ASEE 2014 Zone I Conference, April 3-5, 2014, University of Bridgeport, Bridgeport, CT, USA.

Structural Performance of Highway Bridges under

Given Foundation Settlements
Zhan Su*; Qian Wang, PhD, PE, Assistant Professor

Huanzi Wang, PhD, PE, Senior Bridge Engineer

Department of Civil and Environmental Engineering

Manhattan College
Riverdale, NY 10471
*Email: [email protected]; Tel: 917-860-2218;
Email: [email protected]; Tel: 718-862-7469

AECOM Technology
2101 Webster St, Suite 1900
Oakland CA 94612
Email: [email protected]; Tel: 510-622-6608

AbstractThis paper presents a study of structural

performance of highway bridges under given foundation
settlements. The ultimate objective of this research is to analyze
and develop acceptable levels of bridge foundation settlements
under ultimate and service limit states, which are not adequately
addressed in current bridge design standards. In order to design
modern bridges to accommodate any expected foundation
settlements, structural engineers need to consider foundation
movements and quantify their impacts on the performance of
bridge superstructures. In this paper, one type of widely used
highway bridges, cast-in-place posttensioned concrete box girder
bridges, is considered and settlement analyses are performed to
study the induced bridge girder stresses due to foundation or pier
settlements. Two sets of bridge examples are studied with
different bridge skew angles and horizontal curve radii, to
investigate the effects of these parameters. A state-of-art bridge
analysis finite element (FE) program is used to analyze various
bridge responses. It is shown that the bridge girder curvature
and skew angle can impact the girder stresses due to foundation
or pier settlements, and the trends are plotted using a numerical
parametric study. Based on the evaluation of stress increase due
to settlements, an acceptable pier settlement limit can be further
determined based on allowable girder stresses in the next stage of
the study.
Keywordsfoundation settlement; numerical study; cast-inplace concrete; highway bridges; finite element (FE)

I.

INTRODUCTION

Bridge designs in real world need to consider foundation

movements, since there are many situations where settlements
could occur at the foundations of a bridges abutments or piers.
Foundation settlements can be induced due to the compaction
or consolidation of soils, higher traffic loads, and poor
drainage, etc. In order to design modern bridges to
accommodate the expected overall and differential movements
to bridge performance requirements, structural engineers need
to use the foundation movements estimated by geotechnical
engineers. It is important to accurately predict the short-term
and long-term effects due to settlements from bridge piers or
abutments.

Settlement study of bridge structures and foundations has

been performed for a long time [1-5]. However, the study of
bridge foundation settlements has been primarily focused on
approach slabs in the literature [2-5]. Bridge settlement criteria
are not adequately addressed in the current bridge design
standards, such as the AASHTO specifications [6]. There is a
need to study bridge settlements and improve bridge design
practice to adequately address bridge foundation settlements,
and their short-term and long-term impacts on the bridge
superstructures, such as bridge girders and decks.
The ultimate objective of this research is to analyze and
develop acceptable levels of bridge foundation overall and
differential movements considering service and strength limit
states. In this paper, cast-in-place posttensioned concrete box
girder bridges are considered, which are widely used in
highway bridge constructions. Settlement analyses are
performed to study the induced bridge girder stresses due to
various foundation or pier settlements. Two sets of bridge
examples are adopted with different bridge skew angles and
horizontal curve radii, to study the effects of these parameters
on additional bridge girder stresses due to pier settlements. A
state-of-art FE program for bridge analysis - CSI-Bridge [7] is
used to analyze the bridge responses.
II.

BRIDGE SETTLEMENT STUDY

A. Cast-In-Place Highway Bridge

Highway bridges with various structure types are
constructed in practice, depending on the span ranges of the
bridges. Cast-in-place posttensioned concrete box girder
bridges are widely used for highway bridges spanning up to
600 feet, since they provide many advantages in terms of
appearance, safety, economy, and maintenance [8]. In this
paper, this type of highway bridges is considered. Two
posttensioned concrete highway bridges are shown in Figure
1, representing typical 3-span and 2-span cast-in-place
concrete bridges constructed in California.

B. Two Bridge Models

Example 1 (EX1) is a 3-span posttensioned cast-in-place
concrete box girder bridge with a mid-span of 150 feet and
100 feet on each side span, as shown in Figure 1(a). There are
two piers and two abutments in EX1. Example 2 (EX2) is a 2span concrete box girder bridge with a typical span of 150
feet. There are one pier and two abutments in EX2. In both
examples, the bridge piers are 30 feet tall, with 6 feet in
diameter. A commercial FE code, CSI-Bridge [7] is used to
model and analyze the bridges. In the two examples, the
following general conditions are assumed:
(b) Example 2 (EX2): A Two-Span Bridge

The concrete strength is 4 ksi.

The bases of the piers are considered as fully fixed, and

the abutments are assumed to be roller-supported.

The foundation or pier settlements are defined at the

bases of the piers.

An integral cap beam is adopted for all the bridges.

The piers are modeled with three-dimensional (3D)

frame elements, and the girder slabs are modelled using
3D shell elements.

An overlay of 20 psf is considered. Traffic barriers are

used on each side of the bridge girders, and the
distributed weight is 1.2 klf.

C. Bridge Evaluation Matrix

Various bridge design cases are examined using FE
simulations to provide a basis for evaluating the settlements
and corresponding bridge girder stresses. Table 1 lists the
bridge evaluation matrix defined in this study. Four bridge
girder horizontal curve radii and five bridge skew angles are
selected. Bridges and settlements are evaluated based on the
following parameters:

(a) Example 1 (EX1): A Three-Span Bridge

(c) Cross Section of the Concrete Box Girders

Figure 1. Two Examples: Posttensioned Concrete Bridges

1. Horizontal Curve Radii: the four different bridge girder

curve radii considered are 250 feet, 500 feet, 1000 feet,
and infinite radius, respectively. The last case represents
a straight bridge. The 3-span bridges in Example 1 with
various horizontal curve radii are illustrated in Figure 2.
2. Skew Angles: To study the effects of bridge skew angles
on bridge girder settlement stresses, five different bridge
skew angles are considered in this study, namely 0, 15,
30, 45, and 60. These are shown in Figure 3.
3. Settlement Cases: Bridge pier settlement cases, including
both one-pier settlement and two-pier settlement, are
considered in the FE simulations. Figure 4 shows the
simulation setups for both one-pier settlement and twopier settlement. The settlements are defined at the bottom
of the piers, as seen in Figure 4, representing the
foundation settlement below one pier, or simultaneous
settlements underneath two piers. The magnitude of pier
settlements is denoted as s, as shown in Figure 4. For
Example 2, only one-pier settlement is possible. For both
one-pier settlement and two-pier settlement, two
different pier settlement magnitudes are considered in
this study, namely 0.5 inch and 1.0 inch. The
consideration of a settlement of 0.5 inch is to verify the
bridge FE analysis results obtained using 1.0 inch as the
pier settlement. For the rest of the paper, only the FE
analysis results based on the one-inch settlement are
presented.

Table 1. Bridge Evaluation Matrix

Curve Radius (ft)

250
500
1,000
Infinity

Skew Angle (Degree)

0
15
30
45
60

Pier Settlement (in)

0.5
1.0

(b) Skew Angle = 30 Degrees

(a) Radius = 250 feet

(c) Skew Angle = 45 Degrees

(b) Radius = 500 feet

(d) Skew Angle = 60 Degrees

Figure 3. Example 1: 3-Span Bridges with Various Skew Angles (EX1)
(c) Radius = 1,000 feet

(d) Radius = Infinite (straight bridge)

(a) Example 1: One-Pier Settlement

Figure 2. Example 1: 3-Span Bridges with Various Radii (EX1)

(b) Example 2: One-Pier Settlement

(a) Skew Angle = 15 Degrees

(c) Example 2: Two-Pier Settlement

(c) Example 2: Two-Pier Settlement

Figure 4. Pier Settlement Cases

Figure 5. Bridge Girder Deformed Shapes (Dead Load + Settlement)

III.

RESULTS AND DISCUSSION

A. Bridge Deformation and Stress Plots

The deformed shapes of the bridge girders for the two
examples under both dead load and pier settlements are
plotted in Figure 5. As expected the dead load deformation is
dominant in both examples, where the maximum girder
deformation is found at mid-span between the piers, and piers
and abutments. For Example 2, the deformed shapes for onepier settlement and two-pier settlement cases are similar.
Figure 6 shows the bridge girder stress contour plots under
both dead load and pier settlements for the two examples. It is
found that the girder stress induced by pier settlements has
maximum (tension) and minimum (compression) values at the
pier supports.

(a) Example 1: One-Pier Settlement

(b) Example 2: One-Pier Settlement

(a) Example 1: One-Pier Settlement

(b) Example 2: One-Pier Settlement

(c) Example 2: Two-Pier Settlement

Figure 6. Bridge Girder Stress Contour Plots (Dead Load + Settlement)

B. Bridge Settlement Stress Analysis

The maximum bridge girder tensile and compressive
stresses in the two examples due to one-inch pier settlements
corresponding to various bridge skew angles and horizontal
curve radii are plotted in Figure 7. It can be seen from Figure
7(a) that for both examples, the maximum girder stresses due
to pier settlements increase as the skew angle increases. The
bridge horizontal curve radius does not significantly impact
the maximum girder stresses due to pier settlements, as can be
seen from Figure 7(b). The maximum girder settlement stress
curves become flat when the bridge horizontal curve radius is
more than 500 feet. For Example 1, as expected that two-pier
settlement typically induces larger settlement tensile and
compressive stresses in bridge girders than one-pier
settlement, when skew angle is less than 40 degrees and curve
radius is more than 500 feet.
(a) Maximum Stress Change vs. Skew Angle

(a) Maximum Settlement Stress vs. Skew Angle

(b) Maximum Stress Change vs. Curve Radius

Figure 8. Bridge Girder Stress Change (%) Due to Pier Settlements (1 inch)

(b) Maximum Settlement Stress vs. Curve Radius

Figure 7. Maximum Bridge Girder Stress Due to Pier Settlements (1 inch)

Figure 8 shows the bridge girder stress changes due to oneinch pier settlements. The maximum stress change is
calculated based on the maximum girder stress due to pier
settlements divided by the maximum dead load stress in the
girder, expressed as a percentage. The maximum bridge girder
stress changes in the two examples due to pier settlements
corresponding to various bridge skew angles are plotted in
Figure 8(a). It can be seen that for Example 1, the maximum
girder stress change increases as the skew angle increases,
from about 20% to 40% when skew angle is zero, to 40% to
60% when skew angle is 60 degrees. For Example 2, the
maximum girder stress change increases as the skew angle
increases, from about 3% to 5% when skew angle is zero, to
6% to 10% when skew angle reaches 60 degrees. Figure 8(b)
shows the maximum bridge girder stress change in the two
examples due to pier settlements corresponding to various
bridge girder curve radii. For Example 1, the girder stress

change increases as the curve radius becomes bigger. When

the bridge girder curve radius is more than 500 feet, the
change of girder stress becomes very small, as can be seen
from the flat curves plotted. For Example 2, the girder curve
radius does not significantly impact the bridge girder stress
change due to pier settlements. For Example 1, as expected
that in the two-pier settlement case, larger stress changes in
bridge girders are typically induced than in the one-pier
settlement case, when bridge skew angle is less than 40
degrees and horizontal curve radius is more than 500 feet.
From the observation and analysis above, it can be seen
that:
(1) Bridge girder stress and percentage of stress change
due to bridge pier settlements is affected by bridge skew angle
and horizontal curve radius. Generally speaking as the skew
angle increases, the girder stress and percentage of stress
change due to pier settlements increase. Their magnitudes
increase by 100 percent as the skew angle increases from 0 to
60 degrees. When horizontal curve radius is more than 500
feet, the impact of horizontal curve radius to girder stress
induced by pier settlements is insignificant.

The advantage of the proposed method is the simple

application based on FE analysis and associated software.
More work, especially nonlinear structural modeling and
analysis is deemed to be beneficial to study the topic.
Nonlinear bridge analyses, including concrete creep, shrinkage
analysis, and bridge staged-construction simulation, will
provide useful results in addition to the linear elastic analysis
models. To make the research results applicable for practical
applications, bridges with different structural types and other
forms of bridge settlements such as rotations and differential
settlements of piers shall be considered. Verification is
required to calibrate the numerical models. Note that there is
no verification from field testing is available at this point. We
would like to propose field testing on bridges to verify the
numerical analytical models.
ACKNOWLEDGMENT
Support for this research provided by the Manhattan
College Mahony Foundation (Dr. John Mahony) is gratefully
acknowledged.

(2) Two-pier simultaneous settlements typically induce

larger stress and percentage of stress change in bridge girders
than one-pier settlements, when bridge skew angle is small
and horizontal curve radius is large.

[1]

(3) For two-span bridges, a one-inch pier settlement

produces an average of 5-10% stress increase in bridge
girders. For three-span bridges, a one-inch pier settlement
results in an average of 20-40% stress increase.

[3]

(4) Based on the bridge girder stress increase due to pier

settlements, an acceptable pier settlement limit can be
determined based on allowable girder stresses. This will be
addressed in the next stage of this study.

REFERENCES

[2]

[4]

[5]

IV.

FUTURE RESEARCH

The focus of this research is to study the structural

performance of highway bridges under given foundation
settlements. Two cast-in-place concrete bridge examples were
adopted to study different foundation settlement scenarios and
their impact on the bridge superstructures. FE models of the
bridges were built and the girder stresses were calculated. It
was shown that the bridge girder curvature and skew angle
could impact the settlement stresses and the trends were
plotted using a numerical parametric study.

[6]

[7]
[8]

W. J. Eden and H. B. Poorooshasb, Settlement observations at Kars

Bridge, Canadian Geotechnical Journal, vol. 5(1), pp. 28-45, 1968.
M. Zaman, A. Gopalasingam, and J. G. Laguros, Consolidation
settlement of bridge approach foundation, J. Geotech. Eng., vol. 117(2),
pp. 219240, 1991.
J. L. Briaud, R. W. James, and S. B. Hoffman, Settlement on Bridge
Approaches (the Bump at the End of the Bridge), NCHRP Synthesis
234, Washington, D.C.: Transportation Research Board, National
Research Council, 1997.
H. Ha, J. B. Seo, J. L. Briaud, Investigation of Settlement at Bridge
Approach Slab Expansion Joint: Survey and Site Investigations. Rep.
No. 4147-1 to the Texas Department of Transportation, TX: Texas
Transportation Institute, Texas A&M Univ. System, College Station,
2003.
C. Cai, X. Shi, G. Voyiadjis, and Z. Zhang, Structural performance of
bridge approach slabs under given embankment settlement, Journal of
Bridge Engineering, vol. 10(4), pp. 482-489. 2005.
AASHTO, AASHTO LRFD Bridge Design Specifications, 5th Edition,
Washington, D.C.: American Association of State Highway and
Transportation Officials, 2010.
Computers and Structures, Inc., CSI Analysis Reference Manual for
SAP2000, ETABS, SAFE and CSiBridge, Berkley, CA, 2011.
R. M. Barker and J. A. Puckett, Design of Highway Bridges: An LRFD
Approach, 3rd Edition, Hoboken, NJ: Wiley, 2013.