the flexural integrity of the beam was influenced by Shear failure occurred only at higher degrees of
the shear reinforcement corrosion. This was more pro- shear reinforcement corrosion (>18.7% in this
nounced for the beams with a higher Mt(0) /Mc ratio investigation)
(lower degree of under-reinforcement) and this should Beams with lower Mt(0) /Mc ratios suffered a lower
be taken into account at the design stage. rate of flexural strength loss when corrosion was
The relationships in Figure A generally show a present in the shear reinforcement. Therefore, the
linear decrease in Mt(ShearCorr) /Mc with increasing per- recommendation is to design beams with lower
centages of shear reinforcement corrosion. The best fit Mt(0) /Mc for enhanced residual flexural strength but
linear equation for each series of beams is tabulated the shear capacity should be increased to guard
along with the coefficient of correlation (R2 ). A very against the risk of sudden shear failure at higher
satisfactory coefficient of correlation exists for both levels of shear reinforcement corrosion
660
Fatigue evaluation of steel finger type expansion joints for highway bridges
S. Ono
Japan Construction Method and Machinery Research Institute, Fuji-shi, Shizuoka-ken, Japan
S. Sakai & T. Imamura

Nippon Expressway Research Institute Co., Ltd., Machida-shi, Tokyo, Japan
ABSTRACT
A number of fatigue cracks have been found at steel

finger type expansion joints of Tomei-Expressway in
Japan, recently. Most severe damage was that the face
plate of the joint had failed and dismounted from the
actual expansion joint as shown in Fig. 1. Fatigue dura-
bility assessment is not well evaluated for this kind
of joints, although they are subjected to wheel loads
directly. Steel finger type expansion joints are con-
sisted of several steel partsface plate, web plates, rib
plates and anchor plates, etc. It is important to evaluate
the fatigue strength and fracture mechanism closely for
the suitable maintenance. In this study, the observation
Figure 2. Establish of the fracture mechanism.
of the joints dismounted from existing bridges has been

conducted first. The results have confirmed that the
cause of damage was the fatigue induced by repetitive
wheel loads. Then, the FEM analyses and the fatigue
tests with actual size specimens have been performed.
The failure process, the fatigue life and the fatigue
behavior of this type of the expansion joints have been
established finally. The outline of the obtained results
of this study for steel finger type expansion joints is
presented in this paper.
Fatigue behavior and characteristics of the steel
finger type expansion joints in highway have been eval-
uated by observations of the fractured expansion joints
and FEM analyses in this study.
Fatigue life to the fracture of face plate has been
estimated by fatigue test for Tomei-Expressway.
The results of this study are summarized as follows.
(1) The fracture of the joints is due to fatigue of applied
repetitive wheel load.
(2) The fracture procedure is verified following order
anchor plate, rib plate, web plate and face plate as
shown in Fig. 2.
(3) The mechanism of face plate fracture is the both-
ends-fixed beam bending after the web plate
fractured.
(4) The fatigue life of the face plate is predicted
approximately 3.2 years.
(5) It is necessary to inspect the joints for fatigue
Figure 1. Damage case of finger type joint. damage at least every 3 years.
661
Design of elastic foundation layers for buildings new design

and calculation methods design examples
M. sterreicher
iC-group, Department for Dynamics and Acoustics, Vienna, Austria
W. Unterberger
iC-group, Head of Department for Dynamics and Acoustics, Vienna, Austria
Due to increasing requirements on developed sites the elastic layer can not be achieved with the classical
with good infrastructure, the buildings move more SDOF System calculation.
often near to existing traffic infrastructure. Addition- An other important point of interest represents the
ally capital spending in revitalization or renewal of material used for elastic foundation layers. Generally
existing traffic nodes (i.e. urban railway stations) leads all of the existing materials come with non-linear mate-
to new developed sites near existing traffic infrastruc- rial properties. The bedding modulus strongly depends
ture. On the other hand the requirements concerning on the load (pressure) on the material. In case of
immission levels (vibrations and re-radiated noise) are the classical approach a single bedding modulus is
increasing rapidly. This leads to the fact that, in order assumed for the elastic layer. For buildings where
to fulfil these increased requirements, elastic founda- the loads on the foundation varies considerable the
tion layers for buildings are the only way to achieve classical approach leads to a wrong estimation of the
these requirements. bedding modulus and therefore to a wrong estimation
Therefore new innovative solutions on designing of the system natural frequency. The system natural
are required to fulfil on the one hand the technical frequency is however one of the key values for the
demands and on the other hand the economical require- immissions expected inside the buidling
ments for such projects. Traditionally the design of Thus to meet the immission requirements together
elastic foundation layers is based on the theory of the with economical requirements new approaches com-
SDOF (Single degree of freedom) system. For this the posed of improved FE-calculation considering the
building itself is assumed as a rigid complex. This is non-linear material properties are necessary. Addi-
generally more or less valid for classical stiff buildings tionally special solutions like expansion joints piping
buildings of conrete but leads to an overestimate of the are crossing the elastic foundation layers must be
loads applied to the elastic foundation layer for mod- considered.
ern buildings with columns bearing the load instead The present work shows this new calculation and
of walls. As a consequence the use of the classical design approach theoretically and on an example. The
SDOF System for such design calculations leads to used example comes from a building site on a new
wrong estimations for the loads (pressure) on the elas- developed area, near a railway tunnel in the city of
tic layer. Thus the demand for an economical design of vienna.
662
An attempt of rationalization for maintenance of railway structures using

management supporting system
M. Oyado & M. Miyashita

Railway Technical Research Institute, Tokyo, Japan
S. Ueda
Hankyu Corporation, Osaka, Japan
A. Sakairi
Odakyu Electric Railway Corporation, Tokyo, Japan
ABSTRACT
To utilize the existing civil-engineering infrastruc-

tures, steady maintenance activities of the structures
are quite important. The railway companies in Japan
are doing Regular Inspections for their own struc-
tures by means of the visual survey method periodi-
cally. Because these railway companies in the urban
areas have a number of structures in spite of their lim-
ited manpower, it is very hard for them to keep their
structures safe for a long time. Actually, many inspec-
tion data were not well organized, stored and utilized
depending on the situation. Additionally, highlights of
technical viewpoints or rules of judgments by plural
inspectors were not well unified. Figure 1. Hardware composition of SMS.
In order to bridge the gap between the ideal sta-
tus and the actual one, the information technology
has been applied to build up the Structural Manage-
ment Supporting System (SMS) by the collaboration
between 14 railway organizations in the urban areas
and a technical research institute. Figure 1 shows the
hardware composition of the SMS, and Figure 2 shows
the function of drawing the reference chart of the
graphical image of the deterioration.
This system has two major technical features; one
is that the system has a quite flexible database which is
suitable for each individual management framework of
14 railway organizations. The other is that the system
has a function to input the grounds of the decisions
and output the guide indices of the soundness ranks
based on the Maintenance Standards for railway struc-
tures in Japan. These features can serve to record and Figure 2. An example of reference chart of deterioration
archive the inspection results, and to make mainte- (Translated in English).
nance of structures efficient and rational. On the other
hand, the negative effect of elongation of the working
hours at the inspection sites was produced by using the REFERENCE
on-site inspection terminal of this system. Railway Technical Research Institute 2007. Maintenance
This paper describes outlines and backgrounds of Standards for Railway Structures and Commentary, Edi-
this system in order to give an outlook of the main- torially Supervised by Railway Bureau of the Ministry of
tenance methodology for the railway structures espe- Land, Infrastructure and Transport Government of Japan,
cially focused on the concrete bridges and viaducts. 210.
663
Sensitivity of bridge reliability to parameter variation in systems susceptible

to spatially distributed soil liquefaction
J.E. Padgett, J. Ghosh & L. Dueas-Osorio

Department of Civil and Environmental Engineering, Rice University, Houston, Texas
ABSTRACT
The complex characterization and response of cou-

pled soil-bridge-foundation systems in the face of
seismically induced liquefaction poses unique chal-
lenges when assessing the reliability of critical bridge
components. Uncertainty in modeling parameters for
bridges susceptible to spatially distributed soil liq-
uefaction range from structural modeling parameters
such as material and geometric properties, to soil foun-
dation parameters, such as soil stratigraphy or strength
parameters. However, there is a lack of understand-
ing of the relative importance of different variable
parameters on the reliability of bridges underlain by Figure 1. Column fragility curves at the complete damage
liquefiable soils, and a range of levels of fidelity state from the 32 runs of the sensitivity study.
of uncertainty treatment adopted in current bridge
fragility modeling practice.
The behavior of a representative multi-span con- Table 1. Summary of sensitivity study for median value
tinuous steel bridge (MSCS) typical of the central- fragility ranking the modeling parameter signifi-cance across
eastern U.S. is probabilistically assessed when all components and damage states.
underlain by liquefiable soils. MSCS bridges are
among the most vulnerable bridge classes owing to Rank of Most Significant Parameters
their bearing and abutment inability to accommo-
Undrained shear strength
date excessive demands, further exacerbated by soil Damping ratio
liquefaction. This study tests the sensitivity of con- Shear modulus
ditional reliability (seismic fragility) estimates for Deck-abutment gap
critical bridge components to variation in liquefiable Ultimate capacity of soil
soil, foundation, and bridge modeling parameters. Expansion bearing COF
A two-level fractional factorial design is adopted Fixed bearing COF*
for the sensitivity study, which requires 32 runs with Friction angle
various combinations of high and low levels of the
parameters. In total 13 factors, or modeling param- *Equally as important as preceding parameter.
eters, are considered in the sensitivity study, yielding
32 samples of the bridge-soil-foundation (BSF) model (median value). Table 1 summarizes the results of the
with various permutations of the potentially uncertain analysis, which found that the significance of parame-
parameters. Each sample is used in a full fragility anal- ters remained consistent across the four damage states.
ysis rooted in nonlinear dynamic analysis to evaluate The top seven modeling parameters tend to have a sta-
the probability of limit state exceedance. The critical tistically significant impact on the sensitivity of the
components considered in the sensitivity study include most vulnerable bridge components, as well as other
the fixed bearings, expansion bearings, columns, and components in the bridge. Modeling parameters that
abutments in active action (tension) and passive action do not tend to have a significant impact on any of the
(compression) at four damage states. The resulting log- highly vulnerable bridge components considered in
normal fragility curves vary considerably with each this study include fixed bearing initial stiffness, yield
combination of high and low modeling parameters as strength of steel, soil contraction parameter, and thick-
shown in Figure 1. An analysis of variance is con- ness of liquefiable soil layer. However, the fragility of
ducted to assess the statistical significance of modeling the piles and the bents and abutments are not tested in
parameter variation on the resulting fragility estimates this analysis and will be a focus of future studies.
664
An enhanced XML Schema matching technique for checking of the missing

items in construction document delivery
S.I. Park, B.-G. Kim, H.J. An & S.-H. Lee

School of Civil and Environmental Engineering, Yonsei University, Seoul, Korea
ABSTRACT
An enhanced method for XML Schema matching is

provided for developing a tool checking information
items during a process of electronic document deliv-
ery. This research has adopted an XML Schema match-
ing technique developed by Lee et al. (2006) because
most construction documents are in an unstructured
format. Basic process of the schema matching pro-
cess follows a work done by Yi et al. (2005), and the
matching process consists of similarity measurement
and relaxation labeling process. Main drawback of
previous method is that the computing time for relax- Figure 1. Accuracies according to change of NE ratio in
ation labeling dramatically increases as the number of simplified module.
document element increases because 4th order matrix
should be calculated during the relaxation labeling
process. In order to overcome this drawback, this
research proposed a simplified compatibility model for
the relaxation labeling process.
Distribution of accuracies according to variation
of weight values used in similarity measurement was
analyzed to evaluate accuracy when the new simpli-
fied compatibility model was applied. As shown in
Figure 1, the maximum mean value of accuracy for
the proposed method achieved about 93.61% where
the ratio of weight values on element name, sibling,
child and parent was NE : S : C : P = 1 : 1 : 1 : 1/2.
When the previous method was applied, the maximum
mean value of accuracy was 93.62%.
Whilst the accuracy of new method achieved
as much as the accuracy of previous method, Figure 2. Comparison of computing time due to number of
computing time was dramatically reduced. The sim- element of model.
plified compatibility model could reduce the comput-
ing time of relaxation labeling process to 460 times REFERENCES
where the number of total elements was 218 (see
Figure 2). Lee, S.-H., Kim, B.-K., Kim, D.-H. and Jeong, Y.-S. 2006.
Validation of the proposed method also performed Development of standardized semantic model for struc-
tural calculation documents of bridges and XML Schema
with twenty of sample schemas that were extracted
matching technique. Proceedings of the Third Interna-
from structural calculation document of steel girder tional Conference on Bridge Maintenance Safety and
bridge. The total numbers of document elements were Management (IABMAS): 633634.
29,525. Final accuracy of the proposed method was Yi, S., Huang, B. and Chan, W.T. 2005. XML application
91.0%, and it took 118 seconds per document in the schema matching using similarity measure and relaxation
relaxation labeling process. labeling. Information Sciences 169: 2746.
665
Structural analysis and fatigue life assessment of the Paderno Arch Bridge
A. Pipinato
Department of Construction and Transportation, University of Padova, Italy
ABSTRACT is presented. Critical regions of hot-spot members

were identified using structural finite element anal-
According to a study conducted by the ASCE Commit- ysis, and remaining fatigue life estimation has been
tee on Fatigue and Fracture Reliability (1982), 8090% performed. In this paper the Paderno steel bridge, a
of failures in steel structures are related to fatigue and typical arched railway of the mid-nineteenth century,
fracture. A follow up paper of this study is reported is studied according to a step-level assessment proce-
in Byers et al. (1997). At the same time, iron his- dure proposed in Pipinato (2008). First the bridge is
torical bridges represent a relevant category of the geometrically described and a literature material inves-
international cultural heritage, being the evidence of tigation is carried out. Then, a linear FEM model is
the modern industrial technology, particularly those used to find out critical hot spot stress. Finally hot spot
intended to accommodate activities of an industrial stress data are used in order to perform the reliability
or transport infrastructure. Many of these structures fatigue assessment.
require particular rehabilitation due to design defects,
basic elements deterioration, variation of use or change
of the intensity of the imposed loads. With regard to REFERENCES
Italy, the historical heritage is rich of significant metal
structures, which played an essential role in the growth ASCE (1982), Committee on Fatigue and Fracture Reliability
of industrial civilization: the most part of this heritage of the Committee on Structural Safety and Reliabil-
is represented by bridges, and the 60 per cent of Italian ity of the Structural Division. Fatigue reliability 14.
railway steel bridges has about one hundred years, as Journal of Structural Division, Proceedings of ASCE
they were built between 1900th1920th. 1982;108(ST1): 388.
The Paderno bridge, an arch metal historical struc- Byers, W.G., Marley, M.J., Mohammadi, J., Nielsen, R.J.,
Sarkani, S. 1997. Fatigue reliability reassessment appli-
ture is part of the Monza-Bergamo line and is in cations: state-of-the-art paper, Journal of structural engi-
service from 1889. A stepwise and practical approach neering New York, N.Y., v 123, n 3, p 277285.
for evaluating the structural integrity of historical Pipinato A. (2008), High-cycle fatigue behavior of historical
and deteriorated steel bridges, incorporating ana- metal riveted railway bridges, Ph.D Thesis, University of
lytical, mechanical and structural characterizations, Padova, Italy.
666
New high endurance sliding material for bridge bearings
V. Quaglini, C. Tavecchio, P. Dubini, D. Cuminetti & D. Ferroni

Department of Structural Engineering, Politecnico di Milano, Milano, Italy
ABSTRACT
PTFEsteel sliding bearings are among the most pop-

ular bearings currently used in bridge construction
because of their ability to support heavy loads and
accommodate large movements and rotations.
Thanks to its low friction and good resistance to
environmental influences, since its introduction in
the 60s PTFE has represented the state of the art
low-friction material in sliding bearings, despite of
its moderate load carrying capacity and low wear
endurance, which cause the PTFE element to be the
dimensioning and most stressed part of the bearing.
However in the recent years developments in bridge Figure 1. Friction profiles of dimpled and lubricated slid-
design and increasing traffic conditions have posed ing elements using either Xlide or PTFE at different
new and more severe requirements to sliding bearings. temperatures and accumulated sliding distances.
In very long span bridges and in structures subjected
to very frequent loads the expected movements at the
supports during the service life of the structure are covered by PTFE elements under similar conditions
expected to exceed by far the wear life of PTFE, and (Fig. 1);
PTFE bearings should be replaced several times dur- 3 Xlide offers minimal resistance to sliding even
ing the lifetime of the structure, resulting in excessive at very low temperatures, out of the range of
costs and an extremely high maintenance burden. Fur- application of PTFE;
ther, the coefficient of friction of PTFE increases at low 4 the dynamic coefficient of friction of Xlide holds
temperatures, and the European standard EN 1337 pre- constant over the typical pressure and velocity
scribes a minimum operating temperature of 35 C ranges of bridge bearings;
for PTFE sliding bearings, so excluding the use of 5 Xlide has a good resistance to chemical and
these bearings from regions with continuously low environmental influences.
temperature characteristics.
To overcome these limitations, a special sliding In conclusion, Xlide is suitable to replace PTFE in
material has been developed at Politecnico di Milano sliding bearings subjected to high loads, long distances
and proposed as an alternative to PTFE in sliding bear- of movements during the service life of the structure,
ings. The trade name of the special sliding material is and very low environmental temperatures. Currently
Xlide . An extensive experimental campaign has been use of Xlide in sliding spherical bearings is covered
carried out to assess its functional properties, and the by an European Technical Approval.
main results are summarized as follows:
1 the compressive strength of Xlide at ambient tem-
perature is 190 MPa, which is two times larger than REFERENCES
PTFEs;
Eggert, H. and Kauschke, W., 2002. Structural Bearings.
2 sliding elements made of Xlide plates lubricated Ernst & Sohn, Berlin.
with silicon grease can sustain, with negligible EN 1337-2:2004 Structural bearing. Part 2: sliding elements.
wear, accumulated movements up to a total distance EN 1337-7:2004 Structural bearings Part 7: Spherical and
of 50,000 m which is five times larger the distance cylindrical PTFE bearings.
667
What is bridge durability official regulations and reality
W. Radomski
Warsaw University of Technology, Warsaw, Poland
ABSTRACT traditional methods do not allow to ensure the bridge

durability demanded by the bridge administration
Bridge durability has became during the last decades because they are based on the quite different philos-
one of the most important technical, economical and ophy of design than durability design. The durability
social problems in many countries including the most design is the newest trend and allows to harmonize the
developed ones. It results mainly from the fact that methodology of design with the required durability of
the durability of a large number of the existing bridge the bridge structures with its technical aspect. The fun-
structures has shown to be not enough and they show damentals of the above design method are presented in
to be structurally deficient and, therefore, require to the paper and exemplified by the case of carbonation
be rehabilitated. The durability is multi-component of concrete and corrosion of the steel reinforcement as
problem concerning both the material and structural the factors influencing the bridge durability.
solutions of the bridge structures as well as their main-
tenance. Moreover, it concerns the adequate methods
of design and erection of the new bridges as well as REFERENCES
the rehabilitation of the existing ones. In many cases
the existing bridges are also functionally obsolete and Directive of the Minister of Transport and Maritime Economy
2000. Conditions for highway bridges. Dziennik ustaw
demand to be modernized.Therefore, bridge durability
Rzeczypospolitej Polskiej, 63, 38413932 (in Polish).
can referred also to the modernized structures. ISO 13823:2008. General principles on the design of struc-
According to the authors concept illustrated in tures for durability.
Figure 1, the bridge durability can be defined as Nowak, A.S. & Collins, K.R. 2000. Reliability of structures.
the period of time when the bridge (or its structural Boston: Mc Graw Hill.
member or detail) achieves certain acceptable level Radomski, W. 2004. Old existing bridges to rehabilitate or
of decreasing its required technical and/or functional to replace them? In Eiichi Watanabe, Dan M. Frangopol
characteristics, such as e.g., load-carrying capacity, & Tomoaki Utsunomiya (ed.), Bridge maintenance, safety,
stiffness, crack widths, too limited vertical clearance management and cost; Proc. intern. symp., Kyoto, 1822
October 2004. Leiden: Balkema.
under the structure or too small width of the bridge
Report of RILEM Technical Committee 130-CSL 2005. Asko
deck, etc. Sarja & Erkki Vesikari (ed.), Durability design of concrete
Officially required bridge durability (i.e., their ser- structures. London: F & FN Spon.
vice life) in the relevant regulations does not result Ryzynski, A. 2005. Technical and functional durability of
from the bridge design methods applied so far. These highway bridges and its prediction. Drogownictwo 78,
195198 (in Polish).
Figure 1. Bridge characteristics in a function of time.
668
Effective management of concrete assets concrete durability: Achievement

and enhancement
P.C. Robery
Halcrow Group Limited
ABSTRACT effectively in this field. Equally, asset management

practitioners need assistance from concrete specialists
Effective management of concrete assets is noabout with to predict future deterioration rates for con-
delivering best value return on investment. The safety crete assets, identifying time-to-spalling, reductions
of highway users places an additional burden on bridge in structural capacity with time if they are left unre-
asset managers, who must ensure their assets remain paired, and ultimately, the point in time when the asset
safe, predictable and resilient to changing materials will be cheaper to demolish and replace, rather than
and environments. retain and repair, in whole life value terms.
To achieve safe and predictable bridge assets, own- This paper discusses some of the issues identified
ers and operators need a detailed asset management when considering bridge asset management, in a cli-
strategy, encompassing a database inventory, inspec- mate where owners are trying to get longer lives and
tion reports operation and maintenance plans, a val- reduced maintenance for concrete assets. The paper
uation for each asset and an understanding of the focuses on bridge asset valuation in particular and its
level of service expected. Many industries are already relationship with whole life valued costing of refur-
highly familiar with these concepts, including Avia- bishment options for bridge assets. Examples are given
tion, Water, Rail and the Military. In the UK, even the for cost effective management of bridges.
humble car parks now need an asset management plan
(the Life Care Plan), following the high profile car park
collapse in Wolverhampton. REFERENCES
The Bridges and Highways sector has lagged behind
BS EN 1504: Parts 1 to 10, Products and systems
other industries in the implementation of a full asset for the protection and repair of concrete structures
management strategy. In the UK, bridges are inspected Definitions, requirements, quality control and evaluation
regularly and inspection information is stored on a of conformity, BSI.
central database, along with notional asset value infor- BS EN 206-1: 2000, Concrete. Specification, performance,
mation. This is moving towards the concept of a full production and conformity, BSI, 2000.
depreciated asset valuation, based on actual condition, BS CP 110:1970: The structural use of concrete, BSI, 1970
but it is a long way off. (withdrawn).
The situation for the local roads and bridges is far BS EN 1992-1-1, Eurocode 2: Design of concrete
structures Part 1: General rules and rules for build-
less consistent or as well advanced. Local Govern-
ings, BSI.
ments (run by County, District or City Councils) have Robery, PC, Managing bridges and other highway assets,
been forced to comply with UK Government require- Assets the Magazine of the Institute of Asset Manage-
ments to carry out asset management and valuation ment, July 2007.
to all of their assets. Many Local Government groups Alisa, M., Robery, P., Hamer, B., Tamburic, Z., & Harding,
are well down the road to preparing Highways Asset N., Managing North Londons River Flood Defences -
Management Plans (HAMPs) but there remains a Part 1: focus on asset management, Concrete Journal,
wide gulf between knowing what bridge assets you V40, N3, pp44-45.
have and the depreciated value of that asset: a gulf that Institute of Chartered Accountants, Implementing
Turnbull a boardroom briefing, Institute of Chartered
can only be closed by expensive testing, assessment
Accountants Centre for Business Performance. September
and optioneering. 1999, 36p.
Those tasked with managing bridges, as well as Roads Liaison Group, Guidance Document for Highway
engineers and concrete technologists, need a work- Infrastructure Asset Valuation, County Surveyors Soci-
ing understanding of asset management plans, objec- ety/TAG Asset Management Working Group, July 2005,
tives and valuation principles if they are to engage 113p.
669
Condition inspection, analysis and maintenance of three prominent

railroad bridges
C. Rolwood
Hardesty & Hanover, LLP, Bear Tavern Road, West Trenton, New Jersey
R. Drew
Hardesty & Hanover, LLP, Broadway, New York
ABSTRACT
During 2005 and 2006, Hardesty & Hanover per-

formed bridge condition inspection and load rating
analysis of three prominent railroad bridges. The first
bridge is Amtraks Little Hell Gate Bridge in New
York City. The Little Hell Gate Bridge, completed in
1917, is a three-track, 1,156-foot long, unique under-
deck inverted bow string steel truss structure with
pins and eyebars, soaring high above the urban land-
scape of New York, designed by the renowned Gustav
Lindenthal and Othmar Ammann. The second and
third bridges are the Tunkhannock and Martins Creek
Viaducts near Scranton, Pennsylvania, inspected for Figure 1. End Span of Amtrak Little Hell Gate Bridge.
Canadian Pacific Railway. The viaducts were con-
structed circa 1915 and are massive, world-record,
reinforced concrete structures of 2,375 feet and 1,600
feet length respectively, that carry a single-track rail-
road as high as 240 feet above rural, forested mountain
valleys of northeastern Pennsylvania. Each bridge
posed challenges to the engineers conducting the
inspection, as there were unique access requirements,
specialty testing, operating railroad accommodation,
and issues of special interest involved with each. This
presentation provides information on field inspection
access, specialty testing, and recommended main-
tenance programs on these types of bridges. This
presentation will describe the inspection plans used to
thoroughly and efficiently conduct these bridge inspec-
tions, the specialty testing used on each, and the safety Figure 2. Canadian Pacific Railway Tunkhannock Viaduct.
procedures implemented to mitigate the higher risks
faced on these bridges. This presentation will also
describe the observed maintenance issues and rec-
ommendations to owners/managers to enhance their
maintenance plans in order to increase the lifespan of
their structures and minimize future rehabilitation and
replacement costs.
670
Full scale laboratory testing of replacement orthotropic deck for Verrazano

Narrows Bridge
S. Roy, J.W. Fisher & R.S. Deo Alapati

ATLSS Engineering Research Center, Lehigh University, Bethlehem, PA, United States
ABSTRACT
After a comprehensive study of the site specific load-

ing, it was recommended to replace the 45 year old
concrete filled steel grid deck at the upper level of
the Verrazano Narrows Bridge with a steel orthotropic
deck that is integral with the floor system and the stiff-
ening truss. The new lightweight deck system with
a thin wearing surface and concrete barrier demon-
strated better structural efficiency and life cycle perfor-
mance. The replacement deck designed by the Parsons
Transportation Group of New York City consists of
closed trapezoidal ribs. To accommodate the limita-
tions on vertical clearance, the deck panel incorporates
a shallow sub-floor beam (diaphragm) with a cutout
at the junction of the pass through ribs. The ribs are
provided with a full depth internal bulkheads at the
sub-floor beam locations. One of the primary rea-
sons for selecting an orthotropic deck is that the deck
system, if properly designed, can provide more than
100 years service life with minimum maintenance.
However, the major concern related to orthotropic
deck is the relatively high initial cost owing to inten-
sive fabrication, and higher possibility of in-service
fatigue cracking from a large number of welded con-
nections. Particularly severe of these details is the
bulkhead/sub-floor beam to rib connections that are
subjected to complex in-plane and out-of-plane defor-
mations under localized wheel loads. Other critical tandem axle bogie load moving across the deck at a
detail is the longitudinal rib-to-deck partial penetra- slow rate. The deck is instrumented using more than
tion weld that is known to develop fatigue cracking at 300 sensors at critical locations. In addition the deck
the lack of penetration at the weld root. performance is also monitored using acoustic emis-
Fatigue performance of the replacement orthotropic sion technique. A finite element model of the prototype
deck for the Verrazano Narrows Bridge is being evalu- deck is analyzed to understand the complex behav-
ated on a full scale prototype at theATLSS Engineering ior of the deck under localized and moving wheel
Research Center, Lehigh University. The prototype loads, to verify the measured response under static
deck is about a quarter of the deck between the panel and dynamic loading conditions and to correlate the
points of the stiffening truss, and includes one floor deck response with the observed cracking mode. The
beam and two stringers. The fatigue testing is being prototype testing and the analytical studies is provid-
conducted using six actuators loaded in series, simu- ing critical information on issues related to fabrication
lating the passage of tandem axles. In addition, global and deck installation, design improvement and opti-
displacement of the deck is simulated by three actua- mization, expected response of the orthotropic deck in
tors located under the floor beam and the two stringers. service, and the effectiveness of the replacement deck
The deck response is also determined under a rolling design in achieving more than 100 years service life.
671
Structural hardening for cable elements of cable supported bridges
N.D. Sauer
VSL, Dallas, TX, USA
ABSTRACT products and practices may be adapted to bridge

protection.
The necessity of ensuring transportation structures are Establishment of the proper performance tests and
protected from threats on all levels is an ongoing con- acceptance criteria that should be used to evaluate
cern for those involved in the design, construction, hardening alternatives against specific threats.
operation and maintenance of critical infrastructure. Recommendations for the procurement process,
Cable-supported bridges can be particularly vulnera- including the development of secure plans and spec-
ble to a variety of threats, including large blast and ifications, protection of the design threat(s) and
precision demolition explosives, as well as fire and hardening alternative composition.
a variety of cutting threats. With the proper analy-
sis of these threats and the execution of a thorough
planning process, these structures can be effectively
protected. This presentation will examine the details REFERENCES
involved with each step of this process, including:
Blue Ribbon Panel on Bridge and Tunnel Security. September
Objective evaluation of the vulnerable elements 2003. Recommendations for Bridge and Tunnel Secu-
against generally defined threats through a thorough rity, Report Prepared for The American Association of
risk analysis and detailed structural analysis of the State Highway and Transportation Officials (AASHTO)
bridge. Transportation Security Task Force.
Definition of the specific threat or threats the hard- Bulwa, D. & Fimrite, P. April 29, 2007. Tanker fire destroys
ening solution will be required to defeat to allow for part of MacArthur Maze. San Francisco Chronicle.
Wilson, B. March 2009. Safekeeping: Industry still trying to
objective testing and approval.
secure bridges, tunnels two years after BRP Report, Roads
A discussion of how to select the best structural
and Bridges.
hardening alternative for a particular structure based Zoli T. & Steinhouse, J. November 411, 2007. Some Consid-
on specific evaluation criteria such as price, perfor- erations in the Design of Long Span Bridges Against Pro-
mance, weight, durability, maintainability, and aes- gressive Collapse. Paper presented at The 23rd US-Japan
thetics. The presentation will examine how existing Bridge Engineering Workshop, Tsukuba, Japan.
672
Performance assessment of bridges under progressive

damage and abnormal actions
D. Saydam & D.M. Frangopol

ABSTRACT Alternative damage scenarios including slowly aris-

ing damage, sudden damage, damage occurring in all
This paper presents a framework for evaluation of members and damage occurring in critical members
bridge performance under progressive damage and are considered.
abnormal actions. Evaluation of bridge response under Time variant parameters in the analysis are the
damage requires the use of proper performance mea- resistance of each member and applied load on the
sures. Selected performance indicators are explained structure. Alternative resistance degradation rates and
and used in a numerical example of a truss bridge damage factors are considered while load increase rate
undergoing time-variant damage. Lifetime perfor- is kept constant in all analyses.
mance of the bridge is evaluated using both determin- Probabilistic lifetime profiles of redundancy and
istic and probabilistic approaches. In the probabilistic vulnerability of a truss bridge under different damage
approach, uncertainties are related to loading and scenarios are shown in Figure 1.
member strength. The first yielding probability, the
system failure probability, and the vulnerability and
damage tolerance profiles considering 75 years life- REFERENCES
time of the structure are obtained. The reserve strength
factor, residual strength factor, redundancy factor and Frangopol, D. M. & Curley, J. P. 1987. Effects of Dam-
age and Redundancy on Structural Reliability. Journal of
margin of redundancy profiles are also obtained.
Frangopol, D. M. & Klisinski, M. 1989a. Material Behavior
and Optimum Design of Structural Systems. Journal of
Frangopol, D. M. & Klisinski, M. 1989b. Weight-Strength-
Redundancy Interaction in Optimum Design of Three-
Dimensional Brittle-Ductile Trusses. Computers and
Structures 31(5):775787.
Frangopol, D. M., Klisinski, M., Iizuka M. 1991. Optimiza-
tion of Damage-Tolerant Structural Systems. Computers
and Structures 40(5):10851095.
Frangopol, D. M. & Klisinski, M. 1992. Design for Safety,
Serviceability and Damage Tolerability. Designing Con-
crete Structures for Serviceability and Safety. ACI Spe-
cial Publication, SP 13312, Detroit:American Concrete
Institute.
Frangopol, D. M., Klisinski, M., Lin K. Y. 1996. Incor-
porating Damage Control in Structural Design. Build-
ing an International Community of Structural Engineers
1:598605.
Lind, N. C. 1994. A Measure of Vulnerability and Dam-
age Tolerance. Reliability Engineering and System Safety
43(1):16.
Okasha, N. & Frangopol, D. M. 2010. Time-variant redun-
dancy of structural systems. Structure and Infrastructure
Engineering 6:279301.
Rashedi, M.R. & Moses, F. 1988. Identification of fail-
ure modes in system reliability. Journal of Structural
Figure 1. Bridge lifetime profiles for two probabilistic Engineering 114 (2):292313.
performance indicators: redundancy and vulnerability.
673
Detection of fatigue crack of steel deck plate by ultrasonic

nondestructive testing
H. Shirahata, K. Akasaka & T. Iizuka

Tokyo City University, Tokyo, Japan
ABSTRACT
Steel orthotropic deck systems have been applied in

many countries. However, in recent years, many cases
of fatigue cracking of the deck systems have been
reported in Japan. This study focuses on a rib-to-deck
fatigue crack that is initiated at the weld root and
propagated in the deck plate, as shown in Figure 1. Figure 2. Wave form obtained by SH wave technique, when
Considering inspection of cracks shown in Figure 1, the fatigue crack was 3.20 mm deep.
visual inspection is nearly impossible due to the acces-
sibility. The objective of this study was to investigate
the applicability of the ultrasonic testing to detect those
fatigue cracks. Fatigue tests were carried out. Dur-
ing the fatigue tests, three ultrasonic nondestructive
techniques were applied. Those were creeping wave,
horizontal shear (SH) wave, and surface (Rayleigh)
wave techniques. Detectability of each technique was
investigated.
The specimen was fabricated by welding two plates,
considering a deck plate and a rib plate. Cyclic load-
ing of 200,000 cycles was applied to the specimen.
After loading, the specimen was taken away from the
loading site, and the ultrasonic testing was carried out.
That procedure was repeated until failure. At each time
Figure 3. Echo height and crack depth in cases of (a)
of cyclic loading, the amplitude of the loading was
creeping wave, (b) SH wave, and (c) surface wave methods.
changed to make beach marks. Fatigue failure from
the weld root into deck plate occurred after more than
surface wave techniques also showed a good perfor-
900,000 cycles.
mance. The ratios of detection for a crack of 2 mm or
Figure 2 shows one of the wave forms obtained by
longer were 80 and 75%, respectively.
the ultrasonic testing. Relation between echo heights
and crack depths was obtained. Actual crack depth
could be obtained from the beach mark observation. REFERENCES
Figure 3 shows the echo height and crack depth plot.
The SH wave technique showed the best detection abil- Miki, C. 2006. Fatigue damage in orthotropic steel bridge
ity among three techniques. It was possible to detect decks and retrofit works, International Journal of Steel
crack longer than 2 mm completely. The creeping and Structure, 6(4) 255267.
Murakoshi, J., Arima, N., & Fujiki, O. 2004. Ultrasonic Mea-
surements of Fatigue Crack in Deck Plate of Orthotropic
Steel Deck, Civil Engineering Journal, 46(11) 3439. (in
Japanese).
Sim, H., Uang, C, & Sikorsky, C. 2009. Effects of Fabrication
Procedures on Fatigue Resistance of Welded Joints in Steel
Orthotropic Decks, Journal of Bridge Engineering, 14(5)
366373.
Figure 1. Fatigue crack focused in this study.
674
Fiber-reinforced polymer composites: An effective solution to our nations

crumbling infrastructure
M. Stafford
R.J. Watson, Inc., East Amherst, NY, USA
ABSTRACT International Code Council. The Transportation

Research Board has also published a useful guide
The poor condition of our nations bridges has been for design and inspection. AASHTO is developing a
well documented. According to the National Bridge design guideline as well. There will be a discussion on
Inventory, there are over 600,000 bridges in the United how FRP composites are designed.
States, and 25% of them are determined to be either Today, there are hundreds of installations of FRP
structurally deficient or functionally obsolete. composites that have been used to strengthen bridge
The age, environment, and increasing load require- columns, beams, pier caps, walls and slabs. Several
ments are factors contributing to the deterioration of case studies will be shown to emphasize the effec-
our nations infrastructure. tiveness of strengthening with FRP composites, and
For years, engineers have been looking for a solu- the cost savings for the owner. Many installations are
tion to the above issues, to repair and strengthen our over 10 years old. FRP Composites have proven to
bridges quickly, in a cost effective manner, and with be very effective in strengthening structural members,
minimal disruption to traffic. One of the most tested and extending the life of our nations bridges.
products over the past 20 years is Fiber Reinforced
Polymer (FRP) Composites. FRPs are one of the tolls
that will help us to catch up to the maintenance needs
of thousands of our bridges. REFERENCES
Research and testing of FRP composites began in
the 1980s to strengthen structural elements to with- Homam, S.M., Sheikh, S.A., Pernica, G., and Mukherjee,
stand the forces associated with earthquakes, and P.K., Durability of Fibre Reinforced Polymers (FRP)
Used in Concrete Structures, Research Report HS-01-
strengthening of structural elements that are either
00, Dept. Of Civil Engineering, University of Toronto
deteriorated, under-designed or damaged from impact. (2000).
The long term durability of FRP composites will also Williams, D., Strengthening and Rehabilitation of Highway
be discussed in greater detail. Structures with FRP Composites in West Virginia, West
There are independent design guidelines pub- Virginia Dept. of Highways, Presented at FRP Workshop,
lished by the American Concrete Institute, and the West Virginia University, October 7, 2009.
675
Seismic assessment on the existing highway bridges in Taiwan
Y.-C. Sung
National Taipei University of Technology, Taipei, Taiwan
K.-C. Chang
National Taiwan University, National Center for Research of Earthquake Engineering, Taipei, Taiwan
D.-W. Chang, C.-L. Chen & H.-M. Wu
Taiwan CECI Consultant Inc., Taipei, Taiwan
K.-Y. Liu
National Center for Research of Earthquake Engineering, Taipei, Taiwan
ABSTRACT
The current engineering technology as well as

economy-benefit policy to enhance the seismic capac-
ity of the bridges in Taiwan is introduced in this paper.
A plan of seismic retrofitting for more than 2,200
highway bridges all over Taiwan performed through
2006 to 2008 is the focus to present and discuss. Fif-
teen types of bridge in which were categorized in
advance according to their different seismic responses, Figure 2. Pushover curve in the longitudinal direction.
based on the lessons learnt from the experienced catas- Figure 3. PGA assessment result in the longitudinal
direction.
trophic Figure 1. Analytical bridge model used in
SAP-2000 earthquakes. Some representative bridges
reasonable results. The case study of the Li-Kun bridge
in each type were preceded with either pushover
shows the efficiency of applying the retrofitting mea-
analysis or time history analysis to get the seismic
sure to improve the seismic behavior in terms of the
capacity. Accordingly, the seismic fragility curves for
return period of the earthquake or anticipated service
each group can be established statistically. The feasi-
life. Analytical result shows that the Li-Kun bridge can
ble retrofitting approaches and the cost required for
sustain an earthquake with PGA of 0.488 g and can
the grouped bridges with insufficient seismic capac-
function effectively for more than 107 years after per-
ity were evaluated to serve as the basis of seismic
forming the seismic retrofitting work. Therefore, the
retrofitting for the numerous existing bridges. As a
goal of extending the structural longevity is achieved.
result, the possible resulting benefits of social economy
This paper is useful for engineers to define a higher
can be explored and discussed to support the optimal
performance objective when setting a realistic target
retrofitting strategy.
in the performance-based design approach.
A simplified and novel seismic assessment proce-
dure has been introduced. A state-of-the-art pushover
analysis method is adopted to calculate the seismic REFERENCES
resistance capacity of the bridge. A few modifications
have been made in the pushover analysis to obtain more ATC-40. 1996. Seismic Evaluation and Retrofit of Concrete
Building, Applied Technology Council, 1996.
China Engineering Consultants. Inc. 2000. 921 Chi-Chi
Earthquake Special Report, Taipei, CECI.
Computers and Structures, Inc. 2002. SAP2000, Integrated
Finite Element Analysis and Design of Structures, Analy-
sis Reference Manual, Version 8.1.2, Berkeley, California.
FHWA. 1995. Seismic Retrofitting Manual for Highway
Bridges, Federal Highway Administration, Publication
No: FHWA-RD-94-052, Virginia.
Ministry of Transportation and Communication. 1995. Seis-
mic Design Specification for Highway Bridges.
Sung, Yu-Chi, Liu, Kuang-Yen, Su, Chin-Kuo, Tsai, I-Chaio,
Chang, Kuo-Chun. 2005. A Study on Pushover Analyses
of Reinforced Concrete Columns, Journal of Structural
Figure 1. Analytical bridge model used in SAP-2000. Engineering and Mechanics, 21(1): 3552.
676
Prediction of restrained shrinkage stresses and premature deck

cracking in composite bridges
A. Szwed
Warsaw University of Technology, Warsaw, Poland
M.M. Szerszen
University of Nebraska, Lincoln, NE, USA
ABSTRACT
Reinforced concrete bridge deck slabs made com-

posite with supporting steel or pre-stressed concrete
girders are widely used structural systems in highway
bridges. Current practice in operation shows that ser-
vice life of a reinforced concrete deck is much shorter
than the design life of the bridge, Krauss & Rogalla
(1996). Observed deterioration scenarios indicate that
degradation process of bridge decks is very often initi-
ated by hydration heat and shrinkage cracks. Despite of
many modifications made in standards of bridge deck Figure 1. Relaxed stresses at the top and bottom fibers of
design and curing transverse cracks have continued to the cross-section in the deck of composite bridge.
appear, French et al. (1999), Hadidi & Saadeghvaziri
(2005).
Paper deals with the development of a simple and shrinking bridge deck is less than for steel girders,
practical method to estimate stresses due to restrained Saadeghvaziri & Hadidi (2005).
shrinkage. Proposed approach is based on flexure anal- A method to reduce early-age deck cracking is pro-
ysis of a composite beams section assuming slip at posed. A general idea behind the method is to initially
interface of the deck and girder as well as slip between create stress pattern in the slab of opposite sign to
concrete deck and reinforcement. Displacement and that caused by restrained shrinkage. It is proposed to
strain discontinuity condition at the interfaces, and apply externally controlled forces to the girders to cre-
curvature compatibility for the deck and girder of com- ate the initial camber. In that manner, tensile stresses
posite section are utilized. Linear elastic properties of will be induced at the top part of the girders. After
constituent materials and linear slip constitutive laws the first phase of concrete maturing, the camber will
are assumed. Obtained results indicate the strong pos- be gradually relaxed. A part of tension stress caused
sibility of cracks initiation in the deck slab due to by shrinkage will be compensated by the composite
restrained shrinkage development especially in early action of bridge cross-section. As an example of appli-
age of concrete. The bottom part of the deck section cation of compensating load simply supported beam
is mostly exposed to cracking. The ratio of deck-to- (pin-roller) of a bridge is analyzed, see Figure 1.
girder stiffness strongly affects the final value of the
restrained shrinkage stresses. The analytical expres-
REFERENCES
sion for the coefficient of restraint level is given in
closed form, Krauss H.C. & Rogalla E.A. Transverse Cracking in Newly
Constructed Bridge Decks. NCHRP Rep. 380,Transporta-
tion Research Board, NRC, Washington 1996.
French C., Eppers L., LE Q. & Hajjar J.F. Transverse Crack-
ing in Concrete Bridge Decks. Transportation Research
Record 1688, TRB, Washington 1999, pp. 2129.
Hadidi R. & Saadeghvaziri M.A. Transverse Cracking of
and can be easily used in design practice to measure Concrete Bridge Decks: State-of-the-Art. Journal of
cracking tendency of concrete in bridge decks. Bridge Engineering, Vol.10, 2005, pp. 503510.
The restraint level caused by the steel girders ranges Saadeghvaziri M.A. & Hadidi R. Transverse Cracking of
from 25% to 60% of the full restraint. In case of typi- Concrete Bridge Decks: Effects of Design Factors. Jour-
cal pre-stressed AASHTO girders the constraint to the nal of Bridge Engineering, Vol.10, 2005, pp. 511519.
677
Evaluation of corrosion deterioration of weathering steel bridge

under the environmental corrosiveness
Y. Tamaki, T. Shimozato, Y. Arizumi & T. Yabuki

Department of Engineering, Ryukyu University, Okinawa, Japan
S. Ono
Japan Construction Method and Machinery Research Institute, Japan Construction Mechanization
Association, Shizuoka, Japan
ABSTRACT In this study, we carried out evaluation of corrosion

deterioration for steel girder bridge using this weath-
Weathering steel bridge had been exposed for 28 years ering steel bridge under the real severe accelerated
in Okinawa Japan. Corrosion environment for steel corrosion environment.
material in Okinawa is a severe accelerated corro- In this report, the outline of investigation results
sion environment because of high humidity with high will be presented and discussed based on the decrease
temperature, and huge amount of the air-born salt for of the main girders web plate thickness and the
islands region. As for this bridge, sever corrosion was environmental corrosiveness.
happened the end of girders. Then it collapsed in July
2009.
Figure 2. Corrosion situation (inside and outside).
Photo 1. Panorama bridge (Okinawa in Japan).
Figure 1. Corrosion situation (the end of the girder). Figure 3. Corrosion of the center part and the end.
678
Study on the cause of brittle fracture during earthquakes in steel

bridge bent focusing on stress triaxiality
H. Tamura, E. Sasaki, H. Yamada & H. Katsuchi

Yokohama National University, Yokohama, Japan
ABSTRACT
During the Northridge Earthquake in 1994 and the

South Hyogo prefecture Earthquake in 1995, unex-
pected brittle fractures were observed in some steel
structures. A corner of steel bridge bent was also
cracked seriously.
Based on the multiaxiality, this paper reveals the
effect of some weld bead shape on the risk of brit-
tle fracture during earthquakes on real structure. The
multiaxiality was evaluated by stress triaxiality which
was computed through nonlinear FEM analyses apply-
ing an earthquake waveform. The target structure is Figure 1. Triaxiality distributins.
steel bridge bent which was damaged during the South
Hyogo prefecture Earthquake.
Stress triaxiality is often defined as
where h and are respectively the hydrostatic stress

and the von Mises stress. On the other hands, triaxi-
ality T2 is also defined by Schafer et al. (2000). The
Definition of T2 is as follows:
where max is the maximum principal stress. Figure 2. Maximum principal stress distributions.
T1 and T2 are indexes which have a similar tendency can be found the remarkable difference.This is because
for estimating the triaxiality of multiaxial stress state the effect of the triaxiality which is strongly affected by
(Tamura et al. 2009). the weld bead shape. It implies one of the main causes
In this study, FEM analyses considering geometri- of brittle fracture in the real bent was possibly the tri-
cal and material nonlinearities were conducted using axiality generated by a small radius toe (approximately
the analysis program Abaqus 6.7. In the analyses, a 1 mm) of unequal leg bead.
zooming method was applied to obtain the local stress Similarly, the effects of unequal leg bead and crack-
state of fracture origin during earthquakes. ing were also investigated. As a result, it was indicated
As one of the result of this study, the dependency that there is a possibility that fracture occurs due to
of triaxiality on the weld bead radius is shown in these weld bead shape effect.
Figure 1. The horizontal axis of the figure is the depth
from the weld toe in which the fracture of real bent
started. From the figure, the difference of triaxiality REFERENCES
distributions among four kinds of bead shape (the toe
Schafer, B. W., Ojdrovic, R. P. & Zarghamee, M. S. 2000.
radii are 0.5 mm, 1.0 mm, 2.0 mm, and 5.0 mm) can be
Triaxiality and Fracture of steel moment connections.
confirmed. The degree of triaxiality increase with the Journal of Structural Engineering, pp.11311139.
depth depends strongly on the toe radius. Tamura, H., Sasaki, E., Yamada, H. & Katsuchi, H. 2009.
The triaxiality increases the risk of brittle fracture, Involvements of stress triaxiality in the brittle fracture
because it increases the local normal stress of there. during earthquakes in steel bridge bents. International
Figure 2 shows the effects of triaxiality on the max- Journal of Steel Structures, Vol. 9, No. 3, September,
imum principal distributions at the fracture origin. It pp. 241252.
679
A simplified approach to calculate the secondary moments of continuous

prestressed concrete bridges
V. Tiruchirapalli & R.A. Hindi

Bradley University, Peoria, Illinois, USA
ABSTRACT
This paper provides practical engineers with an equa-

tion to calculate the secondary moments of prestressed
concrete continuous beams. Normally the prestressed
secondary moments are calculated by deriving it
from the net moment and the primary moments. This Figure 1. Two span girder showing span length and
involves extensive calculations and sometimes com- eccentricities.
plex integration techniques but the proposed equation
described in this paper will give the secondary moment 3. Moment distribution
directly. The secondary moment calculated using the 4. Stiffness method
equation is compared with the secondary moment cal- 5. STAAD
culated using various existing methods such as support
displacement method, moment distribution method To summarize the secondary moment of prestressed
and stiffness method, and also using the structural anal- concrete can be determined using the derived expres-
ysis software STAAD. For the purpose of analysis, sion so that the designers can use this equation.
two span, three span and four span beams are con- The proposed equation can be used to accurately
sidered. This equation can be applied to multiple span calculate the secondary moment using the eccen-
beams/girders. Finally the equation is used to show the tricities and span length.
principle of concordant profile where the secondary The span length can be modified to any length but
moments are zero at the intermediate supports. the secondary moment of the system is not affected
The equation for determining the secondary due to this change and when the eccentricities are
moment is derived from the basic principles of moment modified there will be a significant change in the
area and support displacement. For the purpose of secondary moment
analysis a two span continuous prestressed prismatic With these analyses, it has been noticed that the
concrete girder over supports A, B and C is modeled concordant profile can be achieved when the
and shown in Figure 1. The physical parameters of the eccentricity at the center of span is twice of eccen-
girders are shown below. tricity at the intermediate span. (i.e. e1 = e22 and the
L = Span length of AB & BC in ft secondary moment will be zero).
P = Prestressing force on the tendons in ksi
e1 = Eccentricity at mid span in inches
e2 = Eccentricity at intermediate support in inches REFERENCES
Extensive work has been done to study the behav- D.L. Foderberg & D.E. Branson (1968), Secondary moments
ior of prestressed concrete members in the past and in single span prestressed concrete beams and frames
very few of them deal with the calculation of sec- determined by column analogy, Journal of Prestressed
ondary moments. A simplified method to determine concrete Institute, 13(1), February 1968, pp 32.
the secondary moments has not yet been explored. T.Y. Lin & K. Thornton, (1972), Secondary moment and
The following methods are the key methods used in moment redistribution in continuous prestressed concrete
this paper to substantiate the validity of the equation. beams. The Journal of Prestressed concrete Institute,
17(0), Jan 1972, pp 816.
1. Moment area method
2. Support Displacement method
680
Fatigue testing of stay cables at resonant frequency
W. Traeger & J. Kollegger

Institute for Structural Engineering, Vienna University of Technology, Austria
ABSTRACT
Dynamic fatigue tests are used to evaluate the fatigue

behavior of materials and structures. The cycling load
for dynamic fatigue tests is generally applied by means
of servo hydraulic controlled jacks. The testing setup is
very simple but on the other hand the testing frequency
is low for high test loads (max. 1 Hz). The Institute for
Structural Engineering has developed a new method
for dynamic fatigue tests on large specimens like stay
cables (Kollegger, Kberl, Pardatscher & Vill).
Due to the novel method it is possible to increase the
testing frequency up to 40 Hz, also for large specimens.
After tensioning a specimen and an auxiliary cable,
anchored at a coupling unit by means of a hydraulic
jack an unbalanced vibration generator, attached at
the coupling unit applies the vibration loading. In case
the testing frequency of the specimen is equal to the
first eigenvalue of the testing setup the applied load
of the vibration generator is multiplied depending on
the damping of the system. So the novel fatigue test-
ing facility requires less time and a lower demand of
energy than conventional servo hydraulic controlled
testing units.
Due to promising results of preliminary tests and the
support of Vienna University of Technology a testing
facility for large specimens was built in the labora-
tory of the Institute for Structural Engineering. The
u-shaped reaction frame is 16 m long, 2.8 m wide and Figure 1. Testing facility.
2 m high with a total weight of 1500 kN. The reaction
frame is made of reinforced high strength concrete
and is post-tensioned in three directions in order to REFERENCE
remain free of cracks. This is very important to achieve
low damping. The testing facility is mounted on spring Kollegger, J.; Kberl, B.; Pardatscher, H.; Vill, M.: Verfahren
bearings and is therefore isolated from the foundation. zur Durchfhrung von Dauerschwingversuchen an einem
Prfkrper sowie eine Vorrichtung zur Durchfhrung des
The new testing method decreases the duration for
Verfahrens, sterreichisches Patent AT 501 168 B1, 2006.
fatigue tests dramatically and fatigue tests can be
carried out more economically. The testing unit is
dimensioned for a static tensile load up to 20.000 kN,
an upper load for fatigue tests up to 12.000 kN and a
vibration range up to 2.500 kN.
681
Durable connections between precast bridge components: Fundamental

approach
A.E. Ulku
Paul C. Rizzo Associates, Inc., Pittsburgh, PA, USA
U.B. Attanayake & H.M. Aktan

Western Michigan University, Kalamazoo, MI, USA
ABSTRACT region boundaries get narrower. Hence, use of cus-

tomized grout properties for specific applications is
The Federal Highway Administration statistics indi- recommended.
cate that more than 20 percent of the National Highway When two surfaces are debonded, both normal and
System (NHS) bridges and 27 percent of non-NHS longitudinal stresses increase due to a narrower bear-
bridges are structurally deficient or functionally obso- ing area. When component-to-grout thickness ratio
lete. Prefabricated bridge elements/systems are widely increases, the effects of friction are minimized. Bond-
used to accelerate the construction and assure the qual- ing between adjacent precast members is important
ity; thus, increase service life. Accelerated construc- when dry joints are used to connect members of similar
tion as well as the use of prefabricated elements has dimensions such as full-depth deck panels.
many inherent advantages including minimum traffic The joints closest to the load application points
disruption and increased work zone safety.Yet, the fail- would be the most critical due to lack of clamping
ure of empirically designed field implemented joint in between the posttension locations. Using wider
details for connecting precast components is a major fascia members would provide adequate distance for
durability concern. Development of joint details to posttension stress distribution even with large duct
assure monolithic structural behaviour with no cracks spacing.
and no water leakage under service loads require fun-
damental understanding of the joint behaviour against
design parameters.The common connection details are
to use grouted keyways or butted joints in conjunction REFERENCES
with posttension. AASHTO. 2007. AASHTO LRFD Bridge Design Specifica-
The objective of this study is to develop an under- tions, Fourth Edition, Washington, DC.
standing of posttension stress distribution at the joints Culmo, M.P. 2009. Connection Details for Prefabricated
between structural components such as beams, pan- Bridge Elements and Systems. FHWA-IF-09-010, Federal
els, etc., through a fundamental analytical approach. Highway Administration, Mclean, VA 22101.
Understanding of basic relations for stress distribu- Iyengar, S.R. & Alwar, R.S. 1964. Stresses in Layered Half-
tion in panel systems under concentrated loads is vital plane. Journal of the Engineering Mechanics Division,
to develop a fundamental approach for posttension ASCE 90(4): 79-96.
PCI. 2003. Precast Prestressed Bridge Design Manual.
design.
Chicago, IL.
The study evaluates the effects of modular ratio Poston, R. W., Phipps, A. R., Breen, J. E., and Carrasquillo,
of grout and precast member material and interface R., L. 1985. Design Procedures for Prestressed Con-
friction on posttension stress distribution. The paper crete Bridge Decks. Center of Transportation Research,
also deals with regions of uniform compression under Bureau of Engineering Research, The University of Texas
multiple posttension loads. Based on the results of the at Austin.
study, following conclusions can be drawn: Russell, H. G. 2009. Adjacent Precast Concrete Box Beam
When the grout layer has a substantially small Bridges: Connection Details. NCHRP Synthesis 393,
elasticity modulus compared to precast components, Transportation Research Board, Washington D.C.
Tahan, N., Pavlovic, M.N., & Kotsovos, M.D. 1993b. Single
post-tension can be distributed at far distances from the
Fourier Series Solutions for Rectangular Plates under In-
load point. By doing so, longitudinal tensile stresses Plane Forces, with Particular Reference to the Basic Prob-
at the edges of the component also increase consider- lem of Colinear Compression. Part 2: Stress Distribution.
ably. Selecting a grout material that has a comparable Thin-Walled Structures 17(1993): 126.
elasticity modulus to that of precast members is desir- Timoshenko, S.P. & Goodier, J.N. 1970. Theory of Elasticity.
able. If grout layers are stiffer, this time, compressed Third edition, New York: McGraw-Hill.
682
Bridge strengthening by structural change: from continuous beam

to network arch
M. Valenzuela & J.R. Casas

Civil Engineering Department. Technical University of Catalonia, Barcelona, Spain
ABSTRACT original deck using it as temporary support during

construction and permanent tie in service.
The paper focuses on solving problems in bridges due Additionally a transformation of the type of bridge
to damage suffered in intermediate supports, either by and the optimization of the geometry and materials are
damage to the element, or by undermining or scour obtained.
problems. These pathologies can be a problem if the This new alternative repair is sustainable, because
conditions of repetitive runoff are aggressive, gener- reduces the risk for future problems of infrastructure
ating poor security conditions and significant cost in costs and avoids repetitive and expensive reinforce-
reinforcements. ments. Additionally, gives a value to the concept
Traditional repair and strengthening techniques of repair, providing a new aesthetic concept to the
often are ineffective. Therefore the objective of this structure.
paper is to evaluate an alternative to these standard
solutions. The main idea is to remove the intermedi-
ate supports that are damaged by changing the bridge REFERENCES
typology, incorporating arches with hangars arrange-
Carracedo, M. & Valenzuela M. 20062007, Fact sheets
ment that hold the deck structure, thus allowing to and photographs, Santiago, Chile, Department of Bridges,
remove the piers. Ministry of Public Works, Chile.
For the study, a continuous reinforced concrete Moya, B. 2006, Project Basis: Reparacin puente San Luis
bridge, built in the town of Chimbarongo, Chile in (in Spanish). Department of Bridges, Ministry of Public
1930, is used as application example.The bridge shows Works, Chile.
severe damage from erosion, corrosion and scour in Tveit, P. 2007 The Network Arch. Bits of Manuscript
piers. The methodology is the modeling of the origi- after Lectures in 44 Countries, Internet Edition, http://
nal bridge in a structural analysis program, comparing pchome.grm.hia.no/pchome/
Valenzuela, M. 2007, Estudio de Factibilidad de Puentes en
it with a model of the bridge arches with vertical and
Arco tipo Network en Chile, (in Spanish). Thesis. Uni-
network hangars arrangements. versidad de Chile. Internet Edition, http://www.network-
This procedure confirms the possibility of redress arch.com
through the network arches, identifying elements and Valenzuela. M. 2010 Refuerzo de puentes de luces medias
technological procedures for the site, as the use of por conversin en arcos atirantados tipo network. Ph. D.
external brace and construction of the arch on the Thesis. In progress.
Figure 1. Longitudinal view of San Luis strengthened arch bridge.
683
Innovations in fabrication of the self-anchored suspension span of the San

Francisco-Oakland Bay Bridge
M. Wahbeh
Consultant to California Department of Transportation, Shanghai, China
P. Siegenthaler
California Department of Transportation, Oakland, CA, USA
T. Nilsson
American Bridge & Fluor Enterprise, Shanghai, China
M. Nader
T.Y. Lin International, San Francisco, CA, USA
A. Cavendish-Tribe
Hatch Mott MacDonald Limited, Consultant to Bay Area Toll Authority, Birmingham, UK
ABSTRACT combining years of experience that include construct-

ing the original SFOBB. AB/F determined that the
The 2.2-mile-long East Span of the San Francisco- structural steel portion of the SAS would be fabricated
Oakland Bay Bridge (SFOBB) is currently undergoing in China at the Shanghai Zhenhua Heavy Industry Co.
a seismic retrofit that will completely replace the Ltd. (ZPMC), a subsidiary of the China Communica-
existing steel truss bridge that opened in 1936.Approx- tion Construction Company (CCCC) on Changxing
imately 280,000 vehicles cross the SFOBB daily and Island just outside Shanghai. ZPMC is the largest
the bridge corridor is being brought up to current seis- heavy-duty equipment manufacturer in the world and
mic safety standards by replacing the existing East is famous for producing more than 75 percent of the
Span. This work is being accomplished while keeping worlds port cranes used in the shipping industry as
the existing bridge open to traffic. A key component of well as other large scale steel bridges such as the
the New East Span will be a Self-Anchored Suspension Golden Ears Bridge in Vancouver, Canada. ZPMC
(SAS) bridge chosen by the region and legislated to be implemented a number of innovations in order to expe-
the signature span in the retrofit of one of the nations dite the fabrication process and to assure meeting the
busiest bridges. When complete, this structure will be stringent quality requirements of what will soon be
the worlds largest SAS. American Bridge/Fluor Enter- a world-renowned architectural icon while remaining
prises (AB/F) (A Joint Venture) won the bid to build the backbone of regional transportation in the San
the SAS. Both companies have impressive portfolios Francisco Bay Area.
684
Fatigue life evaluation of existing highway reinforced concrete bridges
C. Wang, G. Li, X. Dong, L. Hao & J. Wang

Key Laboratory for Bridge and Tunnel Engineering of Shaanxi Province, Highway College, Changan University,
Xian, Shaanxi Province, China
ABSTRACT ACKNOWLEDGEMENT
Generally, fatigue has not been considered as a key The writers gratefully acknowledge the financial
problem in the design of reinforced concrete bridges. support provided by the Shaanxi Province Trans-
Until the 1960s, reinforcement was mild steel and portation Technology Research Project (Grant No.
the stresses permitted in the steel bar and the con- 07-04k) and Chinese West Transportation Develop-
crete were such that fatigue and fracture failure was ment Research Projects (Grant No. 200831849404,
believed to be impossible. With the developments of 200631822302-07).
reinforced concrete structures, higher working stresses
were permitted and, in particular, high yield reinforc-
ing bars were introduced. In recent years, some studies
showed that fatigue could occur in reinforced concrete REFERENCES
structures in combination with other causes of deteri-
ACI Committee 215., 1974. Consideration for design of con-
oration. In the past ten years, considerable increases
crete structure subjected to fatigue loading. ACI Journal,
in traffic volume and wheel loads have caused obvi- 71(3): 97121.
ously fatigue damage in existing highway reinforced BSI., 2007. Guide on Methods for Assessing the Acceptabil-
concrete bridges in China. Some highway reinforced ity of Flaws in Metallic Structures, BS7910:1999. British
concrete bridges were damaged seriously, leading to Standards Institution, London.
the whole bridge collapse sometimes because of over- ECCS., 1985. Recommendations for the Fatigue Design of
loading and oversize trucks. So how to evaluate the Steel Structures. European Convention for Constructional
fatigue safety of existing reinforced concrete bridges Steelwork, Brussels.
is an urgent problem in China. In the current paper, Helgason, T. and Hanson, J.M., 1976. Fatigue Strength of
High-Yield Reinforcing Bars. National Cooperative High-
the assessment models of existing reinforced con-
way Research Program Report 164, National Research
crete bridges based on S-N curve and linear elastic Council, Washington. D.C.
fracture mechanics (LEFM) were proposed. The two Herwig, A., 2008. Reinforced Concrete Bridges under
methods were applied to the case study bridge, and Increased Railway Traffic Loads. Ph.D. Thesis of cole
the fatigue life of the bridge was estimated using the Polytechnique Fdrale de Lausanne, Lausanne.
monitoring strain data and considering traffic volume Hordijk, D.A., 1991. Local Approach to Fatigue of Concrete.
increase rate. In short, the following conclusions could Ph.D. Thesis of Technische Universiteit Delft, Delft.
be gained: (1) The S-N curve and LEFM approaches Schlfli, M. and Brhwiler, E., 1998. Fatigue of existing rein-
could be effectively applied to evaluate the fatigue forced concrete bridge deck slabs. Engineering Structures.
20(11): 991998.
safety of actual RC bridges. (2) Combined with other
Walker, E.F. Harrison, I.M. and Morley, J., 1975. Fatigue and
factors of deterioration such as corrosion, corrosion Corrosion Fatigue of Reinforcement Bars. Department
fatigue could cause the remarkable reduction of fatigue of Civil and Structural Engineering Report, University
life for reinforced concrete bridges. (3) Furthermore, College, Cardiff, UK.
the effect of overloading and oversize trucks should be Zhao, G.F., 1999. The Advanced Reinforced Concrete Struc-
investigated in-depth. tures. China Electric Publish Press, Beijing.
685
Durability rehabilitation of a reinforced concrete bridge damaged

by corrosion and overload
H.L. Wang & X.Y. Sun

Department of Civil Engineering, Zhejiang University, Hangzhou, PR China
Z.J. Zheng
Hangzhou Urban Construction Design and Research Institute Co. Ltd., Hangzhou, PR China
ABSTRACT
Cracking and spalling of concrete cover caused by

corrosion of steel rebar and overload is the major dam-
age of reinforced concrete (RC) bridge regarding its
durability in the coastal area. To improve performance
of RC structures and prolong their service life, the
cracked members must be repaired effectively. The
conventional method is removing the cover and recast-
ing it with new concrete, which is destructive and
chlorides in the residual concrete can still increase the
risk to the further corrosion of steel rebar. Considering
the corrosion mechanism of RC structures, a non-
destructive electrochemical technique was carried out
to rehabilitate a RC bridge damaged by chloride and
overload after 30 years service in the coastal region. Figure 1. Effect of cracks treatments on the chloride
Some researchers have focused their interests on the removal.
indoor experiments about this method, but there is a
gap among indoor experiments, design and practice.
In order to make a good understand of this method, REFERENCES
this research presented a work on the engineering
applications. Arya C et al. 1996. Factors influencing electrochemical
The rehabilitation process was introduced in this removal of chloride from concrete, Cement and Concrete
study. The effectiveness of this strategy was assessed Research 26(6):10951104.
by inspecting the initial and the final chloride ions Castellote M et al. 2000. Electrochemical removal of chlo-
rides: modelling of the extraction, resulting profiles and
in concrete together with the carbonated depth. The
determination of the efficient time of treatment. Cement
tested results showed that about 22% to 74% of the and Concrete Research 30:615621.
initial chloride was removed from the concrete after Elsener B & Angst U. 2007. Mechanism of electrochemical
40 days treatment, and the chloride contents close to chloride removal. Corrosion Science 49:45044522.
the rebar were reduced by 49% to 81% of the initial val- Li SL et al. 2004. Effects of electrochemical desalination on
ues according to the different treatments. In addition, microstructural characteristics of steel mortar interface.
the carbonated concrete cover had been re-alkalized The Ocean Engineering 22(2):7578.
quickly. Mietz J. 1998. Electrochemical rehabilitation methods for
Cracks and repair materials in the damaged zone reinforced concrete structures: a state of the art report.
London: IOM Communications Ltd.
had significant effects on the desalination efficiency.
Orellan JC et al. 2004. Electrochemical chloride extraction:
An effective path for the transportation of chloride efficiency and side effects. Cement and Concrete Research
must be established between the anode and the rebar 34:227234.
in concrete. Therefore, the cracks should be sealed Polder RB. 1996. Electrochemical chloride removal from
before the application of this technique in order to concrete prisms containing chloride penetrated from sea
avoid the short circuits, and the repair materials should water. Construction and Building Materials, 10(1):8388.
be conducive to the transportation of chloride.
686
Where are the sustainable bridges in the United States?
D. Whittemore
AI Engineers, Inc., Middletown, CT
ABSTRACT goals and objectives in each of these categories is

posed to form in concept a complete standardized
Far from the abstraction of a wooden crossing tucked sustainable bridge metric.
away in a national park, the concept of sustainable Hard and soft advantages for pursuing a sustainable
bridge design is one that needs further investigation bridge design are presented. While this question is one
and application by the bridge design, maintenance, and deserving of further study, hard benefits derive directly
ownership communities. from the proposed metric, while the soft benefits of
Sustainable bridge design can be defined as an this system range from public approval to the open-
engineering project that is conceived, designed, con- ing up of federally earmarked dollars for sustainable
structed, operated, maintained, and eventually put out infrastructure and energy savings.
of service in such a fashion that these activities After defining sustainable design and outlining a
demand as little as possible from the natural, material proposed quantifiable metric, a serious look at the cur-
and energy resources of the surrounding supporting rent standard of sustainable bridge design in the United
community. In order to achieve this goal, the design- States is examined. Without a national standard or best
ers, owners, and maintenance professionals need to practice defined, it is difficult to properly compare any
approach each aspect of their craft from new per- two structures that claim the title of sustainable. Both
spectives during each phase of a proposed structures academic study of the benefits and a national stan-
lifespan. dard spearheaded by AASHTO and embedded in their
From this basic definition, a series of standard current design and maintenance guides in the United
bridge metrics is proposed, and compared against States would be appropriate for any meaningful dis-
the benchmark U.S. Green Building Councils bench- cussion or practice of sustainable bridge design. This
mark LEED standard (U.S. Green Building Council, work would also have the additional benefit of reigning
2006): in the now wide open field of green material vendors
Each of these equivalent bridge design goals is and processes.
delved in further detail, and a series of possible design
REFERENCES
Table 1. LEED Design Goals vs. Sustainable Bridges.
Brundtland Commission. 1987. Our Common Future, Report
Equivalent Bridge by the Brundtland Commission. Oxford University Press
LEED Design Goals Design Goal U.S. Green Building Council. 2006. New Construction &
Major Renovation Version 2.2 Reference Guide. Washing-
Sustainable Sites Sustainable Sites ton, D.C.:U.S. Green Building Council.
Water Efficiency Water Efficiency Center for Environmental Excellence by AASHTO. 2009.
Energy and Atmosphere Energy and Transportation Sustainability. Web. http://environment.transportation.org/
Materials and Resources Materials and Resources environmental_issues/sustainability/
Indoor Environmental Quality N/A
Innovation in Design Innovation in Design
687
A heuristic approach for optimizing bridge inspection route
C.-I. Yen
Department of Research & Development, National Taiwan University of Science and Technology, Taiwan
H.-K. Liao
Department of Civil Engineering, National Central University, Taiwan
J.-J. Chai & N.-J. Yau

Institute of Construction Engineering and Management, National Central University, Taiwan
ABSTRACT component (according to last inspection record);

X3 = number of span.
Bridge inspection, which can identify deteriorations In this research, a heuristic rule was established
and damages, is a major task of bridge maintenance. to identify a feasible initial route. First of all, select
In Taiwan, three types of inspection are performed bridges located on main roads and find the direct dis-
for roadway bridges: (1) daily patrol, (2) regular tance from O based on their GPS coordinates. Second,
inspection, and (3) special inspection. All of them are choose the nearest bridge from O as the first bridge
visual inspection without using heavy equipments or on schedule. Use vehicle speed set by inspector can
instruments. calculate travel time between O and Bridge 1 (RO1 ).
Daily patrol and special inspection are quick and Meanwhile, one can use Equation 1 to estimate the
simple inspection and usually dont need to schedule inspection time of Bridge 1 (T1 ). For the same work
in advance. Compared with daily patrol and special day, subtracting RO1 and T1 from remaining work
inspection, regular inspection requires much more hours to see how many hours left. If there is enough
time to perform detailed visual inspection. In addition, time travel to next bridge and complete the inspection,
there are usually time limitations for bridge manage- then move to the next bridge. Otherwise, the day is
ment agencies to complete their regular inspections, done and start from O to schedule for the next day.
such as before rainy season. Initial route is a good solution but not optimized.
The goal of this research is to establish a two-stage The second-stage utilized GA to improve the initial
heuristic approach that can help bridge management solution. A real world case of 68 bridges were put into
agencies quickly schedule an optimal inspection plan the program and studied. The result of the study case
for daily route. The first stage develops a heuristic pro- is satisfied. The program rearranged the sequence of
cedure to quickly identify a feasible initial solution; the bridges, reduced the duration from 15 days to 13 days,
second stage then takes the initial solution from stage and has 6.31 hours remain for the last day.
one and improves it utilizing Genetic Algorithm (GA). This research established a model to find an optima
There are six constraints and assumptions, includ- route for inspecting a group of bridges. Not only does
ing (1) the inspection crew will start off from the it find the shortest path that connects every bridge,
main office called O, (2) only one inspection crew, but also assign all bridges into a certain work day.
(3) the number of work-hour per day is fixed, (4) every Multifactor linear regression analysis was applied to
bridges GPS coordinates are known, (5) the travel time estimate the bridge inspection time. There are some
between two bridges is calculated by the direct dis- plans for future research. The current model uses direct
tance, and (6) the inspection of a bridge is inseparable distance between two bridges to calculate the travel
unless the bridge is too long to complete in one day. time. The future model will incorporate this model
The bridge inspection time can be estimated by with GIS and navigation system to estimate more
Equation (1) derived from a multifactor linear regres- accurate travel time among bridges.
sion analysis.
REFERENCE
Marco Dorigo & Luca Maria Gambardella 1997. Ant
Where Y = bridge regular inspection time (minute); Colonies for The Travelling Salesman Problem, BioSys-
X1 = slab area (m2 ); X2 = number of deteriorated tems 43: 7381.
688
Assessment of the operation level of a bridge network in postearthquake

scenarios
Y.C. Yue, D. Zonta, F. Bortot & R. Zandonini

DIMS, University of Trento, Italy
ABSTRACT
This paper presents the results of seismic evaluations at

bridge level and network level, carried out on the road
network of the Autonomous Province of Trento (APT).
The APT Transportation Department operates approx-
imately 2340 kilometers of roads and 1000 bridges,
using a real-time internet based Bridge Management
System (BMS). This system (for more details of the
APT stock, see Zonta et al. 2007) incorporates seismic
vulnerability analysis capability based on the defini-
tion of fragility curves for all the bridges inventoried.
Fragility curves are conditional probabilities that a Figure 1. Simulated network and the safest path between
bridge will reach a certain limit state (operational, Lavaz Pass and Riccomassimo.
damage control, life safety, collapse) under a given
ground motion intensity level (Nielson 2005). to prepare a pre-earthquake plan, and for the post-
Currently in the APT-BMS, fragility curves for a earthquake emergency response. Google Earth and
specific bridge are developed by capacity-spectrum GIS tools are used to present the results of seismic eval-
approach according to HAZUS guidelines(HAZUS- uation and network prioritization. This combination
MH MR3 2003; Basz & Mander 1999), which is can offer straightforward visualization, and is a great
a rapid approach seeking to establish dependable help to government officials and non-professionals in
fragility curves. In contrast to other methods that have understanding research results.
been used in the past, such as empirical fragility curves
or analytical fragility curves that require much pre-
vious damage data or extensive computation, only REFERENCES
limited information is needed for this model. Basz, N. & Mander, J.B. 1999, Enhancement of the Highway
Using this approach, the system can generate a Transportation Lifeline Module in HAZUS, Final Pre-
probabilistic damage scenario over the whole network, Publication Draft (#7) prepared for the National Institute
given a simulated design earthquake. We show that, of Building Sciences, March 31, 1999.
in the case of the APT stock, the direct seismic risk Dijkstra, E. W. 1959. A Note on Two Problems in Connexion
involving collapse or loss of life is moderate. By con- with Graphs. Numerische Mathematik l, 269271 (l 959)
trast, we expect a critical problem in network operation HAZUS-MH MR3 2003. Multi-hazard Loss Estimation
in a post-earthquake situation, when it will be neces- Methodology (Earthquake Model). Technical Manual.
Department of Homeland Security Emergency Prepared-
sary to identify the safest path between any two places
ness and Response Directorate. FEMA. Washington, D.C.
in APT region. Nielson, B.G. 2005. Analytical fragility curves for highway
This problem is addressed using Dijkstras graph bridges for moderate seismic zones. Ph.D. Dissertation,
search algorithm (Dijkstra, 1959) to find the shortest Georgia institute of technology, Atlanta, GA.
path between a source-destination pair for a non- Zonta, D., Zandonini, R. & Bortot, F. 2007.A reliability-based
directional and non-negative cost path graph. Giving bridge management concept. Structure and Infrastructure
the simulated network graph of APT region, this Engineering. Structure and Infrastructure Engineering
method can show the best path rapidly and clearly 3(3): 215235.
between any two places on Google Earth, as in fig-
ure 1. The results can be used by decision makers
689
A distributed application for infrastructure risk prevention
M.H. Zaharia & F. Leon

Computer Science & Engineering dept., Technical University Gheorghe Asachi of Iasi, Romania
G.M. Atanasiu
Structural Mechanics dept., Technical University Gheorghe Asachi of Iasi, Romania
G. Pagu & E. Czeran

Microsoft Center, Technical University Gheorghe Asachi of Iasi, Romania
ABSTRACT The paper also illustrates the deployment of infor-

mation into a geographical map where the information,
Risk mitigation during the life cycle monitoring available in real time & instantaneously for a Cen-
of critical infrastructure usually involves an expert tral/Regional Command Office, will put in action an
screen-ing for a proper evaluation. This activity is alert signal, in emergency situations in case of some
needed for the risk management in both pre- and post- critical stages for structures immediately after the
disaster occurrence. The present paper illustrates a seismic event.
distributed application taking into account the crite- The software is a web-enabled GIS-based applica-
ria and approaches of the risk management of built tion. Because it was developed using free maps, it
infrastructure. also has lower implementations costs. Some similar
The software was created to code the importance of applications are already developed but none has the
a report in accordance with the associated risk. So, at combina-tion of facilities centered to the field report
any time, the user can see the most important reports and the real-time transmission using GIS in infrastruc-
or a class of importance selected at his/her choice. ture risk prevention. The approach is feasible because
This will help to easily track the most needed infras- the new Internet access for mobile devices seems to
tructure investment that must be made to avoid for be cheaper but also resilient at various types of normal
example the associated risks if an earthquake with a disasters.
certain magnitude occurs. Therefore, this helps in the
dis-aster prevention phase. When a disaster occur the
most important problem is usual that at the command REFERENCES
center mostly disparate and incomplete information
will came. As a result, the efficiency of central coordi- Angermann, M., Khider, M., Frassl, M. & Lichtenstern M.,
2009. DMT An integrated disaster management tool,
nation is highly decreased due to inherent distribution
published in K. Duncan , C. A. Brebbia Disaster Manage-
and isolation of intervention teams. By using this tool, ment and Human Health Risk: Reducing Risk, Improving
any expert that will have either a mini PC or a PDA Outcomes (Wit Transactions on the Built Environment)
with GPS connected to the Internet can give all needed WIT Press.
information about all kinds of damage and estimation Dawes, S.S., Knafo, L., Levin, P. , Schimke, K. & Steve Kos S.
in accordance with his/her domain. The software is 2004. Learning from Crisis: Lessons from the World Trade
created in such as way that any expert has a tracking Center Response. A Research Symposium Panel Tran-
route that can help a rescue team to save him/her if script Summary, http://www.ctg.albany.edu/publications/
he/she suddenly disappears without a trace. reports/wtc_symposium
FEMA 2009. Get Disaster Information www.fema.gov/hazard
From the point of view of the expert, the applica-
/index.shtm
tion is created to be as simple as possible to be used. IFRC 2009. Disaster management, www.ifrc.org/what/ dis-
A login process is required due to the importance of asters/responding/drs/tools/dmis.asp
the processed data. After that, the user will have the JRC 2009. Earthquakes: support for disaster management,
possibility to see all the reports that are already made, ipsc.jrc.ec.europa.eu/news.php?id=43
to search from the reports and to modify application Kircher, C.A., Whitman R.V. & Holmes W.T. 2006. HAZUS
settings such as the language, for example. Earthquake Loss Estimation Methods, Natural Hazards
The proposed methodology is able to provide the Rev. Volume 7, Issue 2, pp. 4559.
critical information for rescue teams, considering also NEES 2009. About NEES, https://www.nees.org/about/
the unsafe position of expert immediately after an
extreme seismic event, during the recovery phase.
690
The relevance of whole life costs for infrastructure buildings
T. Zinke & T. Ummenhofer

Karlsruhe Institute of Technology, Karlsruhe, Germany
ABSTRACT
Since several years calculation of life cycle costs

becomes more and more important and the appraisal
of these costs is discussed in an enormous num-
ber of papers. But life cycle costs of a bridge are
normally defined as direct costs of construction, oper-
ation, maintenance, rehabilitation and removal. In
contrast whole life costs additionally consider costs
generated due to time-delays of the infrastructure
users, an increasing number of accidents and cumu-
lative air pollution. These costs are summarized in the Figure 1. Classification of costs according to ISO 15686.
category externalities. Furthermore non-construction
costs and benefits resulting from improved time of
travel, reduced traffic hazard and therefore reduction Due to the great importance to consider all cate-
of accidents have to be taken into account. A classi- gories of the whole life costs and the uncertainties
fication considering the different cost components is described in this paper, further examinations espe-
given in ISO 15686. cially for building activities in the infrastructure sector
Since the 1980s different studies for identifying and have to be realized. Therefore a structural bridge
assessing externalities as well as for the determination design has to be developed, which is not only mini-
of external costs were performed in the traffic sector. mizing manufacturing and maintenance costs but also
Up to now no detailed analysis of external costs from arising external costs. The goal must remain to create a
bridges exists. If external costs are to be calculated and holistic assessment system for bridges. The challeng-
internalized a monetarily quantification is required. In ing problem is to perform a quantification of external
the course of cost calculation it is necessary to find a costs and benefits.
proper procedure for the assessment of goods which
are not part of a market and for whom no sales values
are existing. REFERENCES
In most cases external costs arising during the whole Frangopol, D.M. 1999. Life-cycle cost analysis for bridges. In
life cycle of a bridge exceed the original building Bridge Safety and Reliability, edited by D.M. Frangopol,
costs by far. The biggest impacts result from time pp. 210236, 1999 (ASCE: Reston, VA).
delay in the course of an obstruction of traffic. Other Singh, D. & Tiong, R.L.K. 2005. Developement of life
relevant external costs are basically related to the com- cycle costing framework for highway bridges in Myamar.
ponents accidents, air pollution and climate change. International Journal lof Project Management 23: 3744.
However, the amount of costs shows a significant vari- INFRAS/IWW 2000. External costs of Transport. Accident,
ety in different studies and guidelines. Depending on Environmental and Congestion Costs in Western Europe.
Zrich/Karlsruhe: UIC.
the assumptions, considered externalities and the kind
UNITE 2003. Unification of accounts and marginal costs
of approach for a consideration of long term effects for Transport Efficiency. Final Report, European Com-
the results vary within a broad range. The clear results mission, 5th Framework Transport RTD.
point out that an approximate calculation of external
costs is better than neglecting them. Therefore a con-
sideration of external costs is indispensable to achieve
efficient macroeconomic decisions.
691

(Dan M. Frangopol (Ed.), Richard Sause (Ed.), Chad PDF

Uploaded by

Copyright:

Available Formats

(Dan M. Frangopol (Ed.), Richard Sause (Ed.), Chad PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Dan M. Frangopol (Ed.), Richard Sause (Ed.), Chad PDF

Uploaded by

Copyright:

Available Formats

What is the conference about?

What is the conference about?

What are some of the keynote topics discussed?

What are some of the keynote topics discussed?

BRIDGE MAINTENANCE, SAFETY, MANAGEMENT AND LIFE-CYCLE OPTIMIZATION

PROCEEDINGS OF THE FIFTH INTERNATIONAL CONFERENCE ON BRIDGE MAINTENANCE,

Bridge Maintenance, Safety,

2010 Taylor & Francis Group, London, UK

Typeset by MPS Ltd. (A Macmillan Company), Chennai, India

Published by: CRC Press/Balkema

T.Y. Lin Lecture

Bridge maintenance and safety: A practitioners view 3

Fundamentals of suspension bridge retrofit 11

MS2: Monitoring & assessment of bridges using novel techniques

MS3: Present & future of bridge inspection & evaluation

MS4: Research & applications in bridge health monitoring

MS6: Measurement systems for bridge weigh-in-motion (B-WIM)

MS7: Bridge management tools & research

MS8: Uncertainty in bridge damageability modelling

MS9: Performance-based asset & risk management of the highway

MS10: Bridging the data gaps for effective management

MS12: Computational prediction & in field validation of bridge performance

MS13: Management & maintenance of long span bridges

MS14: SmartEN Marie Curie ITN Smart management for

MS16: Safety & management of bridges in Mexico

MS17: COWI Group mini-symposium Cable supported bridges

MS18: Lifetime design of super long span bridges

SS3: Bridge condition assessment

SS4: Bridges for high speed railways

SS5: Industrial smart material applications for civil infrastructure (ISMA)

SS6: Advances in structural robustness: dependability framework

SS8: Nondeterministic schemes for structural safety & reliability of bridges

SS10: Life cycle bridge engineering in Korea

SS13: Current advancements in bridge technology

SS15: Modeling of bridge seismic response

SS16: Recent challenging bridge structures

SS17: Life-cycle design of structural systems

Study of remaining fatigue life of Brazilian concrete bridges 492

SS18: Optical monitoring techniques for bridge maintenance & safety

Base study on measure of transformation of concrete structures used by digital image

Development of approximation process of existing action stress of pre-stressed concrete

Development of a remote laser scanning system for continuous monitoring of

3D profile measurement and buckling simulation on thin-walled cylindrical shells under

Deflection measurement for bridges with frequency-shifted feedback laser 502

SS19: Implementation of bridge management administration in Japan

Evaluation of maintenance cost estimation and feedback to the BMS 506

Preventive maintenance and technical development on long-span bridges 509

Mitigation of terrorist threats to structural components 517

SS21: Structural monitoring of bridges: Hong Kongs experience

Thermal behaviors of Tsing Ma Suspension Bridge 523

Long-term performance of structural health monitoring system in bridges 526

SS23: Performance-based design for steel structures

Reliability evaluation of RC piers of highway bridge using time series multi-state

Performance-based durability design of weathering steel bridge 534

SS24: Steel bridge rehabilitation

SS25: New developments in bridge design codes

SS26: WIM-based live loads for bridges

SS27: High performance concrete lessons of past decades

SS28: Construction, architecture & testing of soil-steel bridges