FEMA P-58-2-SE Volume2 Implementation

Seismic Performance
Assessment of Buildings
Volume 2 – Implementation Guide
Second Edition
FEMA P-58-2 / December 2018

FEMA P-58-2 / December 2018
Seismic Performance Assessment of Buildings

Volume 2 – Implementation Guide
Second Edition
Prepared by
APPLIED TECHNOLOGY COUNCIL
201 Redwood Shores Parkway, Suite 240
Redwood City, California 94065
www.ATCouncil.org
Prepared for
FEDERAL EMERGENCY MANAGEMENT AGENCY
Michael Mahoney, Project Officer
Robert D. Hanson, Technical Monitor
Washington, D.C.
SECOND EDITION PROJECT TEAM
APPLIED TECHNOLOGY COUNCIL STAKEHOLDER PRODUCTS TEAM

Christopher Rojahn Laura Samant (Team Leader)
Jon A. Heintz (Project Executive) David Mar
Ayse Hortacsu (Project Manager) Lori Peek
Maryann T. Phipps
PROJECT MANAGEMENT COMMITTEE Sharyl Rabinovici
Ronald O. Hamburger (Project Technical Director) L. Thomas Tobin
John Gillengerten
William T. Holmes PERFORMANCE PRODUCTS TEAM
John D. Hooper John Gillengerten (Team Leader)
Stephen A. Mahin David R. Bonneville
Jack P. Moehle Dominic Campi
Khalid Mosalam Vesna Terzic
Laura Samant
Steven R. Winkel PRODUCTS UPDATE TEAM
John D. Hooper (Team Leader)
PROJECT STEERING COMMITTEE Russell Larsen
William T. Holmes (Chair) Peter Morris
Lucy Arendt
Deborah Beck
Christopher Deneff
H. John Price
Jonathan C. Siu
Jeffrey R. Soulages
Eric Von Berg
Williston Warren (ATC Board Contact)
STAKEHOLDER PRODUCTS PRODUCTS UPDATE
CONSULTANTS CONSULTANTS
Sandra L. Grabowski Robert Bachman
Taline Mitten Jack Baker
Stacia Sydoriak Dustin Cook
Jennifer Tobin-Gurley Scott Hagie
Angie Harris
PERFORMANCE PRODUCTS Curt Haselton
CONSULTANTS Wyatt Henderson
Shreyash Chokshi Monica Huang
Travis Chrupalo Gilberto Mosqueda
Erica Hays Farzad Naeim
Nirmal Kumawat Barbara Rodriguez
Abe Lynn Kathrina Simonen
Daniel Saldana Siavash Sorooshian
Vinit M. Shah Katherine Wade
Udit S. Tambe Farzin Zareian
Duy Vu To
Peny Villanueva
FIRST EDITION PROJECT TEAM
APPLIED TECHNOLOGY COUNCIL RISK MANAGEMENT PRODUCTS

Christopher Rojahn (Project Executive) TEAM
Jon A. Heintz (Project Manager) John D. Hooper (Co-Team Leader)
Ayse Hortacsu Craig D. Comartin (Co-Team Leader)
Mary Comerio
PROJECT MANAGEMENT COMMITTEE C. Allin Cornell
Ronald O. Hamburger (Proj. Tech. Director) Mahmoud Hachem
John Gillengerten Gee Heckscher
William T. Holmes (ex-officio) Judith Mitrani-Reiser
Peter J. May Peter Morris
Jack P. Moehle Farzad Naeim
Maryann T. Phipps (ATC Board Contact) Keith Porter
Hope Seligson
STEERING COMMITTEE
William T. Holmes (Chair) STRUCTURAL PERFORMANCE
Roger D. Borcherdt PRODUCTS TEAM
Anne Bostrom Andrew S. Whittaker (Team Leader)
Bruce Burr Gregory Deierlein
Kelly Cobeen John D. Hooper
Anthony B. Court Yin-Nan Huang
Terry Dooley Laura Lowes
Dan Gramer Nicolas Luco
Michael Griffin Andrew T. Merovich
R. Jay Love
David Mar
Steven McCabe
Brian J. Meacham
William J. Petak
NONSTRUCTURAL PERFORMANCE NONSTRUCTURAL PERFORMANCE
PRODUCTS TEAM PRODUCTS AND FRAGILITY
Robert E. Bachman (Team Leader) DEVELOPMENT CONSULTANTS
Philip J. Caldwell Richard Behr
Andre Filiatrault Greg Hardy
Robert P. Kennedy Christopher Higgins
Helmut Krawinkler Gayle Johnson
Manos Maragakis Paul Kremer
Eduardo Miranda Dave McCormick
Gilberto Mosqueda Ali M. Memari
Keith Porter William O’Brien
John Osteraas
RISK MANAGEMENT PRODUCTS Elizabeth Pahl
CONSULTANTS John Stevenson
Travis Chrupalo Xin Xu
D. Jared DeBock
Armen Der Kiureghian FRAGILITY REVIEW PANEL
Scott Hagie Bruce Ellingwood
Curt Haselton Robert P. Kennedy
Russell Larsen Stephen Mahin
Juan Murcia-Delso
Scott Shell VALIDATION/VERIFICATION TEAM
P. Benson Shing Charles Scawthorn (Chair)
Mohamed Talaat Jack Baker
Farzin Zareian David Bonneville
Hope Seligson
STRUCTURAL PERFORMANCE
PRODUCTS AND FRAGILITY SPECIAL REVIEWERS
DEVELOPMENT CONSULTANTS Thalia Anagnos
Jack Baker Fouad M. Bendimerad
Dhiman Basu
Dan Dolan
Charles Ekiert
Andre Filiatrault
Aysegul Gogus
Kerem Gulec
Dawn Lehman
Jingjuan Li
Eric Lumpkin
Juan Murcia-Delso
Hussein Okail
Charles Roeder
P. Benson Shing
Christopher Smith
Victor Victorsson
John Wallace
Notice
Any opinions, findings, conclusions, or recommendations expressed in this publication do not
necessarily reflect the views of the Applied Technology Council (ATC), the Department of
Homeland Security (DHS), or the Federal Emergency Management Agency (FEMA).
Additionally, neither ATC, DHS, FEMA, nor any of their employees, makes any warranty,
expressed or implied, nor assumes any legal liability or responsibility for the accuracy,
completeness, or usefulness of any information, product, or process included in this publication.
Users of information from this publication assume all liability arising from such use.
Cover photograph – Collapsed building viewed through the archway of an adjacent building, 1999 Chi-Chi,
Taiwan earthquake (courtesy of Farzad Naeim, Farzad Naeim, Inc).
Preface
In 2001, the Applied Technology Council (ATC) was awarded the first in a
series of contracts with the Federal Emergency Management Agency
(FEMA) to develop Next-Generation Performance-Based Seismic Design
Guidelines for New and Existing Buildings. These projects would become
known as the ATC-58 series of projects. The overall program was separated
into two major phases of work: Phase 1 – Developing a Methodology for
Assessing the Seismic Performance of Buildings; and Phase 2 – Developing
Performance-Based Seismic Design Procedures and Guidelines.
This report is one of three principal products of the Phase 1 work, which
were first published in 2012, and have since been updated under the Phase 2
work:
• FEMA P-58-1, Seismic Performance Assessment of Buildings,
Volume 1 – Methodology, Second Edition
Volume 2 – Implementation Guide, Second Edition
Volume 3 – Supporting Electronic Materials and Background
Documentation, Third Edition
The three volumes together describe the seismic performance assessment

methodology for individual buildings, as well as the development of basic
building information, response quantities, fragilities, and consequence data
used as inputs to the methodology. The procedures are probabilistic,
uncertainties are explicitly considered, and performance is expressed as the
probable consequences, in terms of human losses (deaths and serious
injuries), direct economic losses (building repair or replacement costs), and
indirect losses (repair time, unsafe placarding, and environmental impacts)
resulting from building damage due to earthquake shaking.
The purpose of this Volume 2 –Implementation Guide, Second Edition is to

provide users with step-by-step guidance in the development of basic
building information, response quantities, fragilities, and consequence data
used as inputs to the methodology.
FEMA P-58-2 Preface v

Second Edition
To allow for practical implementation of the methodology, work included the
collection of fragility and consequence data for most common structural
systems and building occupancies, and the development of an electronic
Performance Assessment Calculation Tool (PACT) for performing the
probabilistic computations and accumulation of losses, which were also
updated as part of the Phase 2 work.
Updates include technical improvements to the methodology, and revisions

to the underlying fragility and consequence data, to take advantage of the
latest research, and to bring results into better alignment with expectations
based on the performance of buildings in past earthquakes. Phase 2 also
included the development of additional volumes intended to expand and
complete FEMA P-58 series of products to communicate seismic
performance to stakeholders, quantify the seismic performance capability of
typical code-conforming buildings, and provide performance-based design
guidance to assist decision-makers in choosing between seismic design
criteria and making seismic design decisions.
The FEMA P-58 series of products is the result of the collaborative effort of
more than 200 individuals, across all phases, involved in the development of
the methodology and subsequent products and reports. Phase 1 consultants
were involved in the development of the basic assessment methodology,
collection of available fragility data, estimation of consequences,
development of supporting electronic tools, implementation of quality
assurance procedures, and beta testing efforts. ATC is indebted to the
Phase 1 leadership of Ron Hamburger, who served as Project Technical
Director, Bob Bachman, who served as Nonstructural Performance Products
Team Leader, John Hooper and Craig Comartin, who served as Risk
Management Products Team Leaders, Andrew Whittaker, who served as
Structural Performance Products Team Leader, and the members of the
Phase 1 Project Management Committee, including John Gillengerten, Bill
Holmes, Peter May, Jack Moehle, and Maryann Phipps. ATC is also
indebted to Andy Merovich, Structural Performance Products Team Member,
for his lead role in the development of the First Edition of this volume.
Phase 2 consultants were responsible for the update of all Phase 1 products,
and the development of new performance-based design guidance and
stakeholder guidance. ATC is indebted to the Phase 2 leadership of Ron
Hamburger, who continued to serve as the Project Technical Director, John
Gillengerten, who served as Performance Products Team Leader, John
Hooper, who served as Products Update Team Leader, Laura Samant, who
served as Stakeholder Products Team Leader, and the members of the
vi Preface FEMA P-58-2

Second Edition
Phase 2 Project Management Committee, including Bill Holmes, Steve
Mahin, Jack Moehle, Khalid Mosalam, and Steve Winkel.
ATC would also like to thank the members of the Phase 1 and Phase 2
Project Steering Committees, the Fragility Review Panel, the Nonstructural
Performance Products Team, the Performance Products Team, the Products
Update Team, the Risk Management Products Team, the Stakeholder
Products team, the Structural Performance Products Team, the
Validation/Verification Team, and the many consultants who assisted these
teams. The names of individuals who served in these groups, along with
their affiliations, are provided in the list of Project Participants at the end of
this report.
ATC acknowledges the Pacific Earthquake Engineering Research Center

(PEER), and its framework for performance-based earthquake engineering,
as the technical basis underlying the methodology. In particular, the work of
Tony Yang, Jack Moehle, Craig Comartin, and Armen Der Kiureghian in
developing and presenting the first practical application of the PEER
framework, is recognized as the basis for how computations are performed
and losses are accumulated in the methodology.
Special acknowledgment is extended to C. Allin Cornell and Helmut

Krawinkler for their formative work in contributing to risk assessment and
performance-based design methodologies, and to whom this work is
dedicated.
ATC also gratefully acknowledges Michael Mahoney (FEMA Project

Officer) and Robert Hanson (FEMA Technical Monitor) for their input and
guidance in the conduct of this work, and Bernadette Hadnagy, Peter N.
Mork, and Carrie Perna for ATC report production services.
Ayse Hortacsu Jon A. Heintz

ATC Director of Projects ATC Executive Director
FEMA P-58-2 Preface vii

Second Edition
Table of Contents
Preface.............................................................................................................v
List of Figures...............................................................................................xv
List of Tables ............................................................................................ xxiii
1. Introduction ...................................................................................... 1-1

1.1 Purpose and Scope.................................................................... 1-1
1.2 Limitations ............................................................................... 1-2
1.3 The Performance Assessment Process ..................................... 1-2
1.3.1 Assessment Types ....................................................... 1-2
1.3.2 Performance Measures ................................................ 1-3
1.3.3 Analysis Methods ........................................................ 1-4
1.4 Updates to Implementation Tools ............................................ 1-5
1.5 Organization and Content ......................................................... 1-7
2. Building Performance Model .......................................................... 2-1

2.1 Introduction .............................................................................. 2-1
2.2 Building Characteristics ........................................................... 2-1
2.2.1 Project Information...................................................... 2-1
2.2.2 Building Information ................................................... 2-2
2.2.3 Population Model ........................................................ 2-9
2.3 Fragility Specifications and Performance Groups .................. 2-11
2.3.1 Fragility Specifications Provided in PACT ............... 2-12
2.3.2 Demand Directionality .............................................. 2-16
2.4 Identification of Vulnerable Structural Components and
Systems................................................................................... 2-16
2.4.1 Structural Steel Elements (B103) .............................. 2-17
2.4.2 Reinforced Concrete Elements (B104)...................... 2-20
2.4.3 Reinforced Masonry Elements (B105) ...................... 2-24
2.4.4 Cold-Formed Steel Structural Elements (B106)........ 2-25
2.4.5 Wood Light Frame Structural Elements (B107) ....... 2-25
2.4.6 Structural Performance Groups ................................. 2-26
2.5 Identification of Vulnerable Nonstructural Components
and Systems ............................................................................ 2-28
2.5.1 Exterior Nonstructural Walls (B107 and B201) ........ 2-29
2.5.2 Exterior Window Systems (B202) ............................ 2-31
2.5.3 Roof Elements (B30) ................................................. 2-33
2.5.4 Partitions (C101) ....................................................... 2-33
2.5.5 Stairs (C201) ............................................................. 2-34
2.5.6 Wall Finishes (C301)................................................. 2-35
2.5.7 Raised Access Floors and Floor Flooding (C302) .... 2-36
2.5.8 Suspended Ceilings and Ceiling Lighting (C303) ..... 2-37
2.5.9 Elevators and Lifts (D101) ........................................ 2-38
FEMA-P-58-2 Table of Contents ix

Second Edition
2.5.10 Domestic Water Distribution (D202) ........................ 2-39
2.5.11 Sanitary Waste Piping (D203) ................................... 2-39
2.5.12 Chilled Water and Steam Piping (D205 and
D206) ......................................................................... 2-40
2.5.13 Chillers, Cooling Towers, and Compressors
(D303)........................................................................ 2-40
2.5.14 HVAC Distribution Systems (D304) ......................... 2-42
2.5.15 Packaged Air Handling Units (D305)........................ 2-43
2.5.16 Control Panels (Table D306) ..................................... 2-44
2.5.17 Fire Protection (D401) ............................................... 2-44
2.5.18 Electrical Service and Distribution (D501) ............... 2-45
2.5.19 Other Electrical Systems (D509) ............................... 2-45
2.5.20 Equipment and Furnishings (E20) ............................. 2-46
2.5.21 Special Construction (F20) ........................................ 2-48
2.5.22 Nonstructural Performance Groups ........................... 2-48
2.6 Collapse Fragility Analysis..................................................... 2-49
2.6.1 Nonlinear Response History Analysis ....................... 2-50
2.6.2 Simplified Nonlinear Analysis .................................. 2-50
2.6.3 Judgment-Based Approach ........................................ 2-55
2.6.4 Collapse Modes and PACT Input .............................. 2-57
2.7 Residual Drift Fragility ........................................................... 2-58
3. Building Analytical Model and Performance Assessments .......... 3-1

3.1 Introduction............................................................................... 3-1
3.2 Building Analytical Model ....................................................... 3-2
3.2.1 Nonlinear Response History Analysis ......................... 3-2
3.2.2 Simplified Analysis ..................................................... 3-3
3.2.3 Demand Directionality................................................. 3-4
3.3 Intensity-Based Assessment ..................................................... 3-4
3.3.1 Nonlinear Response History Analysis ......................... 3-4
3.3.2 Simplified Analysis ................................................... 3-11
3.3.3 Review Results .......................................................... 3-28
3.4 Scenario-Based Assessment ................................................... 3-32
3.4.3 Review Results .......................................................... 3-42
3.5 Time-Based Assessment ......................................................... 3-42
3.5.3 Review Results .......................................................... 3-52
4. Example Application: Intensity-Based Performance Assessment

Using Simplified Analysis ................................................................ 4-1
4.1 Introduction............................................................................... 4-1
4.2 Obtain Site and Building Description ....................................... 4-1
4.3 Select Assessment Type and Performance Measure................. 4-3
4.4 Assemble Building Performance Model ................................... 4-3
4.4.1 Project Information ...................................................... 4-3
4.4.2 Building Information ................................................... 4-4
4.4.3 Population Distribution................................................ 4-5
4.4.4 Structural Components ................................................ 4-5
4.4.5 Nonstructural Components ........................................ 4-10
4.4.6 Collapse Fragility and Collapse Modes ..................... 4-16
x Table of Contents FEMA-P-58-2

Second Edition
4.4.7 Residual Drift Fragility ............................................. 4-19
4.5 Select Analysis Method and Construct Building Analytical
Model ..................................................................................... 4-19
4.6 Define Earthquake Hazards .................................................... 4-19
4.7 Analyze Building Response ................................................... 4-24
4.7.1 Estimate Median Story Drift Ratio and Dispersion ... 4-24
4.7.2 Estimate Median Peak Floor Acceleration and
Dispersion.................................................................. 4-26
4.7.3 Estimate Median Residual Story Drift Ratio and
Dispersion.................................................................. 4-27
4.8 Input Response Data and Calculate Performance................... 4-28
4.9 Review Results ....................................................................... 4-29
5. Example Application: Time-Based Performance Assessment

Using Nonlinear Response History Analysis.................................. 5-1
5.1 Introduction .............................................................................. 5-1
5.2 Obtain Site and Building Description....................................... 5-1
5.3 Select Assessment Type and Performance Measures ............... 5-2
5.4 Assemble Performance Assessment Model.............................. 5-2
5.5 Select Analysis Method and Construct Building Analytical
Model ....................................................................................... 5-3
5.6 Define Earthquake Hazards ...................................................... 5-3
5.7 Analyze Building Response ..................................................... 5-9
5.8 Input Response Data and Calculate Performance................... 5-13
5.9 Review Results ....................................................................... 5-14
6. Structural Fragility Calculation ..................................................... 6-1

6.1 Introduction .............................................................................. 6-1
6.2 Building Description ................................................................ 6-1
6.3 Development of Structural Components and System
Fragilities by Calculation ......................................................... 6-5
6.3.1 Plywood Roof Diaphragm ........................................... 6-5
6.3.2 Tilt-Up Walls............................................................. 6-13
6.3.3 Wall to Roof Attachments ......................................... 6-19
6.4 User-Defined Fragilities in PACT .......................................... 6-29
7. Nonstructural Fragility Calculation ............................................... 7-1

7.1 Introduction .............................................................................. 7-1
7.2 Unanchored Components ......................................................... 7-1
7.2.1 Overturning ................................................................. 7-2
7.2.2 Sliding Displacement .................................................. 7-6
7.3 Anchored Components ............................................................. 7-7
7.3.1 Code-Based Limit State Determination of
Anchorage Fragility..................................................... 7-7
7.3.2 Strength-Based Limit State Approach to Anchorage
Fragility Calculation .................................................. 7-14
7.4 Displacement-Based Limit State Approach to Define
Calculated Fragilities.............................................................. 7-16
8. Consequence Function Development.............................................. 8-1

8.1 Introduction .............................................................................. 8-1
8.2 General Considerations ............................................................ 8-1
FEMA-P-58-2 Table of Contents xi

Second Edition
8.3 Calculation of Consequence Functions..................................... 8-2
8.3.1 Repair Cost .................................................................. 8-2
8.3.2 Repair Time ................................................................. 8-4
8.3.3 Unsafe Placards ........................................................... 8-5
8.3.4 Casualties ..................................................................... 8-7
8.3.5 Environmental Impact ................................................. 8-8
8.4 Example Application .............................................................. 8-10
8.4.1 Estimation of Repair Cost.......................................... 8-10
8.4.2 Estimation of Repair Time......................................... 8-12
8.4.3 Estimation of Environmental Impacts ....................... 8-12
8.5 Consequence Functions in PACT ........................................... 8-15
8.5.1 Provided Consequence Functions in PACT............... 8-15
8.5.2 User-Defined Consequence Functions ...................... 8-15
8.5.3 Other Considerations ................................................. 8-19
Appendix A: Structural Component Fragility Specifications ............... A-1
Appendix B: Nonstructural Component Fragility Specifications ......... B-1
Appendix C: PACT User Manual ............................................................ C-1

C.1 Introduction.............................................................................. C-1
C.2 Hardware and Software Requirements .................................... C-1
C.3 Installing PACT ....................................................................... C-2
C.4 PACT Control Panel ................................................................ C-2
C.5 Model the Building and Import Analyses Results ................... C-3
C.5.1 Building Modeler Menus ............................................ C-4
C.5.2 Project Information Tab.............................................. C-6
C.5.3 Building Information Tab ........................................... C-6
C.5.4 Population Tab ............................................................ C-7
C.5.5 Component Fragilities Tab ......................................... C-9
C.5.6 Performance Groups Tab .......................................... C-10
C.5.7 Collapse Fragility Tab .............................................. C-11
C.5.8 Structural Analysis Results Tab................................ C-12
C.5.9 Residual Drift Tab .................................................... C-14
C.5.10 Hazard Curve Tab ..................................................... C-15
C.5.11 Saving the PACT model ........................................... C-15
C.6 Evaluate Performance ............................................................ C-16
C.6.1 Command Line Evaluation ....................................... C-18
C.7 Examine Results .................................................................... C-18
C.7.1 Scenario/Intensity Results ........................................ C-19
C.7.2 Time-Based Results .................................................. C-29
C.8 Fragility Specification Manager ............................................ C-32
C.8.1 Overview Tab ........................................................... C-32
C.8.2 Fragility Specification Details Tab ........................... C-34
C.9 Population Manager ............................................................... C-44
C.10 Reporting ............................................................................... C-45
C.11 A Look under the Hood ......................................................... C-46
C.11.1 Project Files .............................................................. C-46
C.11.2 Results Files .............................................................. C-46
C.11.3 Fragility Specification Files ...................................... C-47
C.11.4 Population Specification Files .................................. C-47
C.11.5 Reporting Template Files ......................................... C-47
C.11.6 Programming Notes .................................................. C-48
xii Table of Contents FEMA-P-58-2

Second Edition
Appendix D: Normative Quantity Estimation Tool User Manual........ D-1
D.1 Introduction ............................................................................. D-1
D.2 Usage Notes............................................................................. D-1
D.3 Normative Quantity Estimate Tab ........................................... D-2
D.3.1 Building Definition Table .......................................... D-2
D.3.2 Component Summary Matrix ..................................... D-2
D.3.3 Additional Features .................................................... D-3
D.4 Normative Quantity Database Tab .......................................... D-5
D.4.1 Component Definition Table ...................................... D-5
D.4.2 Building Occupancy Component Matrix.................... D-6
Glossary ......................................................................................................E-1
Symbols ....................................................................................................... F-1
References .................................................................................................. G-1
Project Participants .................................................................................. H-1
FEMA-P-58-2 Table of Contents xiii

Second Edition
List of Figures
Figure 1-1 Performance assessment process ....................................... 1-4
Figure 2-1 Format of PACT screenshots used, indicating the tab

title location ...................................................................... 2-1
Figure 2-2 PACT Project Information tab ........................................... 2-2
Figure 2-3 PACT Building Information tab ........................................ 2-3
Figure 2-4 Definition of floor and story numbers and floor and
story heights ....................................................................... 2-4
Figure 2-5 PACT Population tab showing commercial office

occupancy ........................................................................ 2-10
Figure 2-6 PACT Population Manager utility ................................... 2-10
Figure 2-7 PACT Component Fragilities tab showing fragility

specification selection ...................................................... 2-15
Figure 2-8 PACT Performance Groups tab. ...................................... 2-28
Figure 2-9 Normative Quantity Estimation Tool, Building

Definition Table. .............................................................. 2-49
Figure 2-10 Normative Quantity Estimation Tool, Component

Summary Matrix. ............................................................. 2-49
Figure 2-11 SPO2IDA idealized pushover curve for hypothetical

structure ........................................................................... 2-51
Figure 2-12 Plan of example structure................................................. 2-52
Figure 2-13 Section of example structure............................................ 2-52
Figure 2-14 Pushover curve for 2-story reinforced masonry

building ............................................................................ 2-53
Figure 2-15 SPO2IDA Input............................................................... 2-54
Figure 2-16 SPO2IDA Output ............................................................. 2-54
Figure 2-17 Comparison of collapse fragility functions obtained

with two different approaches: SPO2IDA and
Judgment .......................................................................... 2-56
FEMA P-58-2 List of Figures xv

Second Edition
Figure 2-18 Illustration of multiple collapse modes ............................ 2-57
Figure 2-19 PACT Residual Drift tab .................................................. 2-59
Figure 3-1 Performance assessment procedure showing steps

covered in this chapter in shaded box ................................ 3-1
Figure 3-2 Example spectra for scaled suite of 5 motions ................... 3-6
Figure 3-3 PACT Structural Analysis Results tab with response

history drift input.............................................................. 3-10
Figure 3-4 PACT Residual Drift input............................................... 3-10
Figure 3-5 Definition of floor, story numbers and floor heights

above grade for two-story building .................................. 3-15
Figure 3-6 Lumped weight distribution ............................................. 3-15
Figure 3-7 Results from linear analysis ............................................. 3-16
Figure 3-8 PACT Structural Analysis Results tab, intensity-based

assessment ........................................................................ 3-26
Figure 3-10 PACT Control Panel ........................................................ 3-29
Figure 3-11 PACT Engine window ..................................................... 3-29
Figure 3-12 PACT Repair Cost tab ...................................................... 3-30
Figure 3-13 PACT Realizations tab ..................................................... 3-31
Figure 3-14 PACT Repair Cost tab showing repair cost by

performance group ........................................................... 3-31
Figure 3-15 PACT Data Drill Down and Exports tab .......................... 3-32
Figure 3-16 Standard deviation of Sa vs. T attenuation plot from

www.opensha.org ............................................................. 3-37
Figure 3-17 PACT Structural Analysis Results tab, scenario-based

assessment, nonlinear analysis ......................................... 3-38
Figure 3-18 Median Sa vs. T attenuation plot from

www.opensha.org ............................................................. 3-39
Figure 3-19 PACT Structural Analysis Results tab, scenario-based

assessment, simplified analysis ........................................ 3-42
Figure 3-20 Seismic hazard plot showing segments ............................ 3-46
xvi List of Figures FEMA P-58-2

Second Edition
Figure 3-21 Variation of mean annual frequency of exceedance for
peak ground acceleration in g .......................................... 3-48
Figure 3-22 PACT Hazard Curve tab .................................................. 3-52
Figure 4-1 Plan view of example four-story office building ............... 4-2
Figure 4-2 Typical frame elevation for example four story office
building .............................................................................. 4-2
Figure 4-3 PACT Project Information tab ........................................... 4-4
Figure 4-4 PACT Building Information tab ........................................ 4-5
Figure 4-5 PACT input screen for beam/column joint fragility .......... 4-6
Figure 4-6 Illustration of reinforced concrete element fragility

specification selections ...................................................... 4-8
Figure 4-7 PACT entries for 1st floor structural performance

groups, direction 1 ........................................................... 4-10

groups, direction 2 ........................................................... 4-10

groups, nondirectional ..................................................... 4-10
Figure 4-10 Normative Quantity Estimation Tool, Building

Definition Table ............................................................... 4-11
Figure 4-11 Normative Quantity Estimation Tool, Component

Summary Matrix showing nonstructural inventory ......... 4-12
Figure 4-12 PACT Performance Groups tab for floor 1,

non-directional ................................................................. 4-13
Figure 4-13 Pushover curve developed by analysis ............................ 4-17
Figure 4-14 SPO2IDA Tool, SPO tab ................................................. 4-17
Figure 4-15 SPO2IDA Tool, IDA results tab ...................................... 4-18
Figure 4-16 PACT Collapse Fragility tab............................................ 4-18
Figure 4-17 USGS hazard data for 1.0 second period at

Site Class B ...................................................................... 4-21
Figure 4-18 Seismic hazard curve for Sa(T = 1.13s), Site Class D ...... 4-22
Figure 4-19 USGS hazard data for peak ground acceleration ............. 4-23
Figure 4-20 Results from linear analysis ............................................. 4-25
FEMA P-58-2 List of Figures xvii

Second Edition
Figure 4-21 PACT peak transient drift ratio input on Structural
Analysis Results tab ......................................................... 4-28
Figure 4-23 PACT Repair Cost tab ...................................................... 4-30
Figure 4-24 PACT Repair Cost tab with realizations .......................... 4-31
Figure 5-1 Seismic Hazard Curve for Sa(T = 1.13s), Site Class D....... 5-4
Figure 5-2 Segmented seismic hazard curve to be used for time-

based assessment ................................................................ 5-5
Figure 5-3 Uniform Hazard Spectrum ................................................. 5-7
Figure 5-4 Deaggregation of segment 3 ............................................... 5-8
Figure 5-5 Conditional mean spectrum for segment 3 ......................... 5-8
Figure 5-6 PACT Structural Analysis Results tab with drift input
for intensity 3 ................................................................... 5-13
Figure 5-7 PACT Residual Drift tab input ......................................... 5-14
Figure 5-8 PACT Time Based Results tab showing annualized

repair cost ......................................................................... 5-15
Figure 6-1 Roof plan ............................................................................ 6-2
Figure 6-2 Cross-sectional elevation.................................................... 6-2
Figure 6-3 In-plane attachment ............................................................ 6-4
Figure 6-4 Out-of-plane attachment..................................................... 6-4
Figure 6-5 Horizontal holdown (out-of-plane) attachment .................. 6-4
Figure 6-6 Plywood diaphragm damage .............................................. 6-6
Figure 6-7 Plywood diaphragm damage .............................................. 6-6
Figure 6-8 Generalized force-deformation relationship adapted

from ASCE/SEI 41-06 ....................................................... 6-7
Figure 6-9 Determining plywood diaphragm damage states ............... 6-8
Figure 6-10 Fragility functions for plywood diaphragm in transverse

direction............................................................................ 6-12
Figure 6-11 Idealized moment-rotation relationship ........................... 6-14
Figure 6-12 Photo showing DS2 level cracking .................................. 6-16
xviii List of Figures FEMA P-58-2

Second Edition
Figure 6-13 Fragility functions for out-of-plane wall flexure ............. 6-19
Figure 6-14 Fragility function for in plane ledger to wall

connection ........................................................................ 6-22
Figure 6-15 Fragility function for out-of-plane wall-to-roof

nailing .............................................................................. 6-29
Figure 6-16 PACT Fragility Specification Manager showing new

specification ..................................................................... 6-30
Figure 6-17 Fragility specification data input form............................. 6-30
Figure 6-18 New fragility damage state definition form ..................... 6-31
Figure 7-1 Overturning parameters for unanchored objects ................ 7-2
Figure 7-2 Fragility function for overturning of 5-shelf metal

bookcase............................................................................. 7-5
Figure 7-3 Fragility function for sliding of 5-shelf metal bookcase .... 7-7
Figure 7-4 Fragility function for anchorage of rigidly mounted

equipment designed under the 1994 UBC ....................... 7-12
Figure 7-5 Fragility function for anchorage of vibration isolated

equipment designed under the 1994 UBC ....................... 7-13
Figure 8-1 Initiating a new fragility specification in Fragility

Specification Manager ..................................................... 8-16
Figure 8-2 Add New Fragility window ............................................. 8-16
Figure 8-3 Consequence Function input window .............................. 8-17
Figure 8-4 PACT graph showing upper and lower bound repair
cost data ........................................................................... 8-17
Figure 8-5 Other Consequences tab, Unsafe Placards input.............. 8-18
Figure 8-6 Other Consequences tab, Casualty input ......................... 8-19
Figure C-1 PACT shortcut icon ...........................................................C-2
Figure C-2 PACT Control Panel ..........................................................C-3
Figure C-3 PACT Building Modeler window ......................................C-4
Figure C-4 File menu ...........................................................................C-4
Figure C-5 Open File menu ..................................................................C-5
Figure C-6 Project Information tab with retrieved data from

opened file..........................................................................C-5
FEMA P-58-2 List of Figures xix

Second Edition
Figure C-7 Edit, Tools, and Help menus ............................................. C-6
Figure C-8 Building Information tab ................................................... C-7
Figure C-9 Population tab.................................................................... C-8
Figure C-10 Population model data screens .......................................... C-9
Figure C-11 Component Fragilities tab ............................................... C-10
Figure C-12 Performance Groups tab .................................................. C-11
Figure C-13 Demand parameter selection data block.......................... C-11
Figure C-14 Collapse Fragility tab ...................................................... C-12
Figure C-15 Structural Analysis Results tab for entry of nonlinear

analysis results ................................................................ C-13
Figure C-16 Structural Analysis Results tab for entry of simplified

analysis results ................................................................ C-14
Figure C-17 Residual Drift tab ............................................................ C-15
Figure C-18 Hazard Curve tab............................................................. C-16
Figure C-19 PACT Engine window .................................................... C-16
Figure C-20 Analysis options menu .................................................... C-17
Figure C-21 PACT Engine progress screen......................................... C-17
Figure C-22 Results window ............................................................... C-19
Figure C-23 Performance Curve display options ................................ C-19
Figure C-24 Results displays ............................................................... C-20
Figure C-25 Changing performance grouping levels .......................... C-20
Figure C-26 Effects of changing performance grouping levels........... C-21
Figure C-27 Repair Cost tab ................................................................ C-22
Figure C-28 Casualties tab .................................................................. C-23
Figure C-29 Repair strategy selection menu ....................................... C-24
Figure C-30 Repair time output assuming floors are repaired in

parallel ............................................................................. C-24
Figure C-31 Repair time output assuming floors are repaired

sequentially ..................................................................... C-25
xx List of Figures FEMA P-58-2

Second Edition
Figure C-32 Environmental Impacts tab ..............................................C-25
Figure C-33 Unsafe Placards tab ..........................................................C-26
Figure C-34 Realizations categories ....................................................C-26
Figure C-35 Individual Realization display of repair cost and repair

time contributions by performance group ........................C-27
Figure C-36 Result Drill Down screen .................................................C-28
Figure C-37 Screen showing options for exporting results ..................C-29
Figure C-38 Time-based results output screen .....................................C-30
Figure C-39 Options menu showing base curve mode selection .........C-30
Figure C-40 Options menu showing histogram bins option.................C-30
Figure C-41 Average annualized loss output screen ............................C-31
Figure C-42 Loss area charts buttons ...................................................C-31
Figure C-43 Loss Area Charts tab ........................................................C-31
Figure C-44 File, Edit, and Tools menus .............................................C-32
Figure C-45 Overview tab ....................................................................C-33
Figure C-46 Fragility Specification Details tab ...................................C-34
Figure C-47 Tree view showing: (a) sequential; and (b)

simultaneous fragilities ....................................................C-35
Figure C-48 Example complicated fragility structures consisting of

multiple sub-damage state groups and sub-sub-damage
state groups of several different types .............................C-36
Figure C-49 Basic damage state level ..................................................C-36
Figure C-50 Consequence functions ....................................................C-36
Figure C-51 Adding damage state to different groups .........................C-37
Figure C-52 Combobox for selecting fragility type .............................C-37
Figure C-53 Creating mutually exclusive damage states .....................C-37
Figure C-54 General Info tab on Fragility Specification Details

screen ...............................................................................C-38
Figure C-55 Create New Demand Parameter window .........................C-38
Figure C-56 Notes tab on Fragility Specification Details screen .........C-39
FEMA P-58-2 List of Figures xxi

Second Edition
Figure C-57 Damage State Group Type dropdown menu ................... C-39
Figure C-58 Damage State Group screen ............................................ C-40
Figure C-59 Fragility data display showing: (a) simultaneous

damage state group; and (b) mutually exclusive damage
state group ....................................................................... C-40
Figure C-60 Damage state information screen .................................... C-41
Figure C-61 Consequence Functions screen........................................ C-41
Figure C-62 Repair Cost Consequence tab .......................................... C-42
Figure C-63 Repair Time Consequence tab......................................... C-43
Figure C-64 Other Consequence tab on Fragility Specification

Details screen .................................................................. C-43
Figure C-65 Building Population Manager window ........................... C-44
Figure C-66 Population model data screens ........................................ C-45
Figure C-67 Create Report window..................................................... C-45
xxii List of Figures FEMA P-58-2

Second Edition
List of Tables
Table 2-1 Height Factor Premium Values for Building Level ........... 2-6
Table 2-2 Occupancy Factors ............................................................. 2-9
Table 2-3 Fragility Groups in PACT................................................ 2-13
Table 2-4 Fragility Classification for Reinforced Concrete

Moment Frames ............................................................... 2-20
Table 2-5 Fragility Classification for Low Aspect Ratio

Reinforced Concrete Walls with Two Curtains of
Reinforcement .................................................................. 2-23
Table 2-6 Fragility Classification for Concrete Walls with Return

Flanges ............................................................................. 2-23
Table 2-7 Fragility Classification for Reinforced Masonry Walls ... 2-24
Table 2-8 Suggested Drift Accommodation Ratio δa ....................... 2-29
Table 2-9 Tested Window Pane Sizes .............................................. 2-32
Table 3-1 Default Descriptions and Values for βc.............................. 3-9
Table 3-2 Default Descriptions and Values for βq.............................. 3-9
Table 3-3 Values of coefficient a ..................................................... 3-12
Table 3-4 Default Dispersions for Story Drift.................................. 3-17
Table 3-5 Default Dispersions for Peak Floor Accelerations........... 3-20
Table 3-6 Default Dispersions for Total Peak Floor Velocity ......... 3-23
Table 3-7 Procedures for Estimating Median Story Ratio at

Yield ................................................................................ 3-24
Table 3-8 Drift Ratio Vectors........................................................... 3-27
Table 3-9 Default Dispersions Ground Motion Variability, βgm ...... 3-37
Table 3-10 Segmented Seismic Hazard Values ................................. 3-47
Table 3-11 Segmented Peak Ground Acceleration Values ................ 3-48
Table 4-1 Fragility Group Selections for the Beam/Column

Components ....................................................................... 4-8
FEMA P-58-2 List of Tables xxiii

Second Edition
Table 4-2 Performance Group Quantities for Reinforced Concrete
Elements ............................................................................. 4-9
Table 4-3 Performance Group Quantities for Plumbing

Components...................................................................... 4-15
Table 4-4 Performance Group Quantities for Distributed HVAC

Components...................................................................... 4-16
Table 4-5 USGS Hazard Data, Adjusted for Building Period and
Site Class .......................................................................... 4-22
Table 4-6 Lumped Weight Distribution ........................................... 4-24
Table 4-7 Median Story Drift Ratio Estimates ................................. 4-26
Table 4-8 Median Floor Acceleration Estimates .............................. 4-27
Table 5-1 Collapse Mode Consequences............................................ 5-2
Table 5-2 Hazard Data, Adjusted for Building Period and Site
Class ................................................................................... 5-4
Table 5-3 Intensity Segment Values ................................................... 5-6
Table 5-4 Deaggregation Summary .................................................... 5-7
Table 5-5 Summary of Structural Analysis Results ......................... 5-10
Table 5-6 Residual Drift, Intensity 1 ................................................ 5-12
Table 5-7 Residual Drift, Intensity 3 ................................................ 5-12
Table 6-1 Basic Building Information ................................................ 6-2
Table 6-2 Damage State Descriptions for Plywood Roof

Diaphragm - Transverse ................................................... 6-12
Table 6-3 Damage State Descriptions for Tilt-Up Walls—Out-of-

Plane ................................................................................. 6-18
Table 6-4 Damage State Description for Ledger to Bolt Wall

Attachment—In-Plane...................................................... 6-22
Table 6-5 Median In-Plane Limit State Strengths for Anchor

Bolt ................................................................................... 6-25
Table 6-6 Limit State Strengths Out-Of-Plane Attachment.............. 6-27
Table 6-7 Damage State Description for Wall Attachment (Purlin

Nailing) – Out-of-Plane.................................................... 6-28
Table 7-1 Static Coefficient of Friction for Common Surface

Combinations ..................................................................... 7-2
xxiv List of Tables FEMA P-58-2

Second Edition
Table 7-2 Calculated Fragility Values for Metal Bookcases.............. 7-5
Table 7-3 Calculated Fragility Values for Filing Cabinets ................ 7-5
Table 7-4 Dynamic Coefficient of Friction for Common Surface

Combinations ..................................................................... 7-6
Table 7-5 Dispersion Values for Failure Mode 1 ............................. 7-15
Table 7-6 Suggested Drift Accommodation Ratios for Precast

Cladding and Other Brittle Cladding Systems ................. 7-16
Table 8-1 Repair Cost Estimate for Wall ......................................... 8-11
Table 8-2 Percentage Cost Allocations for Wall .............................. 8-13
Table 8-3 Resulting Total Cost Allocations for Wall....................... 8-13
Table 8-4 Environmental Impact Factors from USEEIO Database

for Select Industrial Sectors ............................................. 8-13
Table 8-5 Adjusted Environmental Impact Factors.......................... 8-14
Table A-1 Structural Steel Elements (B103)...................................... A-1
Table A-2 Reinforced Concrete Elements (B104) ............................. A-3
Table A-3 Masonry Vertical Elements (B105) .................................. A-6
Table A-4 Cold-Formed Steel Structural Elements (B106) ............... A-6
Table A-5 Wood Light Frame Structural Elements (B107) ............... A-6
Table B-1 Exterior Nonstructural Walls (B201) .................................B-1
Table B-2 Exterior Window Systems (B202) .....................................B-2
Table B-3 Roof Elements (B301, B303, B304) ..................................B-6
Table B-4 Partitions (C101) ................................................................B-7
Table B-5 Stairs (C201) ......................................................................B-7
Table B-6 Wall Finishes (C301) .........................................................B-8
Table B-7 Floor Finishes, Raised Access Floors, and Floor

Flooding (C302).................................................................B-8
Table B-8 Ceilings and Ceiling Lighting (C303)................................B-9
Table B-9 Elevators and Lifts (D101).................................................B-9
Table B-10 Domestic Water Distribution (D202)...............................B-10
Table B-11 Sanitary Waste Piping System (D203) ............................B-11
FEMA P-58-2 List of Tables xxv

Second Edition
Table B-12 Chilled Water Piping (D205) .......................................... B-11
Table B-13 Steam Piping (D206) ....................................................... B-12
Table B-14 Chillers, Cooling Towers, and Compressors (D303) ...... B-14
Table B-15 Distribution Systems (D304)........................................... B-15
Table B-16 Packaged Air Handling Units (D305) ............................. B-16
Table B-17 Control Panels and Instrumentation (D306).................... B-16
Table B-18 Fire Protection (D401) .................................................... B-17
Table B-19 Electrical Service and Distribution (D501) ..................... B-18
Table B-20 Other Electrical Systems (D509)..................................... B-19
Table B-21 Movable Furnishing (E202) ............................................ B-20
Table B-22 Special Structures (F101) ................................................ B-22
Table C-1 PACT Third Party Controls............................................. C-48
Table D-1 Summation Presentations in the Component Summary

Matrix ................................................................................ D-4
xxvi List of Tables FEMA P-58-2

Second Edition
Chapter 1
Introduction
This Second Edition Implementation Guide provides guidance on

Second Edition Update
implementation and use of the methodology set forth in FEMA P-58-1,
Significant updates in the
Seismic Performance Assessment of Buildings, Volume 1 – Methodology, Second Edition are identified
Second Edition, (FEMA, 2018a) to assess the seismic performance of with green margin boxes
individual buildings based on their unique site, structural, nonstructural, and throughout the document.
occupancy characteristics, expressed in terms of the probability of incurring
casualties, repair and replacement costs, repair time, environmental impacts,
and unsafe placarding. This Implementation Guide contains examples
illustrating the performance assessment process, including selected
calculation and data generation procedures, as well user manuals for selected
electronic tools provided in FEMA P-58-3, Seismic Performance Assessment
of Buildings, Volume 3 – Supporting Electronic Materials and Background
Documentation, (FEMA, 2018b).
This Implementation Guide, and its companion Volumes 1 and 3, were first
published in 2012. Shortly after publication, a second phase of work was
initiated to update and complete the FEMA P-58 series of products, including
exercising the methodology to assess the performance of code-conforming
buildings, developing products for communicating seismic performance to
stakeholders, and developing design guidance to assist decision-makers in
choosing between seismic design criteria and making seismic design
decisions. Technical improvements and updates to the methodology were
developed, as necessary, to take advantage of the latest research and to bring
assessment results into better alignment with expectations based on
performance observed in past earthquakes.
1.1 Purpose and Scope
The general methodology presented in Volume 1 can be applied to seismic Second Edition Update
performance assessments of any building type, regardless of age, All chapters have been
construction or occupancy type. Many different means of implementing this updated to present screenshots
general methodology are possible. During the development of the general from PACT version 3.1.2.
methodology, the project development team found it necessary to develop a
tool to implement the methodology, which is provided in Volume 3 as the
Performance Assessment Calculation Tool (PACT). The problem
FEMA P-58-2 1: Introduction 1-1

Second Edition
formulation and execution process outlined in this Implementation Guide is
sequenced to correspond to the input cues provided by PACT.
This Implementation Guide provides a detailed road map for users to follow
in applying the FEMA P-58 methodology to the site, structural,
nonstructural, and occupancy characteristics of their individual building to
obtain intensity-based, scenario-based, or time-based earthquake
performance assessments. Implementation requires basic data on the
vulnerability of structural and nonstructural components to damage
(fragility), as well as estimates of potential casualties, repair costs, repair
times, and environmental impacts (consequences) associated with this
damage. This document also provides examples for calculating user-defined
structural and nonstructural fragilities, and developing consequence
functions.
1.2 Limitations
This document tracks with the provisions of Volume 1, but does not
substantially duplicate its narratives, definitions, equations, or other
provisions. Readers are cautioned to use this document in conjunction with
Volume 1, and not to rely on Volume 2 alone for guidance on executing the
methodology.
1.3 The Performance Assessment Process
Figure 1-1 illustrates the basic steps in the performance assessment process.
Volume 1 describes each of these steps and how they relate to the overall
performance assessment. Three of these steps, assembling building
performance model, defining earthquake hazards, and analyzing building
response, in addition to developing collapse fragility, when necessary, are
performed directly by the user. This Implementation Guide presents
illustrations of these steps with examples ranging from simple to more
complex.
Before implementing the assessment process, however, the user should select
the assessment type, analysis method, and performance measures that will
provide the desired output.
1.3.1 Assessment Types
Volume 1 defines three different seismic performance assessment types.

Each assessment type requires different input and utilizes different
procedures.
1-2 1: Introduction FEMA P-58-2

Second Edition
Intensity-based assessments evaluate a building’s probable performance
assuming that it is subjected to a specified earthquake shaking intensity.
Shaking intensity is defined by 5% damped, elastic acceleration response
spectra. This type of assessment can be used to assess a building’s
performance in the event of design earthquake shaking consistent with a
building code response spectrum, or to assess performance for shaking
intensity represented by any other response spectrum.
Scenario-based assessments evaluate a building’s probable performance
assuming that it is subjected to a specified earthquake scenario consisting of
a specific magnitude earthquake occurring at a specific location relative to
the building site. Scenario assessments may be useful for buildings located
close to one or more known active faults. This type of assessment can be
used to assess a building’s performance in the event of a historic earthquake
on these faults is repeated, or a future projected earthquake occurs.
Time-based assessments evaluate a building’s probable performance over a
specified period of time (e.g., 1-year, 30-years, or 50-years) considering all
earthquakes that could occur in that time period, and the probability of
occurrence associated with each earthquake. Time-based assessments
consider uncertainty in the magnitude and location of future earthquakes as
well as the intensity of motion resulting from these earthquakes.
1.3.2 Performance Measures
The seismic performance of a building is expressed as the probable damage

and resulting consequences of a building’s response to earthquake shaking.
The consequences, or impacts, resulting from earthquake damage considered
in this methodology are:
• Casualties. Loss of life or serious injury requiring hospitalization,
occurring within the building envelope.
• Repair cost. The cost, in present dollars, necessary to restore a building
to its pre-earthquake condition, or in the case of total loss, to replace the
building with a new structure of similar construction.
• Repair time. The time necessary to repair a damaged building to its pre-
earthquake condition. The repair time estimates do not include the time
needed to design, permit, and fund the repair project, the time to procure
long lead time components that require replacement, or the time to
mobilize the contractors and finalize the project.
• Unsafe placarding. A post-earthquake inspection rating that deems a
building or portion of a building damaged to the point that entry, use, or
occupancy poses immediate risk to safety.

Second Edition
Second Edition Update • Environmental Impacts. Environmental impacts, in terms of carbon
Consideration of emissions and embodied energy, to restore a building to its pre-
environmental impacts is earthquake condition, or in the case of total loss, to replace the building
included in accordance with the with a new structure of similar construction.
updated Methodology.
Figure 1-1 Performance assessment process.
Each performance measure requires basic information about the building’s

characteristics.
1.3.3 Analysis Methods
The methodology provides users with a range of options for generating the
above assessments. Options include use of a simplified analytical estimation
of building response or suites of detailed nonlinear response history analyses.
Building assets at risk can be defined by occupancy-dependent typical
(normative) quantities or building-specific surveys. The performance
characteristics of these at-risk assets can be represented by provided

Second Edition
relationships for component fragility and consequences, or component-
specific fragility and consequence functions can be developed and used.
Each building performance assessment can use either of the analysis
approaches or any combination of the options to define component fragility
and consequence characteristics.
The simplest application of the methodology includes use of the simplified

analysis method to estimate building response and the selection of provided,
occupancy-dependent fragility and consequence functions for the building
assets at risk. This streamlined approach may be most appropriate for
circumstances where information about building characteristics is limited, as
is typical during preliminary design of new buildings or in the initial
evaluation stages for existing buildings. In general, the more streamlined the
approach, the more limitations there are in the methodology’s ability to
characterize performance and the larger the inherent uncertainty in the
performance assessments.
1.4 Updates to Implementation Tools
Originally developed in 2012, this Implementation Guide provided example

applications of the FEMA P-58 methodology and tools, using various online
Most tools provided as part of
and external resources available at the time. Phase 2 updates to the fragility the FEMA P-58 series of
database include new fragility and consequence data, and editorial changes to products have been updated
fragilities such as revised numbering and addition or deletion of selected since the original publication in
fragilities. Changes to fragility and consequence information are 2012.
documented in FEMA P-58/BD-3.7.19 (FEMA, 2018e). Because of changes
to the fragility database, a building model developed using a previous version
of PACT may cause an error. This Second Edition Implementation Guide
reflects updates that have been made to the FEMA P-58 methodology, tools,
and underlying data. References to online tools and outside resources have
been updated; however, the ongoing availability of resources developed
external to the project (e.g., online tools for constructing hazard curves) is
subject to change, and current versions may not be comprehensively covered
herein. Other resources may also become available.
Volume 3 – Supporting Electronic Materials and Background

Documentation, Third Edition includes the following electronic products to
assist in the implementation and use of the methodology:
• Performance Assessment Calculation Tool (PACT). PACT is an
electronic calculation tool, and repository of fragility and consequence
data, that performs the probabilistic calculations and accumulation of
losses described in the methodology. It includes a series of utilities used
to specify building properties and update or modify fragility and

Second Edition
consequence information in the referenced databases. An executable file
(.exe) is provided to facilitate installation of the tool. Appendix C of this
Volume presents a User Manual for PACT.
• Provided Fragility Data. This folder contains the following products to
aid management, maintenance of all provided fragility and consequence
data outside of PACT and assist in the development of consequences for
custom fragility specifications:
o Fragility Database is an Excel workbook that is used to review and
maintain all provided fragility and consequence data outside of
PACT. This database has been updated for the Second Edition
release and includes environmental impact consequence information.
o Fragility Specification is a PDF file displaying the contents of the
fragility database. Each fragility specification contains fragility and
consequence data for the component of interest, in a one-page
format. Damage states are illustrated with photos of representative
damage, when available.
o PACT Input Matrix is an Excel workbook that contains the Fragility
Database in a format that is required for upload into PACT.
o Consequence Estimation Summary is an Excel workbook that
provides the basis for provided consequence data, and can be used to
assist in estimating consequences for custom fragility specifications.
• Normative Quantity Estimation Tool is an Excel workbook designed to
assist in estimating the type and quantity of nonstructural components
typically present in buildings of a given occupancy and size. Appendix
D of this Volume presents a User Manual for this tool.
• Performance Estimation Tool (PET). PET is an Excel workbook that
Second Edition Update utilizes assessment results from the application of the FEMA P-58
This is a newly developed tool methodology to a group of archetypical building representative of
and is described in FEMA P-58.
structures conforming to the seismic design requirements of the current
Volume 5 and FEMA P-58,
Volume 6. building code. The tool can be used as a design aid to obtain an initial
estimate of required design strength and drift ratio necessary to achieve a
desired performance objective.
• Static Pushover to Incremental Dynamic Analysis (SPO2IDA).
SPO2IDA is an Excel workbook application that was originally
developed by Vamvatsikos and Cornell (2006). This tool uses empirical
relationships from a large database of incremental dynamic analysis
results to convert static pushover curves into probability distributions for
building collapse as function of ground shaking intensity.

Second Edition
• Collapse Fragility Tool. The Collapse Fragility Tool is an Excel
workbook application that fits a lognormal distribution to collapse
statistics obtained from a series of nonlinear dynamic analyses at
different ground motion intensity levels.
With the exception of the Performance Estimation Tool (PET), this
Implementation Guide illustrates the use of all tools listed above. Use of the
Performance Estimation Tool for viewing assessment results is described in
FEMA P-58-5, Guidelines for Performance-Based Seismic Design of
Buildings (FEMA, 2018c), and use of the tool as a design aid is presented in
FEMA P-58-6, Guidelines for Performance-Based Seismic Design of
Buildings (FEMA, 2018d).
1.5 Organization and Content
This Implementation Guide is organized into the following chapters:

Chapter 2 provides a detailed description of the steps used to develop a
building performance model consisting of basic building information,
structural and nonstructural fragility specifications, consequence functions,
and the building’s collapse fragility and residual drift characteristics.
Chapter 3 provides a detailed description of the steps required to execute
intensity, scenario, and time-based assessments.
Chapter 4 provides a step-by-step application of the methodology for an
intensity-based assessment that uses simplified analysis for estimating
building response and relies on provided fragility and consequence functions
to characterize component vulnerability.
Chapter 5 provides a step-by-step application of the methodology for a time-
based assessment that uses nonlinear response history analyses for estimating
building response.
Chapter 6 illustrates development of structural component fragility functions
by calculation to address unique circumstances or to supplement the provided
fragility functions.
Chapter 7 illustrates development of nonstructural component fragility
functions by calculation to address unique circumstances or to supplement
the provided fragility functions.
Chapter 8 provides guidance on the development of component consequence
functions to accompany user-defined fragility functions and to supplement or
modify provided consequence functions.

Second Edition
Appendix A lists structural components for which fragility and consequence
data are provided.
Appendix B lists nonstructural components and contents for which fragility

and consequence data are provided.
Appendix C provides instructions for the use of PACT.
Appendix D describes the Normative Quantity Estimation Tool that is

designed to assist in estimating the type and quantity of nonstructural
components typically present in buildings of a given occupancy and size.
A Glossary and list of Symbols, providing definitions of key terminology and

notation used in the methodology, along with a list of References, are
provided at the end of this report.

Second Edition
Chapter 2
Building Performance Model
2.1 Introduction
This chapter provides guidance for assembling the building performance

model and is organized to provide direct references to the appropriate input
tabs the user will find in the Performance Assessment Calculation Tool
(PACT) provided in Volume 3. Figure 2-1 illustrates the format of PACT
screenshots provided throughout this document.
Figure 2-1 Format of PACT screenshots used, indicating the tab title location.
The building performance model provides a systematic and quantitative

description of the building assets at risk of damage from earthquake ground
shaking effects. This model includes basic building characteristics (Section
2.2), an overview of fragility specifications and performance groups (Section
2.3), an organized description of the structural (Section 2.4) and
nonstructural (Section 2.5) components, the location of these assets within
the building, an expression of their damageability and the consequences of
this damage, as well as a collapse fragility function (Section 2.6) expressing
the probability of building collapse, and a residual drift function (Section 2.7)
which is a measure of the building’s repairability.
2.2 Building Characteristics
2.2.1 Project Information
Assembly of the building performance model within PACT begins with the
Project Info tab, shown in Figure 2-2. This is used to input basic project
information used to identify the analysis files and results including Project
ID, Building Description, Client, and Engineer fields.
PACT uses the Region Cost Multiplier and Date Cost Multiplier fields to
adjust provided component repair cost consequence functions to appropriate
present values. The provided consequence functions reflect repair costs
appropriate to Northern California in 2011. Users can address escalation and
regional cost variation through cost multiplier input using any suitable cost
FEMA P-58-2 2: Building Performance Model 2-1

Second Edition
index system. The cost modifier applies only to the cost data provided in the
PACT consequence function database. If the user provides building-specific
consequence cost data, these data should directly reflect the cost index
associated with the building’s locality and the assessment time, before
insertion into the performance model. If the user inputs independently
derived consequence functions for all of the performance groups included in
the assessment, the region and date cost multipliers should be input as a
value of 1.0.
Figure 2-2 PACT Project Information tab.
PACT uses the Solver Random Seed Value input to initiate all internally
programmed sequences of random number generation utilized in
performance assessment. If a Solver Random Seed Value of zero is used,
PACT will randomly seed each generation sequence. This will result in
different values for performance assessment results each time the same
problem is executed even if there are no changes to the input. While the
results of these assessments can be expected to be similar, users should input
a single digit non-zero integer to avoid seeing anomalous changes in
predicted performance when multiple evaluations of the same building are
performed. Note that if a sufficiently large number of realizations is used,
this effect is negligible.
2.2.2 Building Information
Figure 2-3 illustrates the Building Info tab used to enter basic building data.
The red exclamation symbol, , appears whenever the value of an input is
outside a reasonable range, indicating a probable input error. This occurs for
many fields before data are entered, as a warning that entry is required.
2-2 2: Building Performance Model FEMA P-58-2

Second Edition
Figure 2-3 PACT Building Information tab.
PACT uses the Number of Stories input as a basic index of the number of
demand parameters, performance groups, and calculations to be performed.
A story is defined as the building volume that extends from the top of slab or
other flooring at one floor level, to the top of slab or flooring at the next
level. It includes all things that are mounted on or above the lower floor and
which are present beneath the top of the higher level, such as the framing
supporting the higher floor or roof. The input value should include all stories
that have vulnerable components and which are to be included in the
performance assessment. If basements are present, and have vulnerable
structural or nonstructural components or occupants susceptible to injury,
these should be included as stories. Similarly, penthouses with vulnerable
components or occupants should be included as stories.
PACT defines the number of floors based on the Number of Stories input,
where floor identifies all those components present within a story that are
located on top of the surface of the identified floor, and beneath the top
surface of the floor above. Thus, the first floor includes fragility groups for
framing that supports (and is beneath) the second floor; as well as
components that are supported on the first floor or suspended from the
second floor.
Figure 2-4 illustrates the PACT definitions of floor and story numbers and
floor and story heights. Floor numbering should be initialized at the lowest
story housing damageable components. In the example of Figure 2-4, it has
been assumed that basement stories of this building do not contain
damageable components.

Second Edition
Figure 2-4 Definition of floor and story numbers and floor and story
heights.
Core and Shell Replacement Cost should reflect a best estimate of the cost
to replace all building core and shell items, including an allowance for
building demolition and site clearance. Core and shell components include
the basic building structure and cladding and all nonstructural components
that are not typically provided by the tenants, such as elevators, stairs, toilet
rooms, and basic electrical and mechanical service. Total Replacement
Cost input includes the cost of core and shell constituents plus the cost to
replace tenant improvements and contents. Tenant improvements commonly
include office partitions, ceilings, light fixtures, HVAC, and electrical
distribution within occupied spaces except in common areas, such as lobbies
or central plants.
The Maximum Workers per Square Foot input is used to calculate repair
time. Values for this parameter should range from 0.0005 (one worker per
2,000 square feet) to 0.004 (1 worker per 250 square feet). During an actual
repair project there can be considerable fluctuation in the number of workers
per square foot of floor area. PACT provides a default setting of 0.001
which corresponds to one worker per 1000 square feet of floor area. Users
should generally execute their assessment with this default value, but can use
denser values when the building occupancy is such that owners will be
willing to bear the cost associated with more rapid repair schedules. Use of
denser values will typically imply that the building will not be occupied
during repair, even if damage is not so great as to warrant this from other
perspectives. Users can perform multiple assessments using different values
of this parameter to understand how it affects the potential repair times.

Second Edition
The Total Loss Threshold is the ratio of repair cost to replacement cost at
which a decision will likely be made to replace the building rather than repair
it. FEMA uses a value of 0.5 for this loss ratio when determining whether
post-earthquake repair should be funded. PACT uses a default value of 1.0
to maximize the amount of assessment information that will be obtained in
an assessment. Volume 1 suggests that when repair costs exceed 40% of
replacement costs, many owners will choose to demolish the existing
building and replace it with a new one.
The Carbon Emissions Replacement is the total carbon impact of replacing

the building. This is calculated by multiplying the Total Replacement Cost Environmental impact
by the appropriate construction sector impacts from the United States considerations were added.
Environmentally Extended Input-Output (USEEIO) database (available at:
https://catalog.data.gov/dataset/useeio-v1-1-matrices, last accessed
September 28, 2018). For example, if the total replacement cost of the
building is $1,000,000 and the global warming potential (labeled in the
dataset as “impact potential/gcc/kg co2 eq”) for the sector
“2332a0/commercial structures, including farm structures/us” is 0.414 kg
CO2e per $1 (reference year 2013), and the inflation factor from 2013 to
2011 is assumed to be 0.88 (based on the ratio of gross domestic product in
the construction sector in 2013 to 2011from U.S. Bureau of Economic
Analysis, available at https://www.bea.gov/, last accessed September 28,
2018), the resulting carbon emissions impact of building replacement would
be (0.414 kg CO2e/$) × ($1,000,000) / (0.88) = 470,454 kg CO2e.
The Embodied Energy Replacement is the total embodied energy impact of

replacing the building. This is calculated by multiplying the Total
Replacement Cost by the appropriate construction sector impacts from the
USEEIO dataset. For example, if the total replacement cost of the building is
$1,000,000 and the energy impact (labeled in the dataset as “resource
use/enrg/mj”) for the sector “2332a0/commercial structures, including farm
structures/us” is 5.729 MJ per $1 (reference year 2013), and the deflation
factor from 2013 to 2011 is assumed to be 0.88, the resulting carbon
emissions impact of building replacement would be (5.729 MJ/$) ×
($1,000,000) / (0.88) = 6,510,227 MJ.
The information placed in the Most Typical Default section is used to

populate a matrix of values for each floor, identified in the lower portion of
the tab. Users can change the values in the matrix by entering other values
directly into the individual cells. Floor Area input is used to estimate the
number of casualties during an earthquake realization. Story Height is not
used within PACT, but input of a reasonable value is required.

Second Edition
In this matrix, the Height Factor is used to reflect increases in repair cost
attributable to:
• Loss of efficiency due to added travel time to get to damaged
components on upper levels
• Material and tool loading and staging, including added cost for hoisting,
elevator loading, pumping, and disposal
• Access costs related to cutting openings or penetrations, removing
windows, loading and moving material to installation areas
• Scaffolding or rigging, including fall protection and protection to lower
areas
Minor cost adjustments may be appropriate for simple interior repairs, where
the impact is a minor loss of efficiency for worker elevator travel.
Significant adjustments may be necessary for exterior cladding repairs on a
high-rise building which could require significant scaffolding for an
otherwise low cost repair item. Both of these extremes are unlikely to
produce a significant total repair cost error, since they are typically combined
with much less sensitive work items. While there can be a significant range
of height premium costs for individual items, the aggregated modifications
will be a relatively small increase for most cases. Table 2-1 presents
suggested Height Factor premiums.
Table 2-1 Height Factor Premium Values for Building Level

Building Level Height Factor
Below grade levels and floors 1-4 1.00
Floors 5-10 1.08
Floor 11 and higher 1.16
The Hazmat Factor field is used to reflect the variable hazardous material
premiums. For new buildings, hazardous materials issues are generally a
function of occupancy. Healthcare and research facilities typically contain
some amount of hazardous material to support their operations. In many
older buildings hazardous materials have been removed as part of recent
tenant improvement and building modernization projects. Unless specific
information is known, a Hazmat Factor of 1.0 is recommended. It is
reasonable to expect this factor to range from 1.00 for modern buildings
without significant hazardous material content to 1.20 for buildings that
contain significant amounts of hazardous material including lead-based paint
and asbestos. Determination of Hazmat Factors should consider the
following:

Second Edition
• Friable asbestos insulation. This material is typically found in pipe
lagging in buildings constructed in 1979 or earlier. Friable lagging could
be present on all pipes in earlier construction. It was more commonly
applied only to boilers, bends and tees, and irregularly shaped elements
in later construction. While the cost for removal or abatement is high,
the extent is typically limited. A recommended mid-point range for
replacement or abatement is $10,000 per boiler or furnace, $50 per lineal
foot of piping, or alternatively $2 to $3 per square foot of overall gross
building area.
• Friable asbestos fireproofing. This material is typically found in
sprayed-on structural steel fireproofing in buildings constructed from the
1940s through the 1970s. Recommended midpoint cost for removal or
abatement is $20,000 per location of steel repair, $15 to $25 per square
foot of overall gross building area (based on 100% abatement).
• Non-friable asbestos cement products. These products are typically
found in flooring, siding or roofing of buildings but may be present in
limited quantities in fire protection. The cost of removal or abatement is
unlikely to be a significant factor in the repair costs for low to
moderately damaged buildings.
• Asbestos containing materials. These materials are common in
buildings constructed from the 1940s through the 1970s and could be
present in earlier buildings. Asbestos containing materials are pervasive
and can be found in ceiling tiles, spray acoustic treatment, flooring
materials, floor mastic, roofing materials, caulk, and drywall taping
compound. A recommended cost for removal or abatement is
approximately $10 to $30 per square foot of overall gross building area.
• Polychlorinated biphenols. This material is commonly found in
electrical equipment that includes fluorescent light ballasts and caulk.
Polychlorinated biphenols (PCB) could be present in buildings
constructed before 1991, but are more likely present in construction
preceding 1980. In many cases, PCB ballasts have been replaced during
energy retrofits, so older buildings may not necessarily contain PCB
materials.
The following midpoint costs for removal or abatement are suggested:

o $100 per ballast for replacement in damaged fixtures. Ballasts in
undamaged fixtures should not need to be replaced following an
earthquake.

Second Edition
o Premium cost of $20,000 per item (in addition to equipment
replacement) of other electrical equipment (transformers,
switchgear).
o Caulk containing PCB has only recently been recognized as a
significant environmental problem. Repair protocols are not well
established. PCB can migrate from the caulk into surrounding
porous materials. This could necessitate not only caulk removal, but
also replacement of surrounding materials, such as masonry,
concrete, or wood. The midpoint costs suggested are for recaulking
only and for abatement or removal is $15 per lineal foot or $3 to $5
per square foot of overall gross building area.
• Lead based paint. In buildings constructed prior to 1980 painted
surfaces, particularly wood and steel, are likely to contain lead based
paint. Midpoint cost for removal or abatement is $3 per square foot of
painted area, or $2 to $5 per square foot for multi-unit residential and
education facilities, and $5 to $8 per square foot of overall building area
for complete structural steel abatement. For structural steel, $5,000 per
repair location may be used.
There are many other hazardous materials and conditions that could be
encountered in repair efforts. In the larger context of seismic repair,
typically these treatments are inconsequential from a cost perspective. These
materials can include mercury in light fixtures and switching devices
(thermostats), dry-cleaning fluids, gasoline, diesel or other hydrocarbons in
generators or fuel storage tanks, chemicals from process or manufacturing,
and radioactive materials (particularly in healthcare or laboratory
equipment).
The Occupancy Factor reflects the added cost of working around ongoing
building operations, equipment, and the collateral protections required for
some construction characteristics. The repair premium is significantly more
pronounced in occupied buildings than in vacated buildings. However, the
level of collateral protection required even in an unoccupied building may be
significant. Table 2-2 lists recommended repair premiums for typical
occupied and unoccupied buildings. PACT does not presently distinguish
between occupied and unoccupied repair conditions.
For time-based assessments, “unoccupied” factors should be used. For

scenario- and intensity-based assessments, “occupied” factors should be used
for low intensity shaking since repair of minor damage is likely to be done
while the building is occupied. For moderate to high intensity shaking, the

Second Edition
use of the “unoccupied” factors is recommended, as buildings are more likely
to be damaged beyond a level at which repair during occupancy is practical.
Table 2-2 Occupancy Factors

Occupancy Factor
Occupancy Category Unoccupied Occupied
Commercial Office 1.0 1.2
Research 1.4 1.8
Healthcare 1.5 2.0
Education K-12 1.0 1.1
Multi-Unit Residential 1.1 1.2
Retail 1.2 1.3
Warehouse 1.1 1.1
Hospitality 1.1 1.3
2.2.3 Population Model
To assess casualties, users must define the population model, i.e., the
distribution of occupants within the building at various times of day. In
PACT, it is possible to use one of the provided building population models or
to develop and input building-specific models. Eight population models are
provided in PACT corresponding to typical commercial office, education
K-12 (elementary, middle, high school), healthcare, hospitality, multi-unit
residential, research, retail, and warehouse occupancies.
Users can assign separate population models to several fractions of each floor
level. Each population model includes the hourly distribution of people per
1,000 square feet for weekdays or weekends and can be adjusted to include
further variation by month. Provided population models can be used directly
or be modified to reflect the unique occupancy characteristics of a specific
building, if known. Figure 2-5 illustrates the commercial office occupancy
provided in PACT. Building-specific population information can be
developed and modified using the Population Manager utility, as illustrated
in Figure 2-6. The provided population models for each occupancy category
can be viewed by selecting from the column on the left and modified by
substituting data on the right side of the window. New occupancy models
can be created by overwriting the existing ones or creating new names for the
occupancy.

Second Edition
Figure 2-5 PACT Population tab showing commercial office occupancy.
Figure 2-6 PACT Population Manager utility.

Second Edition
2.3 Fragility Specifications and Performance Groups
Following the definition of the general building characteristics, it is

necessary to define the quantity, vulnerability, and distribution of
damageable components and contents. PACT organizes this process into two
parts: (1) identification of required fragility specifications for each floor, and
(2) identification of the quantity of components in each performance group at
each floor.
The fragility specification includes a description of the demand parameter

that predicts damage, the types of damage that can occur, fragility functions,
which indicate the probability of incurring each damage state as a function of
demand, and consequence functions, which indicate the probable values of
loss that will occur as a result of each damage state.
Each fragility function specifies damage state probabilities for a single

demand parameter. Typically, peak story drift ratio or peak floor
acceleration parameters are used to determine if a component is damaged.
The demand parameter can have a specific orientation with respect to the
component (directional demand) or the demand can be non-directional. For
example, wall elements will typically be damaged by story drift within their
plane, where suspended ceiling systems are susceptible to damage from floor
acceleration independent of horizontal direction. Users should be aware that
floor acceleration important to the damageability of a component, such as
ceiling mounted nonstructural components, may be associated with the floor
level above the level under consideration. Unanchored nonstructural
components use peak total floor velocity as the demand parameter.
User-defined, building-specific fragility functions can use any appropriate

demand parameter. For example, a user could decide that beam-column joint
plastic rotation is the best predictor of damage for a particular type of
structural component and could develop fragility functions based on that
parameter. If demand parameters other than story drift ratio, floor
acceleration, or velocity are used, the structural analysis used to estimate
building response must provide values for the selected demand parameters.
The quantity and distribution of damageable components conforming to each

selected fragility specification is entered into PACT through the definition of
performance groups at each floor level. A performance group is a set of
components described by a single fragility group that will experience the
same demand. Performance groups are ordered by the direction of
application of their common demand parameter.
In PACT, components can be selected and distributed across the building’s

floors to create a complete representation of the damageable building. To

Second Edition
determine the quantities of vulnerable nonstructural components and contents
within a building, the Normative Quantity Estimation Tool provided in
Volume 3 can be utilized.
Users should be aware that the list of component fragility specifications

provided with PACT does not include all vulnerable building components
that may be present in a building. Users must carefully identify damageable
building features not provided in PACT. User-defined fragility and
consequence functions can be developed following the guidance provided in
Chapters 6, 7, and 8.
2.3.1 Fragility Specifications Provided in PACT
PACT references a data set of more than 700 individual fragility

specifications, containing both structural and nonstructural components,
contained in the Fragility Database provided as part of Volume 3. Table 2-3
provides a list of fragility groups identified in PACT where fragility groups
with fragility specifications provided in PACT are indicated with bold
lettering. Figure 2-7 illustrates the PACT window used to select fragility
groups present in a performance model.
Each fragility group is identified by a unique identification code based on

recommendations contained in NISTIR 6389 Report, UNIFORMAT II
Elemental Classification for Building Specifications, Cost Estimating and
Cost Analysis, (NIST, 1999) where codes take the form: A1234.567. The
first letter in the classification system indicates the overall component
category. The first two numbers provides the next categorization. For
instance, B10 represents superstructure components while B20 represents
exterior enclosures. The next two numbers identify a unique component.
For example, the classification for reinforced concrete shear walls is B1044.
The identifiers after the decimal provide variations of the basic component
and are used to identify different configurations, conditions of installation,
material quantities, demand levels, and other attributes.
Note that many building components are inherently rugged and not subject to
significant damage for credible levels of demand. Rugged components
include such things as plumbing fixtures, electrical raceways, wall-mounted
panels, and some components of gravity framing systems. Volume 1,
Appendix I provides an itemized listing of common building components
that can generally be considered rugged. These components are not typically
susceptible to damage as a result of a building’s earthquake response and
need not be included in the building performance model.

Second Edition
Table 2-3 Fragility Groups in PACT
Component Category Component Name
A Substructure
A10 Foundations
A101 Standard Foundations
A102 Special Foundations
A103 Slab on Grade
A20 Basement Construction
A202 Basement Walls
B Shell
B10 Super Structure
B101 Floor Construction
B102 Roof Construction
B103 Structural Steel Elements
B104 Reinforced Concrete Elements
B105 Masonry Vertical Elements
B106 Cold-Formed Steel Structural Elements
B107 Wood Light Frame Structural Elements
B20 Exterior Enclosure
B201 Exterior Nonstructural Walls
B202 Exterior Window Systems
B203 Exterior Doors
B30 Roof Elements
B301 Roof Coverings
B302 Roof Openings
B303 Masonry Chimney (residential)
B304 Masonry Parapet
B40 Exterior Finishes
B401 Exterior Finishes
C Interiors
C10 Interior Construction
C101 Partitions
C102 Interior Doors
C20 Stairs
C201 Stairs

Second Edition
Table 2-3 Fragility Groups in PACT (continued)
C30 Interior Finishes
C301 Wall Finishes
C302 Floor Finishes, Raised Access Floors, and Floor Flooding
C303 Ceilings and Ceiling Lighting
D Services
D10 Conveying
D101 Elevators and Lifts
D102 Escalators and Moving Walks
D103 Other Conveying Systems
D20 Plumbing
D202 Domestic Water Distribution including hot water heaters
D203 Sanitary Waste Piping System
D204 Rain Water Drainage Piping System
D205 Chilled Water Piping
D206 Steam Piping
D209 Other Piping
D30 HVAC
D301 Energy Supply
D302 Heat Generating Systems (furnaces and boilers)
D303 Chillers, Cooling Towers, and Compressors
D304 Distribution Systems including Fans, Drops and Diffusers, and VAV Boxes
D305 Packaged Air Handling Units
D306 Control Panels and Instrumentation
D309 Other HVAC Systems and Equipment
D40 Fire Protection
D401 Sprinkler Horizontal Distribution including Risers and Drops
D402 Standpipes
D403 Fire Protection Specialties
D409 Other Fire Protection Systems
D50 Electrical
D501 Electrical Service & Distribution
D502 Lighting and Branch Wiring
D503 Communications and security
D509 Other Electrical Systems including Generators, Battery racks, and Chargers

Second Edition
Table 2-3 Fragility Groups in PACT (continued)
E Equipment & Furnishings
E10 Equipment
E101 Commercial Equipment
E102 Institutional Equipment
E103 Vehicular Equipment
E109 Other Equipment
E20 Furnishings
E201 Fixed Furnishings
E202 Movable Furnishing, including Office Furniture and Equipment
F Special Construction and Demolition
F10 Special Construction
F101 Special Structures including Storage Racks
F103 Special Construction Systems
F104 Special Facilities
F105 Special Controls and Instrumentation
Figure 2-7 PACT Component Fragilities tab showing fragility specification selection.

Second Edition
2.3.2 Demand Directionality
Users must identify whether components are sensitive to a direction of

shaking. Vertical elements of the seismic-force-resisting system, including
frames and walls are generally susceptible to damage from in-plane shaking.
Distributed slab column systems can be damaged by drifts along either
building axis. These components should be treated as possessing non-
directional fragility specifications.
Many nonstructural components are assumed to be vulnerable to damage

independent of shaking direction. Damage to nonstructural components may
be a function of story drift ratio, floor velocity, or floor acceleration. While
some components may actually be damaged by a combination of drift, floor
acceleration, and velocity, the methodology only permits use of a single
demand parameter for damage prediction.
2.4 Identification of Vulnerable Structural Components

and Systems
In addition to elements of a structure’s intended seismic-force-resisting

system, damageable structural components should include vulnerable
structural elements normally considered part of the vertical-force-resisting
system.
Appendix A lists the structural fragility specifications provided in the

Fragility Database referenced by PACT, organized into five component
categories, spanning the range of structural materials from structural steel to
wood light frame systems.
Each component category contains multiple fragility specifications, which

reflect a range of configuration and construction variations affecting seismic
performance characteristics, such as special and ordinary masonry walls.
PACT makes further differentiation among some components with identical
damageability but with significant differences in damage state consequences,
such as repair costs, owing to such characteristics as size and finish.
Users should select the most appropriate set of structural fragility

specifications using the descriptions provided in PACT for each principal
building direction. The PACT Fragility Specification Manager utility can
also facilitate the process of identifying appropriate fragility specifications.
This utility presents general information on the provided fragility
specifications and includes a summary of the demand parameter and
characterization of its applicability to demand directionality and also permits
modification of fragility specifications and definition of new ones.

Second Edition
Fragility specifications for horizontal floor or roof construction components
or systems acting as diaphragms are not provided. However, fragility
information for some diaphragm components that act as part of the vertical-
force-resisting system such as slab-column joints and steel beam shear tab
connections, is included, as shown in Appendix A. For buildings with
damageable diaphragms, the user must generate fragility and consequence
functions using either laboratory data or calculations. For further guidance
see Chapter 6 and Chapter 8.
The following sections discuss the general categories of superstructure

components included for each of the provided fragility specifications.
Appendix A provides a summary of construction design characteristics,
information on damage mode, design configuration, and other features,
organized by the fragility classification numbers.
2.4.1 Structural Steel Elements (B103)
This fragility classification (B103) is summarized in Table A-1 and includes

miscellaneous components and structural steel braced frame and structural
steel moment frame systems.
Miscellaneous components include gravity beam connections that feature a

bolted shear tab (B1031.001) and column base plates, differentiated by
column weight as follows:
• B1031.011a: Columns < 150 plf
• B1031.011b: 150 plf < Columns < 300 plf
• B1031.011c: Columns > 300 plf
Column base plates should be included in the performance model when they
are part of the lateral force-resisting system and are designed to provide
Consideration of column base
rotational restraint at the base. plates and splices have been
added.
Also within this component category are three fragility specifications for
welded column splices, again differentiated by column weight:
• B1031.021a: Columns ≤ 150 plf
• B1031.021b: 150 plf < Columns ≤ 300 plf
• B1031.021c: Columns > 300 plf
Only column splices that are part of the lateral force-resisting system should
be included in the performance model.

Second Edition
Structural steel braced frame fragility groups include selections for
concentrically braced (special and ordinary) as well as buckling-restrained
and eccentrically braced and systems. For each brace type and configuration,
selections are further subdivided by brace weight below, where “XXX” in
the fragility classification numbers represents an integer, depending upon the
subcategory represented:
• B1033.XXXa: Brace < 40 plf
• B1033.XXXb: 41 plf < Brace < 99 plf
• B1033.XXXc: Brace > 100 plf
Special concentrically braced systems (B1033.001a – B1033.043c) include

those with balanced designs in which gusset plates are configured to yield in
an elliptical pattern such that gusset strength coincides with brace strength
(Lehman et al., 2008), tapered gussets and those conforming to AISC
minimum standards. These categories are further subdivided by brace types
(wide flange, HSS, double angles) and for each brace type, by configuration
(chevron, single diagonal, x-diagonals).
Ordinary braced frames (B1033.051a – B1033.063c) are subdivided into

three basic categories: compact braces, braces at the ductile slenderness limit,
and braces designed without specific seismic detailing. Each of these
categories includes selections for chevron, single diagonal, and x-diagonal
configurations that reflect wide flange, HSS, or double angle brace sections.
Braced frames designed with factored loads only with no additional seismic
detailing (B1033.071a – B1033.073c) are subdivided by configuration
(chevron, single diagonal, x-diagonals).
Buckling-restrained brace (BRB) frames (B1033.101a – B1033.101c) are

subdivided into two categories based on their configuration (chevron braces
and single diagonal braces).
Story drift ratio is the demand parameter used for all steel braced frame
fragility specifications, except eccentrically braced systems. Steel braced
frame systems are measured in individual units, i.e., “each,” corresponding to
one bay of bracing.
Eccentrically braced frames (EBF) governed by the formation of a shear link

are divided into those with floor framing supported by the link beam and
those where the link beam does not support floor framing. These selections
are further divided by link beam weight below, where “X” in the fragility
classification numbers represents an integer, depending upon the subcategory
represented:

Second Edition
• B1035.06Xa: Link < 100 plf
• B1035.06Xb: 100 plf < Link < 200 plf
Link rotation angle is the demand parameter used for EBF systems. This
parameter can be obtained directly from response history analysis. When
simplified analysis is used, link rotation angle can be estimated using the
median estimate of story drift and the story yield drift. AISC 341-05,
Seismic Provisions for Structural Steel Buildings, (AISC, 2005) Section
C15.2 provides guidance for calculating link rotation angle from the story
yield drift. Because the behavior of the EBF system is governed by the shear
link, the braces themselves are considered rugged and are not included in the
performance model. Section 3.3.2.2.2 provides guidance for estimating story
yield. Section 3.3.1.3 provides guidance on input of structural analysis
results.
Eccentrically braced frame systems are measured in individual units, i.e.,

“each,” corresponding to one bay of eccentric bracing.
Steel moment frame systems include pre-Northridge and post-Northridge

categories. Post-Northridge selections are further subdivided into
configurations with reduced beam sections (RBS) with welded webs and not
RBS configurations. The systems are further subdivided into those with
beams on one side of the column, and those with beams on both sides of the
column and yet further by beam size as follows:
• B1035.001: RBS, Beam one side, Beam ≤ W27
• B1035.002: RBS, Beam one side, Beam > W30
• B1035.011: RBS, Beam two sides, Beam ≤ W27
• B1035.012: RBS, Beam two sides, Beam > W30
• B1035.021: Not RBS, Beam one side, Beam ≤ W27
• B1035.022: Not RBS, Beam one side, Beam > W30
• B1035.031: Not RBS, Beam two sides, Beam ≤ W27
• B1035.032: Not RBS, Beam two sides, Beam > W30
Pre-Northridge steel moment frame assemblies are assumed to have welded

flange bolted web connections without special weld access holes or notch-
through weld metal. Fragility specifications for these components are
subdivided into one- and two-sided configurations and by beam size as
follows:
• B1035.041: Beam one side, Beam ≤ W27

Second Edition
• B1035.042: Beam one side, Beam > W30
• B1035.051: Beam two sides, Beam ≤ W27
• B1035.052: Beam two sides, Beam > W30
Story drift ratio is the demand parameter used for all steel moment frame
fragility specifications. Beam-column moment connections are measured in
individual units, i.e., “each.”
2.4.2 Reinforced Concrete Elements (B104)

reinforced concrete moment frame and shear wall systems. Reinforced
concrete moment frame systems are divided into those with beam elements
and those consisting of slabs acting in conjunction with columns to provide
frame action.
Reinforced concrete moment frame systems with discreet beams are further
subdivided into four basic categories that reflect the modes of failure
anticipated based upon detailing and reinforcement proportioning. These
include: special moment frame, intermediate moment frame, ordinary
moment frame, and non-conforming frame systems. These systems are
subdivided based on the size of the beams and the presence of beams on one
side of the column or two, as illustrated in Table 2-4. In the table, “XX”
represents an integer representing a moment frame subcategory.
Table 2-4 Fragility Classification for Reinforced Concrete Moment Frames

Fragility Classification Number
Beam Size One Sided Two Sided

24×24 B1041.XX1a B1041.XX1b
24×36 B1041.XX2a B1041.XX2b
36×36 B1041.XX3a B1041.XX3b
Reinforced concrete special moment frame systems are subdivided into

frames governed by beam yield and those with relatively weak joints. For
those systems governed by beam yield, selections are further subdivided by
the presence of beams on one side of the column or two and beam size, and
XX in Table 2-4 is taken as 00, e.g., B1041.001a for a moment frame with
24×24 beam size with beams on one side of the column only.
For special moment frame systems judged to have relatively weak joints with
(∑Mc/∑Mb) > 1.2 Pu (col ) < 0.6f c′ and a joint shear ratio, (V/Vn), less than 1.2
with ACI 318-11, Building Code Requirements for Structural Concrete and

Second Edition
Commentary, (ACI, 2011) compliant transverse reinforcement fragility
specifications are similarly subdivided by beam size and configuration and
XX in Table 2-4 is taken as 01, e.g., B1041.011a for a moment frame with
24×24 beam size with beams on one side of the column only.
Fragility classification of intermediate moment frame systems includes

consideration of frame responses governed by beam-column shears and joint
shear damage. Fragility specifications in this category are further subdivided
by beam size and configuration and XX in Table 2-4 is taken as 02, e.g.,
B1041.021a for a moment frame with 24×24 beam size with beams on one
side of the column only.
Ordinary moment frame systems are subdivided into 5 subcategories that

reflect the governing behavior characteristics. These 5 subcategories are:
• Weak joints, beam yield (XX in Table 2-4 is taken as 03)
• Weak joints, column yield or joint shear yield (XX in Table 2-4 is taken
as 04)
• Beam yield (flexure or shear) with weak joints (XX in Table 2-4 is taken
as 05)
• Column yield with strong joints (XX in Table 2-4 is taken as 06)
• Weak columns with high axial loads and strong joints (XX in Table 2-4
is taken as 07)
Selections from among these subcategories can be made by reviewing the

strength ratios and notations in column 4 of Table A-2. Each subcategory is
further subdivided by beam size and configuration.
Non-conforming moment frame systems include 6 subcategories based on

anticipated behavior characteristics. These 6 subcategories are:
• Weak joints, beam yield (XX in Table 2-4 is taken as 08)
• Weak joints, column yield (XX in Table 2-4 is taken as 09)
• Weak beams, strong joints (XX in Table 2-4 is taken as 10)
• Weak columns, low to moderate axial loads (XX in Table 2-4 is taken as
11)
• Weak columns, strong joints, high axial loads (XX in Table 2-4 is taken
as 12)
• Inadequate reinforcement development (beam or column) (XX in Table
2-4 is taken as 13)

Second Edition
Selections from among these subcategories can be made by reviewing the
strength ratios and notations in column 4 of Table A-2. Each subcategory is
further subdivided by beam size and configuration.
Story drift ratio is the demand parameter used for all reinforced concrete
moment frames. Beam-column moment connections are measured in
individual units, i.e., “each.” For moment frames with beams, separate
performance groups must be designated in each direction.
Slab-column joints in flat slab systems are subdivided into the following 5
categories:
• Flat slab without shear reinforcement at columns (B1049.001a –
B1049.003b)
• Flat slab with shear reinforcement at columns (B1049.011 – B1049.012)
• Post-tensioned slab without shear reinforcement at columns (B1049.021a
– B1049.22b)
• Post-tensioned slab with shear reinforcement at columns (B1049.031 –
B1049.032)
• Flat slab with drop panel or capital without shear reinforcement at
columns (B1049.041a – B1049.042b)
Each of these subcategories is further subdivided by the ratio of gravity shear

to total shear (factored) and whether integrity reinforcement over columns is
present or not. This results in 18 selections that encompass classification
numbers B1049.001a through B1049.042b.
Story drift ratio is the demand parameter used for all reinforced concrete
moment frames. Slab-to-column connections are measured in individual
units, i.e., “each.” Slab-column moment frame fragility specifications use
non-directional demands.
Reinforced concrete shear wall systems include concrete link beams, low
aspect ratio walls (aspect ratio < 1.0), low-rise walls, and slender walls
(aspect ratio ≥ 2.0). Reinforced concrete wall systems are measured in units
of H×H panels, where H is the height of the panel. Concrete link beams are
subdivided based on their reinforcement (diagonal or conventional, and
further subdivided based on the aspect ratio of the link beam and the sizes of
the beam and column. This results in 12 selections that encompass
classification numbers B1042.001 – B1042.022b.
Nine subcategories of low aspect ratio walls reflect reinforcement and

configuration options (single curtain or double curtain of reinforcement and

Second Edition
the presence of flanges or boundary columns), wall thickness, and height.
Planar single curtain walls include walls 8 inches thick or less at heights
including 15 feet, 16-24 feet or 25-40 feet corresponding to fragility
classification numbers B1044.001, B1044.002 and B1044.003. Table 2-5
indicates the fragility categories available for concrete walls with two
curtains of reinforcement. Story drift ratio is the demand parameter for low
aspect ratio reinforced concrete wall systems.
Table 2-6 indicates the further sub-categorization of wall fragility

specifications to account for configurations with return flanges including
single curtain and double curtain reinforcement subcategories thickness and
height subdivisions:
Table 2-5 Fragility Classification for Low Aspect Ratio Reinforced Concrete
Walls with Two Curtains of Reinforcement
Thickness Height Fragility Classification Number
8-16 inches ≤ 15 feet B1044.011
8-16 inches 16 to 24 feet B1044.012
8-16 inches 25 to 40 feet B1044.013
17-24 inches ≤ 15 feet B1044.021
17-24 inches 16 to 24 feet B1044.022
17-24 inches 25 to 40 feet B1044.023
Table 2-6 Fragility Classification for Reinforced Concrete Walls with

Return Flanges
Fragility
Classification
Curtains of Reinforcement Thickness Height Number
≤ 15 feet B1044.031
Single ≤ 8 inches 16 to 24 feet B1044.032
25 to 40 feet B1044.033
≤ 15 feet B1044.041
8-16 inches 16 to 24 feet B1044.042
25 to 40 feet B1044.043
Double
≤ 15 feet B1044.051
17-24 inches 16 to 24 feet B1044.052
25 to 40 feet B1044.053
Similar to walls with flanges, walls with boundary columns are subdivided
into categories that reflect single and double curtains of reinforcement,

Second Edition
further subdivided by thickness and height. This results in 12 selections that
encompass classification numbers B1044.061 – B1044.083.
Slender walls (with height to length ratios in excess of 2) are subdivided by

thickness (12, 18, and 30 inches) and further subdivided by wall length (15,
Second Edition Update 20, and 30 feet). This results in 9 selections that encompass classification
Effective drift was introduced numbers B1044.091 – B1044.113.
as the demand parameter for
reinforced concrete slender Effective story drift ratio is the demand parameter for reinforced concrete
wall systems. slender wall systems.
2.4.3 Reinforced Masonry Elements (B105)

fully grouted and partially grouted reinforced masonry walls. Each of these
categories is further subdivided into shear dominated or flexure dominated
behavior. Shear dominated walls will fail in shear rather than yield in
flexure. Flexure dominated walls have sufficient shear strength to force
flexural hinging of the wall as a limit state. ACI 530-08, Building Code
Requirements for Masonry Structures and Specifications for Masonry
Structures, (ACI, 2008) equations can be used to establish the appropriate
subcategory selection. Within these subcategories (shear or flexure
dominated) further subdivisions are made based on wall thickness and height
as shown in Table 2-7.
Table 2-7 Fragility Classification for Reinforced Masonry Walls

Fragility
Construction Damage Classification
Characteristics Mode Thickness Height Number
≤ 12 feet B1051.001
4-6 inches
> 12 feet B1051.002
Shear ≤ 12 feet B1051.011

8-12 inches
dominated > 12 feet B1051.012
≤ 12 feet B1051.021
16 inches
Partially > 12 feet B1051.022
Grouted ≤ 12 feet B1051.003
4-6 inches
> 12 feet B1051.004
Flexure ≤ 12 feet B1051.013

8-12 inches
dominated > 12 feet B1051.014
≤ 12 feet B1051.023
16 inches
> 12 feet B1051.024

Second Edition
Table 2-7 Fragility Classification for Reinforced Masonry Walls (continued)
Fragility
Construction Damage Classification
Characteristics Mode Thickness Height Number
≤ 12 feet B1052.001
8-12 inches
Shear > 12 feet B1052.002
dominated ≤ 12 feet B1052.011
16 inches
> 12 feet B1052.012
Fully Grouted
≤ 12 feet B1052.003
8-12 inches
Flexure > 12 feet B1052.004
dominated ≤ 12 feet B1052.013
16 inches
> 12 feet B1052.014
Story drift ratio is the demand parameter used for masonry walls. Reinforced
masonry walls are measured in units of H×H panels, where H is the height of
the panel.
2.4.4 Cold-Formed Steel Structural Elements (B106)

repetitively framed wall systems that derive lateral resistance from wood
panel siding, flat strap x-bracing, and steel sheathing. Sufficiently strong
holdowns are assumed so that damage arises in the fasteners, studs, and
sheathing. Subcategories for these basic fragility specifications are included
to reflect performance of walls with stucco on one side (exterior). This
results in 6 selections that encompass classification numbers B1061.001a –
B1061.021b. Fragility specifications for interior walls with gypsum board
finishes (partitions) can be found in Section 2.5.4 and Table B-4.
Story drift is the demand parameter for cold-formed steel wall systems.
Cold-formed steel wall systems are measured in units of H×H panels, where
H is the height of the panel.
2.4.5 Wood Light Frame Structural Elements (B107)

repetitively framed walls with structural panel sheathing, gypsum board
sheathing, and let-in diagonal strut bracing. Structural panel sheathing
includes oriented strand board (OSB) or plywood. Fragility specifications
are provided for systems with and without holdowns for systems with stucco,
and for double side gypsum board sheathed walls without holdowns. This
results in 5 selections that encompass classification numbers B1071.001 –
B1071.031. Fragility specifications for wood stud walls with single sided
gypsum board only (partitions) can be found in Section 2.5.4 and Table B-4.

Second Edition
The wall fragility specifications provided are applicable to wall piers with
height to length ratios up to 2.0. For ratios from 2.0 to 3.5, users can adapt
the fragility specifications provided using the PACT Fragility Specification
Manager utility. This adaptation requires scaling the median value of each
damage state by the ratio 2L/H, where L is the wall length, and H is the wall
height. Section 6.3 provides guidance on the input of user defined fragility
information.
Table B-6 provides finish specifications for light framed walls. Finishes are
identified as individual components for input of fragility specification and
performance group assignment.
Story drift ratio is the demand parameter for wood light frame systems.
Wood light frame systems are measured in units of H×H panels, where H is
the height of the panel.
2.4.6 Structural Performance Groups
After identifying the fragility groups present, it is necessary to distribute

these into individual performance groups and populate these groups with the
quantities of components present in each.
Structural performance groups are entered into PACT on a floor-by-floor

basis for each of: direction 1, direction 2, and non-directional, as appropriate.
A performance group of like components should be assembled for all
elements resisting earthquake deformation in each direction at each floor.
For steel and reinforced concrete moment frames, the number of interior
(beams framing to both sides of the column) and exterior (beams framing to
one side of the column) joints is input for each direction on each floor. For
steel braced frames, the number of bays of framing is the quantity used.
Flat slab column joints are susceptible to damage associated with the
maximum story drift ratio in any horizontal direction. Such components are
defined in PACT as non-directional.
For wall systems, the unit of measure is the number of panel elements active
in a particular direction of response. For example, low aspect ratio shear
walls are measured in units of H×H panels because damage in these walls
often occurs in the form of diagonal “X” pattern cracking extending across
panels of this size. For a building with a 36 foot long wall and 12 foot story
height, the number of panels input into PACT would be three (36ft/12ft) in
that direction.

Second Edition
Slender walls having aspect ratios greater than 2:1 typically exhibit
flexurally-controlled behavior resulting in damage occurring predominantly
in a flexural hinge region at the base of the wall. Aspect ratio is defined as
the overall length of the wall divided by its length. The flexural hinge region
is typically taken as the greater of one half the wall height or a single story
height. Given that these walls generally only exhibit flexural damage in the
hinge region, structural performance groups need only be defined at these
floor locations. If secondary hinges or shear damage are anticipated at other
levels of the building due to higher mode effects or changes in the wall
geometry, structural performance groups for these stories should also be
defined. Other portions of the wall need not be included since damage is not
anticipated.
If performance group members are identified as “correlated,” PACT will

determine a single damage state for each realization that applies to all
members of the performance group. If performance group members are
identified as “uncorrelated,” PACT will independently evaluate the damage
state for each member of the group on the basis of the demand value and the
component’s fragility function. Designating performance groups as
correlated can achieve a significant reduction in data handling time and
required computer disc space. There will, however, be a significant loss of
variability in the resulting assessment. By selecting an uncorrelated
relationship among group members, greater variability can be expected in the
outcome of each realization; however, the mean consequences can be
expected to be similar to the result if the members are considered as
correlated. If the distribution of loss is important to the user, performance
groups should be identified as uncorrelated. When casualty and unsafe
placard postings are selected as performance measures, the user should not
use correlated components in a performance group, unless there is reason to
believe that behavior really is correlated. For example, a series of identical
braces along line framing can be considered correlated because all braces
will experience the same deformation and one brace cannot exhibit
significant buckling deformation unless all of the braces in this line do as
well. In general, components should be considered uncorrelated unless other
compelling reasons exist for a correlated designation.
Figure 2-8 illustrates structural performance group selections for direction 1

on the first floor of a two-story building. Special reinforced masonry shear
walls (B1052.003) have been identified with a quantity of 12.8 units. This
quantity is number of H×H panels. For this example, the story height is 10
feet; therefore, the quantity reflects 128 feet of 10 foot high masonry wall in
direction 1. Note that for uncorrelated components, PACT considers only

Second Edition
integer values of these quantities. If non-integer quantities are input, PACT
will round these to the nearest integer.
Figure 2-8 PACT Performance Groups tab.
2.5 Identification of Vulnerable Nonstructural

Components and Systems
Appendix B lists the nonstructural fragility specifications provided in the

Fragility Database referenced by PACT, organized into 25 component
categories, spanning the range of exterior curtain wall window systems to
desktop electronics.
Each component category contains multiple fragility specifications, which

reflect a range of configuration and material variations affecting seismic
performance characteristics. PACT makes further differentiation among
some components with identical damageability but with significant
differences in damage state consequences, such as repair costs, owing to such
characteristics as size and finish.
User-defined fragility specifications can be added as needed or existing

specifications can be modified using the PACT Fragility Specification
Manager utility. The 25 groups of nonstructural components and systems
reflect the range of damageable items generally found in buildings. Users
should trace an organized path that will comprehensively define the
damageable nonstructural building components and systems.
There are many components for which fragility specifications are not
provided. Users can develop fragility functions for these items using the
guidance provided in Volume 1, Section 3.8. Chapter 7 provides guidance
on development of nonstructural component fragility parameters by

Second Edition
calculation and Chapter 8 provides guidance for development of consequence
functions.
2.5.1 Exterior Nonstructural Walls (B107 and B201)
This fragility classification (B107 and B201) is summarized in Table B-1 and
includes the following types of exterior nonstructural walls:
• Gypsum with wall studs (B1071.041).
• Cold-formed steel walls (B2011.001a – B2011.021b). These
specifications are the same as those in category B106.
• Wood light frame walls (B2011.101 – B2011.131) with structural panel
sheathing, gypsum board sheathing, and let-in diagonal strut bracing.
These specifications are the same as the wood light frame walls in
category B107.
• Precast concrete cladding panels (B2011.201a – B2011.201b).
For walls in this category, damageability is often a function of detailing.

2.5.1.1 Perimeter Cladding Systems
Fragility and consequence functions for these systems should include

consideration of the size of the gap provided in the panel system to
accommodate horizontal drift. The in-plane distortion required to bring
adjacent panels into contact can be converted to an equivalent story drift
using the geometry of the panel. Contact points can occur at the panel
corners and at the interface with column cover panels. System suppliers
often can indicate the amount of drift that can be accommodated before
adjacent panels contact. The design story drift for the structure should be
used for estimating the drift accommodation ratios of cladding systems. If
this unknown, Table 2-8 provides guidance for commonly constructed gap
sizes as a function of building category (according to ASCE/SEI 7) and the
operative design code for the period of construction. The table expresses the
gap in terms of story drift ratio.
Table 2-8 Suggested Drift Accommodation Ratio δa
Building Occupancy Category per ASCE/SEI 7
Building Code II III IV

1994 UBC (ICC, 1994) or before 0.012 0.012 0.012
1997 UBC (ICC, 1997) 0.020 0.020 0.020

2000 IBC (ICC, 2000) – 2009 IBC (ICC, 2009) 0.020 0.015 0.010
2012 IBC (ICC, 2012) or thereafter 0.020 0.019 0.015

Second Edition
In-plane story drift ratio is the demand parameter for perimeter cladding
systems. The area of cladding per floor is dependent on the floor perimeter
in each direction and the height of the floor (less the area of glazing).
Perimeter cladding systems are measured in units of 100 square feet, i.e., the
net cladding area (in square feet) in each direction between floors divided by
100.
Perimeter cladding systems may also be vulnerable to damage when out-of-

plane floor accelerations or drifts exceed the component or anchorage
capacity. To avoid double counting damage to the same component, only the
specification for the governing demand should be input to the assessment
model. In cases where vulnerability to in-plane and out-of-plane demands is
comparable, some of the components can be assigned to one performance
group having in-plane vulnerability, and the balance to a second group
having out-of-plane vulnerability.
2.5.1.2 Precast Concrete Cladding Panels
Two fragility specifications are provided for precast concrete cladding panels
(B2011.201a and B2011.201b). One is associated with damage in-plane
while the other fragility is associated with out-of-plane damage. Users must
provide the median demand and dispersion values based on the details of the
system present.
Unless panel detailing indicates greater capacity, it is suggested that the

median in-plane demand requiring panel replacement be taken as (δa + 0.005)
where δa is the code design story drift used for the precast cladding system,
or if this unknown, the suggested drift accommodation as indicated in Table
2-8. A dispersion value of 0.50 is suggested.
Peak floor acceleration in the direction normal to the wall at the top floor
adjacent to a panel is the demand parameter used for out-of-plane damage.
Section 7.3 provides guidance on the calculation of anchorage fragilities. For
most situations, a crude but adequate calculation can be made by using the
code seismic design acceleration coefficient for which the anchorage was
presumed to be designed (see Section 7.3.1) and converting this to a median
estimate of capacity. Precast concrete cladding panels are measured in units
of 30 lineal feet.
Failure of precast concrete cladding has no life safety consequences within

PACT because the exterior population model is unknown.

Second Edition
2.5.2 Exterior Window Systems (B202)
This fragility classification (B202) is summarized in Table B-2 and includes

exterior window systems. The specifications are based on testing of modern
metal framed exterior window systems. Testing was performed on glazing
with square cut corners (the most common form of exterior window
construction).
A total of 48 fragility specifications are provided, and encompass

classification numbers B2022.001 through B2023.011. Forty-four of these
fragilities reflect specific glazing installation details while four are provided
for situations when the glazing installation detail is not defined and
represents a composite of all possible details.
Fragility data for two basic window system types are provided: curtain wall
systems and storefront systems. Both are for window systems supported and
contained within aluminum framing.
Twenty-eight of the fragility specifications are provided for site-built (mid-

rise stick built) curtain wall systems, which is the most common form of
construction in the commercial building industry and for which most testing
has been performed. The site-built curtain wall system specifications are
divided into 6 configuration types: non-laminated monolithic (single pane),
symmetric (dual pane with equal thicknesses) insulated glazing units (IGU),
asymmetric (dual pane with unequal thicknesses) IGUs, and laminated
versions of each of these. These system types are further subdivided into
three types of glass: annealed, heat-strengthened, and fully-tempered.
Separate specifications are provided for variations in glazing edge
clearances, glazing aspect ratios, and glass thicknesses (B2022.011 through
B2022.101).
One specification is provided for unitized curtain wall (B2022.201).
The storefront window system specifications are also subdivided into

configuration types including: monolithic, symmetric, and asymmetric IGUs
with laminated glass for some selections, with further subdivision into
annealed and fully tempered glass types. Separate specifications are
provided for variations in glazing edge clearances, glazing aspect ratios, and
glass thicknesses (B2023.011 through B2023.058).
Particularly for existing buildings, in many situations, the details of glazing

installation may not be known. Four additional fragility specifications have
been provided for these situations. Two of these are for curtain wall systems,
one for monolithic, and one for IGU glazing, B2022.001 and B2022.002,

Second Edition
respectively. There is no subdivision for laminated glazing. It is assumed
that the glazing aspect ratio is 6:5. The glazing edge clearance and glass
thicknesses are assumed as unknown and variations are incorporated into the
provided specifications. Similarly, two fragility specifications are provided
for generic storefront window systems, B2023.001 and B2023.002.
The fragility information provided is based on the size of glazing panels

tested. The median drifts associated with the damage states should be
adjusted if smaller or larger glazing panels are used and gap clearances are
basically the same as those for which fragilities are provided. The
adjustment factor is the ratio of the tested panel height divided by actual
panel height. This ratio is multiplied by the median drift value for each
damage state in the specification. Table 2-9 lists the heights of tested panels
for each aspect ratio.
Table 2-9 Tested Window Pane Sizes

Aspect Ratio Tested Panel Height
6:5 6 feet
2:1 8 feet
1:2 4 feet
Where specific knowledge or details of the exterior window system is

unavailable, it is suggested to use the IGU specifications with a 6:5 aspect
ratio for curtain wall applications and the monolithic fragility specifications
for storefront applications.
Metal panels are frequently used in conjunction with exterior window

systems. Generally, the exterior window system will exhibit damage before
the metal panel system. Fragility functions for the window system can be
used for the combined system and the user can modify the consequence
function in PACT to reflect the consequences to the entire system. Chapter 8
provides guidance on modifying consequence functions.
Story drift ratio in the plane of the wall is the demand parameter for exterior
window systems. The area of glazing per floor is dependent on the perimeter
length of the floor in the direction under consideration and height of the floor
less the area of perimeter cladding or other non-glazed areas. The exterior
window systems are measured in units of 30 square feet, i.e., the net window
area (in square feet) divided by 30.

Second Edition
2.5.3 Roof Elements (B30)
2.5.3.1 Roof Coverings (B301)

fragility specifications for sloping roofs with concrete or clay tiles. The
specifications are further subdivided based on whether the attachment to the
structure is compliant with the requirements of the 1994 UBC, Uniform
Building Code, (ICC, 1994) and later codes or not.
Peak roof acceleration (non-directional) is the demand parameter for roof

coverings. Roof coverings are measured in units of 100 square feet, i.e., tiled
roof area (increased for slope) in square feet divided by 100.
2.5.3.2 Unreinforced Masonry Non-Industrial Chimneys
(B303)

unreinforced non-industrial masonry chimneys. The specifications are
subdivided by chimney height above the roof, including: 5 feet, 10 feet, and
15 feet. Each of these categories is further subdivided into two selections
that include reconstruction of the chimney with reinforced masonry or
replacement with a framed alternative. Most jurisdictions no longer permit
new construction using these types of chimneys in areas of significant
seismic risk.
Peak roof acceleration (non-directional) is the demand parameter used for

these chimneys. Chimneys are measured in individual units, i.e., “each.”
2.5.3.3 Masonry Parapets (B304)
Fragility specifications are provided for unreinforced, unbraced masonry

parapets. Peak roof acceleration normal to the parapet wall is the demand
parameter used for these parapets. Masonry parapets are measured in units
of 100 lineal feet. As with precast cladding, failure of masonry parapets does
not have any life safety consequences in PACT, although real consequences
can exist.
2.5.4 Partitions (C101)
This fragility classification (C101) is summarized in Table B-4 and includes

gypsum wallboard partitions on metal and wood studs.
Four subcategories of gypsum wallboard partitions on metal studs are

provided: (1) full height partitions fixed to the underside of the floor at their
top (C1011.001a); (2) partial height partitions sheathed with gypsum
wallboard 9 feet vertically and braced to the underside of the above floor

Second Edition
(C1011.001b); (3) full height partitions without corners where slip track top
connections restrain out-of-plane movement but not in plane horizontal or
vertical movement (C1011.001c); and (4) full height partitions with slip track
connections at their tops and returns at their ends (corners) (C1011.001d). It
should be noted that slip track connections are often installed such that
horizontal movement is restrained and therefore behave more like full-height
partitions fixed to the floor above than partitions that have been designed to
accommodate horizontal drift. Even when effective, slip track connections
are intended to permit in-plane movement, not out-of-plane movement. As a
result, such walls will behave as if they had fixed connections, particularly
near intersections with other walls.
Fragilities for true “drift-protected” partition walls are not presently provided
but can be added as user-defined when appropriate. The fragility and
consequence functions for all four provided partition types assume gypsum
wallboard on both sides of the partition and include allowances for all typical
accessories including doors and embedded mechanical and electrical utilities.
For damage states where the consequence includes partition replacement, the
repair cost includes allowance for these accessories.
A single fragility specification (C1011.011a) is provided for full-height wood

frame partitions fixed to the floor above. The fragility and consequence
functions assume gypsum wallboard on both sides and include allowances
for all typical accessories, including doors and embedded mechanical and
electrical utilities. For damage states where the consequences include
partition replacement, the repair cost includes allowance for these
accessories.
It is assumed that partitions will be constructed the same regardless of finish

type and will include a painted finish with taped joints. All other finishes are
addressed separately under the wall finishes component category (C301) in
Table B-6. The partition fragility specifications in Table B-4 must be
separately included as partition walls.
Story drift ratio in the plane of the wall is the demand parameter used with
partitions. Partitions are measured in units of 100 lineal feet. For estimating
purposes, the walls are assumed to be 9 feet tall.
2.5.5 Stairs (C201)

four types of interior egress stairs: (1) prefabricated steel stair assemblies
(C2011.001a and C2011.001b); (2) prefabricated precast concrete stair
assemblies (C2011.011a and C2011.011b); (3) cast-in-place concrete stairs

Second Edition
(C2011.021a through C2011.021d); and (4) hybrid stairs with steel stair
stringers with precast concrete treads (C2011.031a and C2011.031b). Cast-
in-place concrete stairs placed monolithically with the structure should be
included in the structural analysis model. Stair specifications are further
subdivided based on support condition, including stairs fixed at both ends
and stairs with bottom ends having slotted joints designed to accommodate
story drift. For existing buildings, if the details of stair installation are not
available, it is reasonable to assume the stairs are a hybrid type without
slotted joints, i.e., fixed at both ends. Consequence functions are provided
for all stair types.
Fragility functions are only provided for prefabricated stair assemblies (steel,
precast, hybrid) composed of individual stringers, stair treads, and landings
assembled without seismic joints. Fragility functions are not provided for
cast-in-place and monolithic precast stairs. These components require user
specified fragility functions. Chapter 6 and Chapter 7 provide guidance on
developing fragility functions by calculation.
Fragility functions for stairs with seismic joints are not provided and need to
be specified by the user. A suggested approach for establishing a fragility
function for this condition requires estimating the story drift the joint is
designed to accommodate and adding the value to the drift associated with
the provided component fragility without a seismic joint.
Story drift ratio in the direction of stair alignment is the demand parameter
for these systems. If the directional orientation of the stair is unknown, the
demand parameter can be adjusted to non-directional. The number of stairs
is based on gross square foot of building floor area and occupancy.
Volume 1, Appendix F provides quantity guidance for stairways per floor
based on building floor area and occupancy. For buildings that have the
same floor area at all floors, the same number of stairs should be specified at
each level. Stairs are measured in individual units of stairways per floor, i.e.,
“each.”
2.5.6 Wall Finishes (C301)

finishes other than paint: wallpaper (C3011.001), ceramic tile (C3011.002),
and high end finishes including marble or wood paneling (C3011.003).
These fragility specifications are further subdivided depending on whether
the finishes are applied to partitions that are: (1) full height and fixed at the
top; (2) partial height and laterally braced at the top; or (3) full height with
slip track top connections.

Second Edition
The fragility and consequence functions assume the finish occurs on one side
of the partition only, allowing different wall finishes to be specified for
opposite sides of the same wall. Note that only damage states and
consequences associated with damage to the finishes themselves are
considered in these fragility specifications. The partitions themselves must
be input separately.
Story drift ratio in the plane of the wall is the demand parameter used with
wall finishes. Wall finishes are measured in 100 lineal feet of wall, similar to
partitions. If both sides of a wall have finishes, the quantities should be
increased accordingly.
2.5.7 Raised Access Floors and Floor Flooding (C302)
2.5.7.1 Floor Flooding
A fragility category is provided to account for floor flooding as a result of

Second Edition Update fire sprinkler drop or pipe leakage. Flooding fragility specifications are
Considerations regarding floor provided for each of the 8 occupancy types. These specifications are further
flooding has been updated.
subdivided based on a high or low likelihood of mold developing subsequent
to the flooding. Development of mold is highly likely in geographic
locations of high humidity. The flooding fragilities are non-directional.
The median demand for floor flooding is selected by the user as equal to the
lowest median value of a piping component damage state included in the
performance model that will result in a major leak. For Seismic Design
Categories A, B, and C (no ceiling bracing) sprinkler drops will typically
govern the demand selection. For Seismic Design Categories D, E, and F the
demand selection is likely governed by threaded water piping.
Flooding damage is presumed to occur on only the floor where the major
leak occurs. The demand parameters and dispersion are the same as the
values for the piping component with the lowest median damage state
demand. The unit of measure is one square foot. For floor flooding
consequences, if the assumption is that the entire floor is flooded, the
performance group quantity should be set to the area of entire floor. The user
may assume only a part of floor is flooded by modifying the performance
group quantities. The likelihood of floor flooding throughout a floor level is
related to the presence of building maintenance personnel capable of rapidly
shutting down sprinkler flow.
2.5.7.2 Raised Access Floors

raised access floors.

Second Edition
Two types of fragilities are provided: unanchored floors (C3027.001),
including floors with adhesive attachment, and seismically-rated access
floors with pedestals anchored and braced to the floor (C3027.002). Current
seismic codes strongly encourage seismically-rated floors for buildings
assigned to Seismic Design Categories D, E, or F. However, anchorage
conditions vary considerably.
Peak floor acceleration (non-directional) is the demand parameter for raised

access floors. Raised access floor systems are measured in units of 100
square feet, i.e., the floor area (in square feet) divided by 100.
2.5.8 Suspended Ceilings and Ceiling Lighting (C303)
This fragility classification (C303) is summarized in Table B-8 and

comprises suspended lay-in ceilings and lighting systems.
Four types of suspended lay-in ceiling specifications are provided

(C3032.001 through C3032.004), based on conformance to the prescriptive
seismic detailing requirements of recent building codes. Types included are
ceilings satisfying detailing requirements for: (1) Seismic Design Categories
A and B; (2) Seismic Design Category C; (3) Seismic Design Categories D
and E; and (4) Seismic Design Category F and where the importance factor Ip
= 1.5 applies. Each of these ceiling fragility specifications is further
subdivided by area of individual ceiling size. Ceiling size is a function of
whether ceilings run continuously over partitions or are independent within
each room on a floor. Four ceiling area sizes are provided: (1) less than or
equal to 250 square feet; (2) 250 to 1000 square feet; (3) 1000 to 2500 square
feet; and (4) greater than 2500 square feet. These area sizes reflect the
increasingly more detailed prescriptive requirements contained in current
codes corresponding to the ceiling area. Lighting fixtures are included in the
lay-in ceiling specifications. For buildings of recent construction, if the
ceiling type is not known, the specification associated with the building’s
current Seismic Design Category and appropriate ceiling area should be
selected. For ceilings that predate the adoption of codes with seismic
detailing criteria, fragilities associated with Seismic Design Category A
should be selected.
Peak floor acceleration (non-directional) of the floor above is the demand

parameter for ceiling systems. Volume 1, Appendix F provides guidance on
total ceiling area per floor area, but does not address how to split the total
ceiling area into different ceiling area subgroups based on room size. Users
must exercise judgment in assigning these quantities. Each area subgroup
uses a different quantity unit to measure this fragility. For areas less than or

Second Edition
equal to 250 square feet, the subgroup unit measure is 250 square feet. For
ceiling areas between 250 square feet and 1000 square feet the performance
group unit is 600 square feet. The unit for areas between 1000 square feet
and 2500 square feet is 1800 square feet. The unit for areas greater than
2500 square feet is 2500 square feet.
Two fragility specifications are provided for suspended pendant lighting

fixtures. One specification is provided for fixtures not seismically designed
or braced (C3034.001) and another for seismically-rated fixtures with
seismically designed restraints (C3034.002). Current seismic codes require
seismically-rated lighting fixtures and pendant light fixtures for buildings in
Seismic Design Categories D, E, or F, unless there is no potential for impact
interaction. If the lighting fixture is free to swing during an earthquake and
the components are detailed to accommodate the swing without impact, the
user can consider the fixtures rugged and not include them in the assessment
model.

parameter for suspended pendant lighting systems. Lighting systems are
measured in individual units, i.e., “each.”
2.5.9 Elevators and Lifts (D101)
This fragility classification (D101) is summarized in Table B-9 and includes

selections for hydraulic and traction elevators. Traction elevators are further
divided into those with seismic provisions and those without. Seismic
requirements were added to the design requirements for elevators in the 1975
California Building Code. These requirements substantially reduced the
vulnerability of elevator systems and were adapted for other seismic zones in
the ASTM A17.1, Safety Code for Elevators and Escalators, (ASME, 1981).
If the type of elevator is unknown, the user can assume that buildings with
more than 5 stories will have traction elevators and buildings with 5 stories
or less will have hydraulic elevators.
The number of elevators present in a building is a function of many factors

but the primary determinants are the building’s gross square footage and
occupancy type. Volume 1, Appendix F provides guidance on selecting the
number of elevators based on floor area and occupancy. Performance groups
for elevators should be defined only for the first floor.
Peak floor acceleration (non-directional) at the first floor is the demand

parameter for elevators. Elevators are measured in individual units, i.e.,
“each.”

Second Edition
2.5.10 Domestic Water Distribution (D202)

cold and hot water piping. The specifications are divided into piping for cold
and hot potable service, and heating hot water piping. The specifications are
further divided into subgroups depending on the seismic design category and
importance factor as follows: (1) Seismic Design Categories A and B; (2)
Seismic Design Category C; (3) Seismic Design Categories D and E; and (4)
Seismic Design Category D, E, or F where the importance factor, Ip, is 1.5.
Further subdivision is made for water piping based on the pipe diameter and
the use of threaded or welded steel connections. Separate specifications are
provided for the piping itself and, in most cases, for anchorage and bracing
supporting the piping. For these components, both specifications should be
included in the performance model.
For recently constructed buildings, if the type of distribution system detailing

is not known, the type appropriate to the building’s Seismic Design Category
and importance should be selected. For construction that predates the
adoption of codes with seismic detailing for these systems, Seismic Design
Category A should be assumed.
Peak floor acceleration (non-directional) in the floor above is the demand

parameter for domestic water piping systems. Domestic water piping
systems are measured in units of 1000 lineal feet, i.e., the total length of
piping per floor (in lineal feet) divided by 1000.
As previously discussed in Section 2.5.7, it is possible to include a floor

flooding fragility specification to account for flooding damage.
2.5.11 Sanitary Waste Piping (D203)

sanitary waste piping. These specifications are further divided into
subgroups depending on seismic design category and importance factor
similar to other piping types. Further subdivision is made for cast iron piping
with flexible couplings and those with ball and spigot couplings. In most
cases, separate specifications are provided for anchorage and bracing of ball
and spigot couplings. Both specifications should be included in the
performance model for these components.

parameter for these systems. Sanitary waste piping systems are measured in
units of 1000 lineal feet, i.e., the total length of piping per floor (in lineal
feet) divided by 1000.

Second Edition
2.5.12 Chilled Water and Steam Piping (D205 and D206)
These fragility classifications (D205 and D206) are summarized in

Table B-12 and Table B-13 and include chilled water and steam piping
systems. These specifications are further divided into subgroups depending
on seismic design category and importance factor similar to other piping
types. Further subdivision is made for welded and threaded connections and
for pipe diameters greater than 2.5 inches and less than 2.5 inches. Separate
specifications are provided for both the piping itself and for anchorage and
bracing supporting the piping. For these components, both specifications
should be specified.
For recently constructed buildings, if the type of distribution system detailing

is not known, the type appropriate to the building’s Seismic Design Category
and importance should be selected. For construction that predates the
adoption of codes with seismic detailing for these systems, Seismic Design
Category A should be assumed.
The demand parameter for these systems is peak floor acceleration (non-
directional) in the floor above. These piping systems are measured in units
of 1000 lineal feet, i.e., the total length of piping per floor (in lineal feet)
divided by 1000.

2.5.13 Chillers, Cooling Towers, and Compressors (D303)

chillers, cooling towers, and compressors. Each of these specifications is
further subdivided by the installation characteristics including:
(1) unanchored and not isolated; (2) isolated and not restrained; and (3) hard
anchored or isolated and restrained. Each of these subcategories is further
subdivided by unit or by whether the unit services small or large medical
systems (large and small are general terms subject to user judgment).
The final category used to differentiate among fragility specifications for

these components is the dominant damage mode. Three possible damage
modes are identified. These include:
• Anchorage failure: Equipment anchorage is relatively weak and fails
well before the equipment is damaged.

Second Edition
• Internal damage: Equipment anchorage is relatively strong and the
equipment suffers internal damage well before the anchorage fails.
• Simultaneously occurring anchorage failure and internal damage:
Equipment anchorage fails and is accompanied by internal damage to the
equipment.
Consequence functions are provided for each of these fragility specifications;

however, users are required to define the median acceleration value
associated with anchorage failure for each piece of equipment. With this
calculated value, the Fragility Specification Manager utility can assist in
determining the best selection for the dominant damage mode by comparing
median floor accelerations for the anchorage and the equipment damage
modes (PACT fragility selections for the equipment damage mode include
median floor acceleration values and dispersion values). Users should:
• Select the anchorage failure damage mode if the median value of peak
floor acceleration associated with anchorage failure is less than 30% of
the value associated with internal damage to the equipment.
• Select the equipment internal damage mode if the median peak floor
acceleration for internal equipment damage is less than 30% of the
median value of acceleration associated with anchorage failure.
• Select the simultaneous anchorage failure and internal equipment
damage mode for all other median acceleration ratios.
For equipment that is unanchored, only damage modes that include internal
damage are provided in PACT. Users can define additional damage modes
associated with failure of attached piping due to equipment sliding. Section
7.3 provides guidance on calculation of anchorage fragilities.
Peak floor acceleration (non-directional) is the demand parameter for these

systems. Estimates of the size and number of equipment items likely to
occur given the size and occupancy of the building under construction can be
found in Volume 1, Appendix F and the Normative Quantity Estimation
Tool. The capacity of an individual equipment item is expressed in terms of
the gross square footage of building area it can serve. Depending on the total
demand, the size and number of equipment units can be assigned. For
equipment with capacity of less than 100 tons, the unit of measure is 75 tons.
For equipment with capacity between 100 tons and 350 tons, the unit of
measure is 250 tons. For equipment with capacity between 350 tons and 700
tons, the unit of measure is 500 tons. For equipment with capacity between
750 tons and 1,000 tons, the unit of measure is 850 tons. If the demand
exceeds the capacity of any available unit (greater than 1,000 tons), two

Second Edition
items of equipment equal in size that provide the needed total capacity can be
assigned. Most equipment identified in this category will be located on the
first floor, basement or roof. Items should only be identified once for a given
building. These systems are measured in individual units, i.e., “each.”
2.5.14 HVAC Distribution Systems (D304)

components of HVAC distribution systems including: in-line fans, ducting,
drops and diffusers, variable air volume boxes, and fans.
In-line fan specifications are divided into subgroups that include units that
are independently supported with vibration isolators (D3041.001) and those
without vibration isolators (D3041.002). These systems are further divided
depending on Seismic Design Category as follows: (1) Seismic Design
Categories A and B; (2) Seismic Design Category C; (3) Seismic Design
Categories D and E; and (4) Seismic Design Category D, E, or F.
Distribution ducting is divided into galvanized (D3041.011 and D3041.012)

and stainless (D3041.021 and D3041.022) steel sheet metal. These
specifications are further subdivided by size (less than or greater than 6
square feet in cross sectional area). Similar to in-line fans, ducting
specifications are further subdivided by Seismic Design Category.
Drop and diffuser specifications are divided into categories of the installation
location: in suspended ceilings (D3041.031) or without ceilings (D3041.032).
These specifications are further subdivided by Seismic Design Category.
Variable air volume boxes (VAV) (D3041.041) are divided into two
categories: Seismic Design Categories A or B and Seismic Design Category
C or above.
HVAC fan specifications are divided into three categories: (1) unanchored
and not isolated (D3041.101); (2) isolated but not restrained (D3041.102);
(3) and hard anchored or isolated and restrained (D3041.103). These three
categories are further subdivided by dominant damage mode (anchorage
failure, internal damage, or simultaneous damage) in a manner similar to that
done for the components in Section 2.5.13. In an analogous fashion, the user
must determine the median value of peak floor acceleration that will produce
anchorage or isolator failure and use that value to select the dominant
damage mode most appropriate.

parameter for these systems. The fragilities are measured with the following
units:

Second Edition
• In-line fans – number of 10 units
• HVAC ducting – number of 1,000 lineal feet units
• Drops or diffusers – number of 10 units
• VAV boxes – number of 10 units
• HVAC fans – number of items, “each”
2.5.15 Packaged Air Handling Units (D305)

packaged air handling units. These specifications are divided into categories
for equipment that is unanchored and not isolated (D3052.011) and
equipment that is hard anchored or isolated and restrained (D3052.013).
Each of these categories is further subdivided by equipment size as follows:
• Equipment size < 5,000 cfm
• 5,000 cfm ≤ Equipment size < 10,000 cfm
The hard anchored or isolated and restrained category is further subdivided

by the dominant damage mode. These selections include anchorage failure,
internal damage, or the simultaneous occurrence of both. Note that for
unanchored, not isolated equipment, only an internal damage mode is
provided. Section 2.5.13 provides further information on estimating the
median floor acceleration associated with anchorage failure and the use of
that information to select the most appropriate damage mode.
Peak acceleration of the supporting floor (non-directional) is the demand

parameter for these systems.
The size of an individual equipment item is expressed in terms of the gross

square footage of building area it can serve. Depending on the total demand,
the size and number of equipment units can be assigned. For equipment with
a capacity smaller than 5,000 cfm, the unit of measure is 4,000 cfm. For
equipment with a capacity between 5,000 cfm and 10,000 cfm, the unit of
measure is 8,000 tons. For equipment with a capacity between 10,000 cfm
and 25,000 cfm, the unit of measure is 20,000 cfm. For equipment with a
capacity between 25,000 cfm and 40,000 cfm, the unit of measure is 30,000
cfm.

Second Edition
2.5.16 Control Panels (Table D306)

control panels. This category is further subdivided into unanchored and
anchored panels. For unanchored systems, only one specification is provided
(B3067.011a); however, for hard anchored panels, the user should select the
most appropriate fragility specification based on a dominant damage mode
that includes anchorage failure, internal damage, or the simultaneous
occurrence of both. Section 2.5.13 provides further information on
estimating the median floor acceleration associated with anchorage failure
and the use of that information to select the most appropriate damage mode.
Peak acceleration of the supporting floor (non-directional) is the demand

parameter for control panels. Control panels are measured in individual
units, i.e., “each.”
2.5.17 Fire Protection (D401)

fire protection system components. Testing of connections indicate that
vertical standpipes designed and installed in accordance with NFPA 13,
Standard for Installation of Sprinkler Systems, (NFPA, 2007) have a median
story drift ratio capacity exceeding 0.05 and are therefore considered
seismically rugged.
Specifications for thin wall piping systems with older Victaulic type
connections are subdivided by Seismic Design Category. The fragility
specifications provided for the horizontal distribution system includes its
bracing.
Fire sprinkler drop fragility specifications are subdivided into the following
categories: (1) lay-in tile soft ceilings; (2) lay-in tile hard ceilings; and (3) no
ceilings. Each of these groups is further divided based on Seismic Design
Category. Seismic Design Categories A, B, or C imply no seismic
requirements or lateral bracing of the ceiling while Seismic Design Category
D, E, and F imply that lateral seismic bracing is provided with a stringent
level of design and construction installation review.

parameter for these systems. Fire sprinkler piping systems are measured in
units of 1,000 lineal feet per floor. Fire sprinkler drops are measured in units
of 100.

Second Edition
2.5.18 Electrical Service and Distribution (D501)

the following building electrical components: motor control centers, primary
service transformers, low voltage switchgear, and distribution panels.
Each of these fragility specifications are subdivided by the following

considerations: (1) unanchored, not isolated; and (2) hard anchored or
isolated and restrained.
For transformers, further subdivisions are made based on unit size as follows:
• 100 to 350 kVA – measured in units of 250 kVA
• 750 to 1,200 kVA – measured in units of 1000 kVA
• 1,200 to 2,000 kVA – measured in units of 1,500 kVA
For low voltage switchgear and distribution panels, the following

subdivisions are used:
• 100 to 350 amp – measured in units of 225 amp
• 350 to 750 amp – measured in units of 400 amp
• 750 to 1,200 amp – measured in units of 800 amp
• 1,200 to 2,000 amp – measured in units of 1,600 amp
All of the fragility specifications are further divided by the dominant damage
mode (anchorage failure, internal damage and simultaneous occurrence of
both). Section 2.5.13 provides information on estimating the median floor
acceleration associated with anchorage failure and the use of that information
to select the most appropriate damage mode.
Peak floor acceleration (directional) of the supporting floor is the demand

parameter for these systems. Motor control centers are measured in
individual units, i.e., “each.” The size of transformers, switchgear, and
distribution panels depend on the total demand. The units of measure for
these systems are indicated above.
2.5.19 Other Electrical Systems (D509)

the following building electrical components not included in Section 2.5.18:
battery racks, battery chargers, and diesel generators.

Second Edition
Fragility specifications for these items are subdivided into the following
categories: (1) unanchored, not isolated; and (2) hard anchored or isolated
and restrained.
Diesel generators include specifications for conditions where the equipment

is isolated and not restrained, as well as the above conditions. Diesel
generator fragility specifications are further subdivided by equipment size
into the following groups:
• 750 to 1,200 kVA – measured in units of 1000 kVA
• 1,200 to 2,000 kVA – measured in units of 1,500 kVA
All fragility specifications in Table B-20 are further subdivided by dominant

damage modes that include anchorage failure, internal equipment damage, or
the simultaneous occurrence of both. Refer to Section 2.5.13 for further
information on estimating the median floor acceleration associated with
anchorage failure and the use of that information to select the most
appropriate damage mode.
Peak floor acceleration (directional) of the supporting floor is the demand

parameter for these systems. The units of measure for these diesel generators
are indicated above.
2.5.20 Equipment and Furnishings (E20)
This fragility classification (E20) is summarized in Table B-21 and includes

various building furnishings and contents, such as filing cabinets, bookcases,
electronic equipment, fragile contents on shelves, and modular office
workstations.
One fragility specification is provided for modular office work stations

(E2022.001). The number of modular workstations per floor is user-
specified, as is the replacement cost of each station.
Peak floor acceleration (non-directional) of the supporting floor is the

demand parameter for these systems. Modular office work stations are
Three types of specifications are provided for valuable fragile objects on

shelves. These include: (1) unsecured objects on shelves with low friction
surface (E2022.010); (2) secured objects on shelves (E2022.011), and
(3) objects on cabinet shelves with latched doors (E2022.012). The quantity

Second Edition
and replacement cost of each group of objects of each type is specified by the
user.

demand parameter for these systems. Fragile objects on shelves are
Three categories are provided for electronic equipment (E2022.020):

(1) home entertainment equipment on shelves; (2) electronic equipment
mounted to walls (E2022.021); and (3) desktop electronic equipment
including computers, monitors, and stereos. The third category is subdivided
into equipment mounted on a slip-resistant surface (E2022.022) and
equipment mounted on a smooth surface (E2022.023). The quantity and
replacement cost of each item is specified by the user.

demand parameter for these systems. Electronic equipment is measured in
individual units, i.e., “each.”
Two types of specifications are provided for bookcases (E2022.102a –

E2022.106b). One is for unanchored bookcases and the other is for anchored
bookcases. The specifications are further subdivided by the size of the
bookcases. Fragility specifications are provided for five bookcase sizes.
Peak total floor velocity of the supporting floor is the demand parameter for
unanchored bookcases. Fragility functions for these components were
derived by calculation. Section 7.2.1 illustrates this calculation.
For anchored bookcases, the demand parameter is peak floor acceleration

(non-directional) of the supporting floor. Fragility specifications are defined
by the user and can be developed following the guidance provided in
Chapter 7. Users also need to define the number of bookcases of each type
on each floor and the replacement cost of the bookcase contents.
Two fragility specifications types are provided for filing cabinets

(E2022.112a – E2022.125b). One type is for unanchored filing cabinets and
the other is for anchored filing cabinets. The specifications are further sub-
divided depending on the size and orientation of the filing cabinets.
Specifications are provided for two sizes each of vertical and lateral filing
cabinets filing cabinets.
Peak total floor velocity of the supporting floor is the demand parameter for
unanchored filing cabinets. Fragility specifications for these components
have been derived by calculation. Section 7.2.1 illustrates this calculation.

Second Edition
For anchored filing cabinets, the demand parameter is peak floor acceleration
(non-directional) of the supporting floor. Fragility specifications are user-
defined and can be developed following the guidance in Chapter 7. The user
needs to define the number of filing cabinets of each type on each floor. The
performance group unit of input is the number of file cabinets of each type
per floor.
2.5.21 Special Construction (F20)
This fragility classification (F20) is summarized in Table B-22 and includes

industrial type storage racks designed and installed to satisfy the seismic
design requirements enforced prior to 2007. Users must identify and input
the number of racks and the replacement cost of the rack contents.
Peak floor acceleration (non-directional) of the floor supporting the rack is

the demand parameter for storage racks. Storage racks are measured in units
of 50 lineal feet, i.e., the total length of storage racks (in lineal feet) divided
by 50.
2.5.22 Nonstructural Performance Groups
Nonstructural components are assembled into performance groups in a

fashion similar to structural components. The Normative Quantity
Estimation Tool provided in Volume 3 facilitates identification of
performance group quantity information and distribution of common
nonstructural building components.
To use this tool, users identify the floor area associated with each of 8
different occupancy types at each building level. The tool produces a floor-
by-floor list of the performance group quantities of nonstructural components
based on the floor area and occupancy type. The user should review values
listed by the tool to confirm that the components and quantities selected are
appropriate given the building size and quantity of discreet units required.
Appendix D provides detailed instructions on the use of the Normative
Quantity Estimation Tool.
Figure 2-9 shows the Building Definition table of the Normative Quantity
Estimation Tool completed with number of floors and square foot for each
occupancy present at each floor. Selection of occupancy categories for each
floor will generate a listing of nonstructural fragility specifications and the
quantities of components conforming to each specification as shown in
Figure 2-10. The results from the tool should be reviewed for
appropriateness with the building information available, such as seismic
category and year of construction, and revised accordingly.

Second Edition
Figure 2-9 Normative Quantity Estimation Tool, Building Definition Table.
Figure 2-10 Normative Quantity Estimation Tool, Component Summary Matrix.
The Normative Quantity Estimation Tool assigns a measure of variability to

the median quantities reported. For distributed systems, such as piping,
ducting, and ceilings, the quantity variability can take a normal or lognormal
form. Dispersion and coefficient of variation values are provided for these
components and can be assigned to the performance group quantity input into
PACT. For discrete items, such as boilers and cooling towers, dispersion
values are not provided. Users should adjust the whole integer quantity of
these items directly. For distributed systems, such as piping and partitions,
the workbook will typically provide quantities in decimal fractions of a unit.
If the fragility of such components is assumed to have correlated fragilities,
these decimal fractions can be entered into PACT directly. However, if the
fragility functions of such components are considered uncorrelated, only
integer values should be entered.
2.6 Collapse Fragility Analysis
Once the building performance model has been developed and input into
PACT, a building collapse fragility function must be developed to permit
assessment of casualties. The collapse fragility function expresses the
probability of building collapse, in one or modes, as a function ground
motion intensity.

Second Edition
Collapse fragility functions include definition of the median spectral
response acceleration at the building’s effective fundamental period Sâ (T )
and a dispersion β associated with collapse, the possible collapse modes
(expressed as a percentage of each floor area impacted by collapse and the
number of stories involved in the collapse), the probability of each collapse
mode given that collapse occurs, and the probability of incurring serious
injuries and fatalities in an area of collapse.
The following sections discuss three general approaches to developing

collapse fragility functions, as described in Volume 1, Chapter 6.
2.6.1 Nonlinear Response History Analysis
This approach uses results from a large number of nonlinear response history
analyses to identify the probability of building collapse at different seismic
intensities. A lognormal distribution is fit to the collapse cases generated for
each intensity. The Collapse Fragility Tool provided in Volume 3 is an
Excel workbook application that fits a lognormal distribution to collapse
statistics obtained from a series of nonlinear dynamic analyses at different
ground motion intensity levels.
The validity of the fragility parameters derived from this approach is

dependent upon the number and quality of analyses performed and the
number of intensities evaluated.
2.6.2 Simplified Nonlinear Analysis
Development of collapse fragility parameters using nonlinear static analysis

is best suited to low-rise buildings of regular configuration. The approach
relies upon archetypical incremental dynamic analysis (IDA) curves
developed from nonlinear single-degree-of-freedom models of varying
properties that are accessible through the Static Pushover 2 Incremental
Dynamic Analysis (SP2OIDA) Tool provided in Volume 3. This tool is based
on Vamvatsikos and Cornell (2006). The process for constructing a collapse
fragility function using this approach consists of the following steps:
Step 1. Develop an appropriate mathematical model of the building for non-

linear static analyses. The procedures and recommendations contained in
ASCE/SEI 41-06, Seismic Rehabilitation of Existing Buildings, (ASCE,
2007) for the nonlinear static procedure (NSP) can be used to generate the
mathematical model.
Step 2. Following the procedures of ASCE/SEI 41-06, perform nonlinear

static analyses to identify the sequence of simulated and non-simulated

Second Edition
failures for each principal building direction. Extend the analyses to a
deformation such that collapse is judged to occur.
Step 3. Construct a force-displacement (pushover) relationship in each

principal direction of building response.
Step 4. Approximate the force-displacement relationship(s) into idealized

piecewise linear representations that include the reference control points for
use as input parameters for the Static Pushover 2 Incremental Dynamic
Analysis (SPO2IDA) tool provided in Volume 3. The tool utilizes the
idealized pushover curve representation illustrated in Figure 2-11. Four
control points define the general characteristics of the piece-wise linear
curve.
Figure 2-11 SPO2IDA idealized pushover curve for

hypothetical structure.
The branch of the curve from the ordinate to point 1 is defined as the elastic
segment. Point 1 is defined as the yield point. The curve segment from point
1 to point 2 is termed the hardening segment, where point 2 is defined as the
point of peak strength. The segment from point 2 to point 3 is termed the
softening branch, where point 3 represents the onset of residual strength
response. The segment from point 3 to point 4 represents the residual
strength plateau and terminates abruptly at point 4. Point 4 represents the
ultimate deformation at collapse. Definition of the control points can be such
that slopes of the various branches are substantially different than those
shown in the figure, and control points can coincide to eliminate branches, if
appropriate.
After definition of the control points, the tool will require the following
additional input information:
• Building weight,
• Building height, and

Second Edition
• Fundamental period.
Step 5. Execute SPO2IDA and extract the median collapse capacity. The
median collapse estimate for Sa is found on the IDA results Tab as the last
row entry in the Sa column.
Step 6. Define the collapse fragility function using the median collapse
capacity from SPO2IDA and an assigned dispersion of 0.6.
The example below illustrates this procedure for a two-story, reinforced

masonry shear wall building with a reinforced concrete slab supported by
perimeter masonry walls and a system of reinforced concrete gravity frames,
as shown in Figure 2-12 and Figure 2-13. The building has been designed to
resist seismic forces prescribed by ASCE/SEI 7-10, Minimum Design Loads
for Buildings and Other Structures, (ASCE, 2010) including minimum base
shear and story drift limitations. Basic design characteristics are as follows:
R = 5, Tx = Ty = 0.13s, Cv = 0.20
For SDS = 1.0g, W = 1133 kips (per wall) and V =227 kips (per wall)
Figure 2-12 Plan of example structure.
Figure 2-13 Section of example structure.
Step 1. A mathematical model is constructed in general accordance with the nonlinear static analysis
provisions of ASCE/SEI 41-06.
Step 2. Perform nonlinear static analyses (for each principal direction) of the building to develop a
force-displacement characterization of response (pushover curve). For the example structure, the
response is identical about each principal building direction.

Second Edition
Step 3. Construct the force-displacement relationships for each principal building direction and plot
the data. Figure 2-14 illustrates the results of a pushover analysis of the subject structure.
For this example, collapse is a non-simulated rupture of flexural wall reinforcement. Reinforcement
rupture is judged to occur at an extensional strain of 0.05. Collapse is defined at a deformation level at
which 30% of the cross section reinforcement rupture.
Figure 2-14 Pushover curve for 2-story reinforced masonry building.
The pushover curve data is input into SPO2IDA using the SPO tab. Note that drift index is used
(Column C) to define deformation data. See Figure 2-15 for data input. Other data needed to
complete the input include:
• Building weight: 1,133 kips
• Building height: 20 feet
• Fundamental period: 0.26s
Step 4. Identify the control points used in SPO2IDA to generate an idealized pushover curve by
entering base shear and roof drift data into cells L30 through L33 and M30 through M33. These
control points should be selected to best match the actual pushover curve. Superposition of the
original curve and the control points is illustrated in the graphical window. See Figure 2-15.
Step 5. Execute SPO2IDA and extract the median collapse capacity. See Figure 2-16. The results are
found on the IDA results tab, as shown in Figure 2-16. The last row of the Sa column provides the
median estimate of spectral acceleration at collapse.

Second Edition
Step 6. Construct the fragility using the median estimate of Sˆ a (T ) = 1.8g and an assigned dispersion
of 0.6. See Figure 2-17.
Figure 2-15 SPO2IDA Input.
Figure 2-16 SPO2IDA Output.

Second Edition
2.6.3 Judgment-Based Approach
This approach to collapse fragility function development is best suited for

buildings that conform to the requirements of recent building codes. The
judgment-based process consists of the following steps:
Step 1. Estimate the highest value of base shear, V, the structure can resist in
each principal direction while just in full conformance with the provisions of
ASCE/SEI 7-10 and its reference standards.
Step 2. Calculate the effective value of the design spectral acceleration at

the structure’s fundamental period, SaD, at which the structure satisfies all
applicable ASCE/SEI 7 criteria as follows:
V
SaD = R
W
where V is the base shear determined in Step 1; W is the structure’s seismic

weight and R is the response modification coefficient of the structure as
defined in ASCE/SEI 7.
Step 3. Calculate the inferred median collapse capacity, Ŝa(T) at the Second Edition Update
building’s fundamental period, T, using the formula below: This example has been
updated in accordance with the
Ŝa(T) = 4SaD
updated definition of inferred
Step 4. Convert the computed median collapse capacity at the building’s median collapse capacity.
fundamental period, T, to that at T .
T
=Sâ (T ) S a (T ) , where T ≥ 0.6 seconds
T
=Sâ (T ) S a (T ) , where T < 0.6 seconds
Step 5. Construct the collapse fragility curve using the median values
derived in Step 4 with a dispersion of 0.6 assigned for regular, well
configured construction and 0.7 or higher for irregular structures or those
with configurations/attributes suggesting poorer seismic performance than
structures of a regular nature.
The example below demonstrates the derivation of a judgment-based

collapse fragility using the sample structure previously described.
As previously described, the structure is a 2-story reinforced masonry shear

wall building designed for ASCE/SEI 7 seismic forces with the following
design characteristics:
R=5, Tx = Ty = 0.13s, Cv = 0.20, SDS = 1.0g, W=1133 kips

Second Edition
Step 1. Define the highest value of base shear the structure can resist while
in full conformance with ASCE/SEI 7-10:
V = CvW = 0.20(1133) = 227 kips
Step 2. Calculate the effective value of the design spectral acceleration at
the structure’s fundamental period
V 227
S aD
= = R (5) 1.00 g
=
W 1133
Step 3. Calculate Sâ (T ) for T = 0.13 seconds
Sâ (T ) = 4C4 S aD
2R − 1 (2)(5) − 1
=C4 = = 2.40 g
R 5
Step 4. Convert Sâ (T ) to Sâ (T )
For 0.6 s ≤ T : Sâ (T ) =

Sâ (T )
Yielding Sâ (T ) = 2.40 g
Step 5. Assign a dispersion of 0.6 in recognition of the regular configuration

of the building.
Figure 2-17 Comparison of collapse fragility functions obtained with two

different approaches: SPO2IDA and Judgment.
Figure 2-17 shows the collapse fragility functions derived from both the
nonlinear static procedure using SPO2IDA and the judgment-based

Second Edition
procedure. As can be seen there is significant difference between the two
functions. This difference can be attributed to a combination of the
inaccuracies inherent in the present basis for code-specified response
modification factors and the assumptions inherent in the nonlinear static
procedure as to what constitutes collapse.
2.6.4 Collapse Modes and PACT Input
When buildings collapse, they can collapse in different modes. For example,
multi-story structures can experience single-story collapse in any of the
stories or the entire structure can collapse. To quantify the impact of partial
or total building collapse, it is necessary to define the various collapse modes
that can occur, and to identify the fraction of building floor area at each level
that experiences collapse in each mode. For each realization in which
collapse is predicted, PACT utilizes a matrix of collapse ratios to determine
the extent of building area involved in the collapse, and the number of
occupants at risk. The following steps can be used to define the collapse
ratios for input to PACT:
Step 1. Identify the likely collapse modes based on the results of analysis
and judgment. This includes identification of building portions that are
prone to collapse. Figure 2-18 illustrates some representative collapse modes
for a 2-story building.
Figure 2-18 Illustration of multiple collapse modes.
Step 2. For each potential mode of collapse, estimate a probability of

occurrence assuming the collapse modes are mutually exclusive, that is, the
probabilities of all modes must sum to a value of 1.0.
Step 3. For each collapse mode, identify the fractional area of each floor that
will either collapse directly or will be covered by debris from the collapsing
floor above. In the illustration of Figure 2-18, collapse mode 1 is a 1st story
collapse, in which 100% of the first story floor area is compressed, but none
of the second. Collapse mode 2 is a 2nd story collapse in which 100% of the

Second Edition
second story floor area is compressed but none of the 1st. Collapse mode 3 is
a total collapse in which 100% of the areas of both floors is involved in the
collapse. Collapse mode 4 is a partial collapse of the 2nd story, in which
perhaps 40% of the 2nd floor area is subject to space compression.
When collapse occurs, PACT requires definition of the Mean Fatality Rate
(and associated COV) and the Mean Injury Rate (and associated COV).
The Mean Fatality Rate is the fraction of occupants inhabiting the impacted
collapse area who are fatally injured by the collapse. Users should select
values based on judgment considering the types of debris likely to be
generated given the construction type and its lethality. A similar judgment is
required for Mean Injury Rate. These two values should sum to a value of
1.0.
2.7 Residual Drift Fragility
The final building characteristic necessary for PACT input is a measure of

the building’s global reparability. This measure of reparability is expressed
on the Residual Drift tab (See Figure 2-19).
This fragility function expresses the probability that for a given value of
residual story drift ratio, the building will be repaired or not. As the value of
residual drift becomes larger, it will become impractical to repair the
damaged building components while ignoring the permanent building
distortion. Building recentering to correct residual drift is not generally
practical and therefore large residual drifts can result in the building damage
state being irreparable.
PACT provides an assumed fragility function for residual drift ratio that uses
a median value for irreparability of 1% residual story drift and a dispersion of
0.3, as shown in Figure 2-19. Volume 1 Appendix C provides discussion of
residual drift.
Construction of the building performance model is complete with the

definition of the Residual Drift Median and Residual Drift Dispersion.

Second Edition
Figure 2-19 PACT Residual Drift tab.

Second Edition
Chapter 3
Building Analytical Model and
Performance Assessments
3.1 Introduction
This chapter presents procedures necessary to evaluate a building’s probable

performance for different assessment types with different analysis methods,
and illustrates data entry into the Performance Assessment Calculation Tool
(PACT) provided in Volume 3. Each of the three assessment types
(intensity-based, time-based, and scenario-based) can be executed using
either the simplified analysis procedure or the more detailed nonlinear
response history analysis method. Figure 3-1 illustrates the steps common to
all assessment types and analysis methods. Chapter 2 provided guidance for
assembling the building performance model.
Figure 3-1 Performance assessment procedure

showing steps covered in this
chapter in shaded box.
FEMA P-58-2 3: Building Analytical Model and 3-1

Second Edition Performance Assessments
3.2 Building Analytical Model
An analytical model of the building is required for both the nonlinear

response history and simplified analysis approaches. Volume 1, Chapter 5
provides guidance on building and component modeling. This section
summarizes some basic modeling considerations that support execution of
the analyses presented in later sections.
For use in nonlinear response history analysis, a building should be modeled

as a three-dimensional assembly of components including foundations and
soils, as appropriate. Two-dimensional models are not recommended, and if
used, an individual model is required for each of the building’s primary
response directions, with ground motion pairs applied, such that one
component is used in the model representing direction 1 and the second in
the model representing direction 2.
Components should be modeled over their full range of expected

deformation response using hysteretic properties based upon test data.
Where test data are lacking, idealized backbone curves, as described in
ASCE/SEI 41-13, Seismic Evaluation and Retrofit of Existing Buildings,
(ASCE, 2014) can be used. The model should explicitly include component
strength and stiffness degradation. Volume 1, Chapter 5 provides a more
detailed discussion of component degradation modeling.
Median values of component properties, such as yield deformation, yield

strength, and capping strength, should be used. Mean or expected values
may be substituted where insufficient information is available to accurately
capture a median value.
Models should include dead and expected live loads acting throughout the
building. Live loads can be taken as 25% of the code specified design live
load without further reduction for influence area.
Equivalent viscous damping should generally be limited to 1% to 5% of

critical damping to account for the effects of nonstructural components and
soil-structure interaction not explicitly included in the analytical model. The
majority of energy dissipated during building response should be generated
by the hysteric behavior of the modeled components.
The response of relatively stiff buildings on relatively soft soils and buildings
with a basement that extends more than one story beneath grade can be
significantly affected by soil-structure interaction. Soil-structure interaction
and foundation embedment effects can be modeled explicitly or accounted
3-2 3: Building Analytical Model and FEMA P-58-2

Performance Assessments Second Edition
for using approximate methods. Volume 1, Appendix B provides guidance
and discussion of these topics.
Models should be structured to return values of the demand parameters

associated with the fragility specifications used to assess performance. For
example, fragility functions for eccentrically braced frame systems use link
beam rotation as the demand parameter. Models of eccentrically braced
frame structures should be configured to provide this demand value. Values
for story drift ratio should reflect only the racking drift ratio component of
total story drift ratio and not the contribution attributable to rotation from
lower floors of the building. For taller structures and those that include wall
systems, one means of obtaining this drift ratio is to include elastic cross
brace members with low stiffness whose elongation will not be influenced by
the rotational effects of the components below. The elongation (or
shortening of these members) can be used to calculate shear drift ratio.
To the extent possible, the model should explicitly include the range of
component deformation to failure. Where this cannot be done, users should
estimate the deformation limits at which failure of these components is likely
to occur. Users should carefully review analysis results and adjust them to
account for these non-simulated component failures.
3.2.2 Simplified Analysis
Simplified analysis uses a linear elastic structural model and an estimate of

the building’s lateral yield strength to estimate response, and determine
estimated median values of peak story drift ratios, residual drift ratios, peak
floor acceleration, and peak floor velocity. Building response is assumed to
be uncoupled along the principal horizontal building axes. Separate analyses
are performed for each response direction. Simplified analysis assumes the
following:
• Building response along each horizontal axis is uncoupled.
• The building is regular in plan and elevation (i.e., there are no substantial
discontinuities in lateral strength and stiffness).
• Story drift ratios do not exceed four times the corresponding yield drift
ratio, so that assumptions of elastic-plastic (bilinear) response at the
component level are not substantially compromised by excessive
degradation in strength and stiffness.
• Story drift ratios are less than 4%, beyond which P-∆ effects may
become important.

• The building is less than 15 stories in height, minimizing the potential for
significant higher mode effects.
If the building being assessed does not conform to these assumptions, the
simplified procedure may produce unreliable results.
Mathematical models should use best-estimate distributions of mass and

lateral stiffness. All elements that contribute significantly to the building’s
lateral strength or stiffness should be included in the model, whether or not
these elements are considered to be structural or nonstructural, or even part
of the seismic-force-resisting system. Strength and stiffness modeling
guidance provided in ASCE/SEI 41-13 for typical components can be used in
simplified analysis.
3.2.3 Demand Directionality
Users must enter peak floor acceleration, velocity, and story drift ratio values
into PACT in each direction. PACT computes non-directional drift ratio,
velocity and acceleration as 1.2 times the maximum of the two values
obtained at each floor along the two horizontal axes.
3.3 Intensity-Based Assessment
Intensity-based assessments provide expression of a building’s probable

performance for a specified intensity of shaking at the building site. The
specified shaking intensity is represented by a user-defined 5% damped
elastic acceleration response spectrum.
3.3.1.1 Define Earthquake Hazards
Intensity is characterized by pairs of ground motion records that are selected

and scaled for compatibility with a 5% damped elastic acceleration response
representing the shaking intensity. The spectrum is assumed to represent the
geometric mean of two orthogonal horizontal components of ground motion.
The guidance for ground motion selection and scaling provided in Volume 1
is summarized below and presented with the following steps:
Step 1. Determine the building’s fundamental translational periods in two

orthogonal directions ( T1X and T1Y ).
Step 2. Determine the effective first mode period T as the average of T1X
and T1Y , if response in the two directions are uncoupled and as the larger of
T1X or T1Y , if response is coupled.
Step 3. Using the target response spectrum, extract the S a (T ) value.

Step 4. Select a candidate suite of recorded ground motion pairs having
geomean spectral shape similar to the target spectrum. Recorded ground
motions are available from a number of sources including: https://ngawest2
.berkeley.edu/ and www.cosmos-eq.org (last accessed March 15, 2018).
Step 5. Construct the geomean response spectrum for each ground motion
pair using Volume 1, Equation 4-1:
gm ( T )
S= S x (T ) × S y (T )
where S x (T ) and S y (T ) are orthogonal components of spectral response

acceleration at period T.
Step 6. Determine the periods Tmin and Tmax over which the suite of ground
motion pairs are scaled to be consistent with the target spectrum. Tmax is
obtained from the equation:
Tmax = 2T
Period Tmin should typically be taken as the smaller of 0.2 T1X or 0.2 T1Y . If
substantial damage can occur due to response in modes having periods
smaller than Tmin, e.g., in high rise buildings where multiple higher modes
may contribute significantly to the acceleration and drift response, select Tmin
sufficiently small to capture these important response modes.
Step 7. Compare each pair’s geomean spectral shape with that of the target
spectrum over the range Tmin to Tmax. Select ground motion pairs with
geomean spectra that are similar in shape to the target response spectrum
within that range. Discard ground motion pairs that do not fit the shape of
the target spectrum adequately.
Step 8. Amplitude-scale both components of each ground motion pair by the

ratio of S a (T ) obtained from the target spectrum to that of the geomean
spectrum for the pair.
Figure 3-2 shows scaled spectra for a suite of 5 motions selected for
intensity-based assessment of a structure with T = 0.5 seconds. The ground
motion pairs were selected with geomean spectral shape similar to the target
(a code-spectrum with Sa(0.2) = 1.0g and Sa(1.0) = 0.6g) over the period
range 0.1 second to 1.0 second. For the selected suite of motions and scale
factors, the average value of Sa for the suite is 98% of the target value and
over the range of periods (0.1 second to 1 second) the average ordinate of the
geomean spectrum has an error of 1.5%. Note that at any period, some
records are substantially higher than the target spectrum and some, less.

Figure 3-2 Example spectra for scaled suite of 5 motions.
The number, n, of ground motion pairs necessary to obtain valid estimates of

median response depends on the extent that the geomean spectral shape of
the scaled motions matches that of the target spectrum in the period range
Tmin to Tmax. Limited study suggests that when a good match between the
spectral shape of the selected motions and the target spectrum is obtained,
use of relatively few records can provide a reasonable prediction of median
response along the building height. When ground motions are selected
without consideration of spectral shape, analyses using eleven pairs of
motions, scaled in accordance with the recommendations of Volume 1,
Chapter 4 can provide a reasonable estimate (+/-20%) of median response
with 75% confidence (Huang et al., 2008). Use of fewer than 7 pairs is not
recommended regardless of the goodness of fit of the spectra to that of the
target because of the extremely poor predictions of record-to-record
variability obtained from small suites of ground motions.
3.3.1.2 Analyze Building Response
Response calculations are performed with the n ground motion pairs

described in Section 3.3.1.1. Peak transient drift ratios, peak floor
accelerations, peak total floor velocities, and residual drift ratios are obtained
as follows:
Step 1. Execute an analysis of the structure for each of the ground motion
pairs.
Step 2. Extract the peak transient drift ratio, peak total floor velocity, and
peak total floor accelerations from the analysis for each principal building
direction and at each floor and story.

Note that many structural analysis software packages do not provide peak
total acceleration or peak total velocity output. Peak total floor acceleration
is the acceleration of a floor relative to a fixed point in space, rather than
relative to the ground. Peak total floor velocity is the velocity of a floor
relative to a fixed point in space, rather than relative to the ground. If the
software used to perform the analysis does not provide total acceleration or
total velocity as an output quantity, these must be derived from the output
that is available. At each analysis time step, peak total velocity can be
derived as the first derivative of the total displacement, taken as the
difference between two subsequent total displacement values divided by the
time step. Note that the total displacement must be determined as the
difference between the story floor displacement and the ground
displacement, on a time step by time step basis. Peak total acceleration can
be derived in a similar manner as the second derivative of total floor
displacement of first derivative of velocity.
Step 3. Estimate residual drift ratio from the peak transient drift ratios
according to Volume 1, Equation 5-25:
∆r = 0 for ∆ ≤ ∆y
∆r 0.3 ( ∆ − ∆y )
= for ∆y < ∆ < 4∆y
∆=
r ( ∆ − 3∆ )
y for ∆ ≥ 4∆y
where:
Δ = values of median story drift ratio determined in Step 2
Δy = median story drift ratio calculated at yield

( )
At values of ground shaking intensity S a T that approach the structure’s
median collapse capacity, it is likely that one or more of the n sets of ground
motion pairs will result in prediction of structural collapse, evidenced by
solution instability, predicted story drift ratio that exceeds the structural
model’s range of analytical validity, or predicted demands that suggest the
onset of one or more non-simulated collapse modes, such as punching shear
failure of a flat slab–column joint that could lead to local collapse.
Volume 1, Chapter 6 provides additional discussion of non-simulated
collapse modes. While it is possible to obtain additional ground motion pairs
and perform additional analyses to replace the motions that produce collapse,
that is not always necessary, and for shaking intensities where the collapse
probability is high, it is impractical. The following approach is
recommended when analytically predicted collapse is encountered in q or

more sets of the n ground motion pairs selected for a ground motion
intensity:
1. Discard the analysis results from any ground motion pair that results in
predictions of collapse.
2. If the remaining number of ground motion pairs for which collapse did
not occur (n-q) is greater than or equal to 7, use the remaining (n-q)
analyses to obtain estimates of demand at this ground motion intensity.
No further analyses are required at this ground motion intensity level.
3. If the remaining number of ground motion pairs for which collapse did
not occur (n-q) is less than or equal to 7, then additional ground motion
pairs should be selected until at least 7 analyses that do not predict
collapse are obtained. It is important to note that if the ground shaking
intensity at which analysis is performed is associated with a collapse
probability greater than 0.5, a very large number of analyses may be
necessary to obtain 7 runs that do not predict collapse. At such ground
motion intensities, building performance will be dominated by the
probability of collapse. Therefore, it is recommended that ground
motion intensities associated with collapse probabilities greater than 0.5
not be used to determine demand. Volume 1, Chapter 6 provides
information on determining the collapse probability at different ground
motion intensities.
Step 4. Estimate residual drift ratio values using peak transient drift ratios
for each principal building direction as described above. Residual drift ratios
should not be directly extracted from the response history analysis results as
presently most response history analysis tools do not produce reliable
estimates of residual drift ratio. Volume 1, Appendix C provides expanded
discussion on this topic.
Step 5. Demand dispersions are assigned based on judgment as follows:

1. Select values from Volume 1, Table 5-1 and Volume 1, Table 5-2
(reproduced here as Table 3-1 and Table 3-2) for βc and βq
2. Calculate βm = βc 2 + βq 2
βm should not be taken greater than 0.5.
3. Assign a dispersion associated with residual drift, βRD, equal to 0.8.

Table 3-1 Default Descriptions and Values for βc
Building Definition and Construction Quality Assurance βc

Superior Quality, New Buildings: The building is completely designed and will
be constructed with rigorous construction quality assurance, including special
inspection, materials testing, and structural observation.
Superior Quality, Existing Buildings: Drawings and specifications are available 0.10
and field investigation confirms they are representative of the actual
construction, or if not, the actual construction is understood. Material
properties are confirmed by extensive materials testing.
Average Quality, New Buildings: The building design is completed to a level
typical of design development; construction quality assurance and inspection
are anticipated to be of limited quality.
0.25
Average Quality, Existing Buildings: Documents defining the building design
are available and are confirmed by visual observation. Material properties are
confirmed by limited materials testing.
Limited Quality, New Buildings: The building design is completed to a level
typical of schematic design, or other similar level of detail.
Limited Quality, Existing Buildings: Construction documents are not available 0.40
and knowledge of the structure is based on limited field investigation.
Material properties are based on default values typical for buildings of the
type, location, and age of construction.
Table 3-2 Default Descriptions and Values for βq
Quality and Completeness of the Analytical Model βq

Superior Quality: The numerical model is robust over the anticipated range of
response. Strength and stiffness deterioration and all likely failure modes are
explicitly modeled. Model accuracy is established with data from large-scale
component tests through failure. 0.10
Completeness: The mathematical model includes all structural components
and nonstructural components in the building that contribute to strength or
stiffness.
Average Quality: The numerical model for each component is robust over the
anticipated range of displacement or deformation response. Strength and
stiffness deterioration is fairly well represented, though some failure modes
are simulated indirectly. Accuracy is established through a combination of
judgment and large-scale component tests. 0.25
Completeness: The mathematical model includes most structural components
and nonstructural components in the building that contribute significant
strength or stiffness.
Limited Quality: The numerical model for each component is based on
idealized cyclic envelope curves from ASCE/SEI 41-13 or comparable
guidelines, where strength and stiffness deterioration and failure modes are
not directly incorporated in the model. 0.40
Completeness: The mathematical model includes structural components in
the seismic-force-resisting system.
3.3.1.3 Input Response and Calculate Performance
For intensity-based assessment, PACT requires input of n demand vectors,

one for each analysis, and the dispersion quantities discussed above. The

vectors should include values for each of the demand parameters required by
the set of performance groups contained in the building performance model.
The basic steps for demand parameter input are:
Step 1. Identify the number of demand vectors, n, to be input. This is equal

to the number of analyses that do not produce collapse.
Step 2. Input the value for each demand parameter for each level and
direction, from each analysis in the Structural Analysis Results tab shown
in Figure 3-3.
Figure 3-3 PACT Structural Analysis Results tab with response history drift
input.
Step 3. Identify and input the value of Sa (T ) and the dispersion in the
Residual Drift input tab shown in Figure 3-4.
Figure 3-4 PACT Residual Drift input.

The ground motion parameters used to assess response are the 5% damped
elastic spectral response accelerations at the building’s fundamental period in
each of two orthogonal response directions, obtained by performing the
following three steps:
Step 1. Identify the 5% damped elastic acceleration response spectrum that

represents the desired intensity of shaking.
Step 2. Determine the building’s fundamental translational period in each of

two orthogonal directions ( T1X , T1Y ).
Step 3. Obtain Sa( T1X ) and Sa( T1Y ) from the selected spectrum.
Analysis is used to produce estimates of story drift when the model is

subjected to specified static lateral forces. These story drifts, together with
estimates of the structure’s yield strength and the peak ground acceleration
and velocity are then used in regression equations to develop estimates of
median story drift ratio, floor acceleration, and velocity in each of two
orthogonal directions. Sections 3.3.2.2.1 through 3.3.2.2.6 below describe
the steps of the procedure described in Volume 1, Section 5.3.2. Each step is
illustrated with an example application using the 2-story reinforced masonry
wall structure introduced in Section 2.6.1.
3.3.2.2.1 Determine Pseudo Lateral Force
Determine the total pseudo-lateral load, V, as defined in Volume 1,

Equation 5-3 to be applied along each principal building axis (Vx, Vy):
V = C1C2Sa(T1)W1
For this illustration Sa(T1) is taken as 1.0g, T1 = 0.13 seconds, and W =

1133 kips (per wall).
W1 is taken as CmW with Cm from Table 3-1 of ASCE/SEI 41-13. For the
2-story shear wall structure, Cm is 1.0 and thus W1 = 1.0W = 1133 kips.
C1 is an adjustment factor for inelastic displacements; given by Volume 1,

Equation 5-5:

S −1
1+
C1 = for T1 ≤ 0.2sec
0.04a
S −1
1+
= for 0.2<T1 ≤ 1.0sec
aT12
= 1 for T1 > 1.0sec
where T1 is the building’s fundamental period in the direction under

consideration; the coefficient a is obtained from Table 3-3.
Table 3-3 Values of coefficient a

ASCE/SEI 7-16 Site Class a
A 130
B 130
C 90
D 60
E 60
and where S is a strength ratio given by Volume 1, Equation 5-6:

S a (T1 )W
S=
Vy1
In this equation, Vy1 is the story yield strength in the direction under
consideration. Vy1 can be calculated using the following three different
approaches.
Plastic Analysis Method
For a shear wall system, the equivalent lateral force distribution produces a
base moment, M0, on the wall. The ratio of this moment to the base shear
is designated the spear span, H0. Up until yielding at the wall’s base
occurs, the vertical force distribution procedure defined in the next step
(Section 3.3.2.2.2) can be used to determine the shear span. The shear
span can be used to define the relationship between base shear at yield and
wall moment at yield.
Structural systems that include concentrically braced steel frames, low

aspect ratio walls, or moment frames generally will not require an
estimation of the vertical distribution of forces on the building. The
calculation of story yield for these systems is predominantly dictated by
the story shear only. For taller structures, the effects of overturning may
reduce the story yield force and thus should be considered for all structural
systems.

As defined in Section 3.3.2.2.2, the story force distribution of a 1 kip base
shear is 0.63 kips at the roof and 0.37 kips at the second floor. For a 1 kip
base shear, the flexural demand at the base of the walls will be:
M0 = 0.63(20) + 0.37(10) = 16.3 kip-ft
Shear span is defined as 16.3/1 = 16.3 feet
The calculated yield capacity using the nominal yield properties of one of the
masonry walls in the building is equal to 5189k-1. The shear force is
required to produce this moment is
M y 5189
= = 318 kips
H 0 16.3
Since the calculated value of Vy1 is based on nominal rather than expected
values, a conversion is required. The expected strength to nominal strength
ratios given in ASCE/SEI 41-13 can be used for this purpose. In this case,
flexural yielding will be controlled by the yield strength of the reinforcing
steel, for which the expected strength to nominal strength ratio is 1.25,
providing an estimated shear yield strength of 1.25(318 kips) = 398 kips.
Nonlinear Static Analysis Method
Using the SPO2IDA Tool provided in Volume 3, the yield value, Vy1, can be
calculated as 480 kips.
Using Judgment
The lower bound value Vy1 is given by the following expression:

1.55 S a (T )W
Vy1 ≥
R
I
where R, and I are the design coefficients for the building’s structural system
obtained from ASCE/SEI 7-16, Minimum Design Loads and Associated
Criteria for Buildings and Other Structures (ASCE, 2017).
For the example structure
S a (=
T ) 1.0, =
W 1133,=
R 5,=I 1.0
1.55 (1.0 )(1133)

Vy1 ≥ 351.2 kips
=
5
The upper bound value Vy1 is given by:

Ω0 Sa (T )W
Vy1 ≤
R
I
where Ω0 is obtained from ASCE/SEI 7-16 as Ω0 =2.2
2.2(1.0)(1133)
=Vy1 = 498.5 kips
5
Summary
Plastic Analysis: Vy1= 398 kips
Nonlinear Static Analysis: Vy1 = 480 kips
Judgment: 351.2 kips ≤ Vy1 ≤ 498.5 kips
For illustration in the continuation of this example, Vy1 will be taken as 398
kips, based on the plastic analysis method above.
(1.0)(1133)
=
Returning to the calculation of S, S = 2.85
398
Since T1 < 0.2s
S −1
C1 = 1 +
0.04a
The structure is located on Site Class D soil, therefore from Table 3-3, a = 60
and
2.85 − 1
1+
C1 = 1.77
=
0.04(60)
For S greater than 1.0, C2 (an adjustment factor for cyclic degradation) is
given by Volume 1, Equation 5-7 as follows:
( S − 1)
2
C2 1+
= for T 1 ≤ 0.2 seconds
32
1 ( S − 1)
2
=1+ for 0.2 < T1 ≤ 0.7 seconds

800 T12
=1 for T1 > 0.7 seconds
Since T1 ≤ 0.2 s
( S − 1) (1.85)
2 2
C2 1+
= 1+
= 1.11
=
32 32
thus V = (1.77)(1.11)(1.0)(1133)
V = 2226 kips

3.3.2.2.2 Estimate Median Story Drift Ratio and Dispersion
The median story drift ratio is estimated following the steps below.
Dispersion is estimated by using the default values given in Volume 1,
Chapter 5.
Step 1. Determine the vertical distribution of force over the height of the
building according to Volume 1, Equation 5-8 where Fx is the pseudo lateral
load at each floor level x:
Fx = CvxV
where Cvx is defined according to Volume 1, Equation 5-9:
wx hxk
Cvx = N +1
∑wh
i =2
i i
k
where wi is the lumped weight at floor level i, hi (hx) is the height above the
effective base of the building to floor level I, and k is equal to 1.0 for T1 less
than or equal to 0.5 seconds.
The floor and story heights are defined in accordance with Figure 3-5:
Figure 3-5 Definition of floor, story numbers and floor heights above grade
for two-story building.
Thus h2 = 10ft and H = 20ft
The weight distribution is as illustrated in Figure 3-6:
Figure 3-6 Lumped weight distribution.
The weight distribution and the story heights yield a ratio of 0.63 for the roof
and 0.37 for the first floor.
FR = 0.63(2226) = 1402.4 kips
F2 = 0.37(2226) = 823.6 kips

Step 2. Compute floor displacements and story drift ratios by using a linear
model with lateral loads FR and F2.
Figure 3-7 illustrates the calculated floor displacements from linear analysis.
Figure 3-7 Results from linear analysis.
0.214
=∆1 = 0.00178
(10)(12)
0.414 − 0.214
=∆2 = 0.00167
( E )(12)
Calculated drift ratios are valid for both of the building’s principal axes
because of the building’s symmetry (see Figure 2-12).
Step 3. Obtain estimates of median story drift ratio, Δi*, at each story by
correcting for inelastic behavior and higher modes using Volume 1, Equation
5-10:
=∆i* H ∆i ( S , T1 , hi , H ) × ∆i i = 1 to N
where H ∆i ( S , T1 , hi , H ) is the drift modification factor for story i computed
using Volume 1, Equation 5-11:
2 3
hi +1  hi +1   hi +1 
ln ( H ∆i ) =a0 + a1T1 + a2 S + a3 + a4   + a5  H  , S ≥1, i =1 to N
H  H   
with T1, S, and H as previously defined.
For a shear wall building of less than 10 stories tall, coefficients a0, a1, a2, a3,
a4, and a5 are determined from Volume 1, Table 5-4 as:
a0 = 0.92, a1= -0.036, a2 = -0.058, a3 = -2.56, a4 = 1.39, a5 = 0
For i=1, h2 = 10 feet, H = 20 feet:
2
 10   10 
ln H ∆1 =− 0.92 0.036(0.13) − 0.058(2.85) − 2.56   + 1.39  
 20   20 
ln H ∆ 1 = −0.18248
H ∆1 = 0.833

=∆1* H=
∆1 ( ∆1 ) 0.833(0.00178)
∆1* = 0.00148
For i = 2, hroof =H = 20 feet:
2
 20   20 
ln H ∆2 =−
0.92 0.036(0.13) − 0.058(2.85) − 2.56   + 1.39  
20
   20 
ln H ∆2 = −0.420
H ∆2 = 0.657
=∆2* H=
∆2 ( ∆2 ) 0.657(0.00167)
∆2* = 0.00110
Step 4. As a result of uncertainty in ground motion, mathematical response
modeling, and material property variability, the actual building response is
likely to vary from estimated median response. These uncertainties are
characterized as record-to-record variability, βaΔ, and modeling uncertainty,
βm. Total dispersion, βSD, associated with the peak transient drift is given by
Volume 1, Equation 5-19:
β SD
= β a2∆ + β m2
Values for βaΔ, βm, and βSD are presented in Table 3-4. The first four columns
of this table are excerpted from Volume 1, Table 5-6, and the last column is
calculated with the above equation. For T1 = 0.13s and S = 2.85,
interpolation yields βSD = 0.47.
Table 3-4 Default Dispersions for Story Drift
Sa ( T1 ) W
S=
T1 (sec) Vy1 βaΔ βm βSD
≤1.00 0.05 0.25 0.26
2 0.35 0.25 0.43
0.2 4 0.4 0.35 0.53
6 0.45 0.50 0.67
≥8 0.45 0.50 0.67
≤1.00 0.1 0.25 0.27
2 0.35 0.25 0.43
0.35 4 0.4 0.35 0.53
6 0.45 0.50 0.67
≥8 0.45 0.50 0.67

Table 3-4 Default Dispersions for Story Drift (continued)
Sa ( T1 ) W
S=
T1 (sec) Vy1 βaΔ βm βSD
≤1.00 0.1 0.25 0.27
2 0.35 0.25 0.43
0.5 4 0.4 0.35 0.53
6 0.45 0.50 0.67
≥8 0.45 0.50 0.67
≤1.00 0.1 0.25 0.27
2 0.35 0.25 0.43
0.75 4 0.4 0.35 0.53
6 0.45 0.50 0.67
≥8 0.45 0.50 0.67
≤1.00 0.15 0.25 0.29
2 0.35 0.25 0.43
1 4 0.4 0.35 0.53
6 0.45 0.50 0.67
≥8 0.45 0.50 0.67
≤1.00 0.15 0.25 0.29
2 0.35 0.25 0.43
1.5 4 0.4 0.35 0.53
6 0.45 0.50 0.67
≥8 0.45 0.50 0.67
≤1.00 0.25 0.25 0.35
2 0.35 0.25 0.43
2.00+ 4 0.4 0.35 0.53
6 0.45 0.50 0.67
≥8 0.45 0.50 0.67
3.3.2.2.3 Estimate Median Peak Floor Acceleration and Dispersion
Peak acceleration at the building base is taken as peak ground acceleration.

At other floor levels, the median peak floor acceleration is estimated by
following the steps below. Dispersion is estimated by using the default
values given in Volume 1, Chapter 5.
Step 1. Obtain estimates of median peak floor acceleration, ai*, at each story
above the building base by correcting peak ground accelerations using
Volume 1, Equation 5-12:

=ai* Hai ( S , T , hi , H ) × PGA i = 2 to N + 1
For this example, PGA = 0.40g.
Hai is the acceleration correction factor for floor i given by Volume 1,

Equation 5-13:
2 3
hi h  h 
ln ( H ai ) =a0 + a1T1 + a2 S + a3 + a4  i  + a5  i  , S ≥ 1 , i =2 to N + 1
H H H
with T1, S, hi and H as previously defined.
a4 ,and a5 are found in Volume 1 Table 5-4 as:
a0 = 0.66, a1 = -0.15, a2 = -0.084, a3 = -0.26, a4 = 0.57, a5 = 0
For i = 1 (at grade):
a1* = PGA
a1* = 0.40g
For i = 2, h2 = 10 ft, H = 20ft
h2 10
= = 0.5
H 20
0.66 − 0.15(0.13) − 0.084(2.85) − 0.26(0.5) + 0.57 (.5 )

2
ln H a2 =
ln H a2 = 0.4136
H a2 = 1.512
=a2* 1.512(0.4)
= 0.605g
For i = 3 (roof), hroof = H = 20ft
h3 20
= = 1.0
H 20
0.66 − 0.15(0.13) − 0.084(2.85) − 0.26(1.0) + 0.57 (1)

2
ln H a3 =
ln Ha3 = 0.7111
Ha3 = 2.036
=a3* 2.036(0.4)
= 0.814g
Step 2. Total dispersion, βFA, associated with peak floor acceleration is given
by Volume 1, Equation 5-20:
β FA
= β aa2 + β m2

where values for βaa and βm are presented in Table 3-5. The first four
columns of this table are excerpted from Volume 1, Table 5-6, and the last
column is calculated with the above equation.
For T1 = 0.13s and S = 2.85, interpolation yields βFA = 0.31.
Table 3-5 Default Dispersions for Peak Floor Accelerations
Sa ( T1 ) W
S=
T1 (sec) Vy1 βaa βm βFA
≤1.00 0.10 0.25 0.27
2 0.10 0.25 0.27
0.20 4 0.10 0.35 0.36
6 0.10 0.50 0.51
≥8 0.05 0.50 0.50
≤1.00 0.15 0.25 0.29
2 0.15 0.25 0.29
0.35 4 0.15 0.35 0.38
6 0.15 0.50 0.52
≥8 0.15 0.50 0.52
≤1.00 0.20 0.25 0.32
2 0.20 0.25 0.32
0.50 4 0.20 0.35 0.40
6 0.20 0.50 0.54
≥8 0.20 0.50 0.54
≤1.00 0.25 0.25 0.35
2 0.25 0.25 0.35
0.75 4 0.25 0.35 0.43
6 0.25 0.50 0.56
≥8 0.25 0.50 0.56
≤1.00 0.30 0.25 0.39
2 0.30 0.25 0.39
1.00 4 0.30 0.35 0.46
6 0.30 0.50 0.58
≥8 0.25 0.50 0.56
≤1.00 0.35 0.25 0.43
2 0.35 0.25 0.43
1.50 4 0.35 0.35 0.49
6 0.35 0.50 0.61
≥8 0.25 0.50 0.56

Table 3-5 Default Dispersions for Peak Floor Accelerations
(continued)
Sa ( T1 ) W
S=
T1 (sec) Vy1 βaa βm βFA
≤1.00 0.50 0.25 0.56
2 0.45 0.25 0.51
2.00+ 4 0.45 0.35 0.57
6 0.40 0.50 0.64
≥8 0.35 0.50 0.61
3.3.2.2.4 Estimate Median Peak Total Floor Velocity and Dispersion
The median peak total floor velocity is estimated by following the steps
below. Dispersion is estimated by using the default values given in
Volume 1, Chapter 5.
Step 1. A reference floor velocity at level i is obtained from Volume 1,

Equation 5-18:
T1 Vy1 δ 
vsi PGV + 0.3
= ( Γ 1) i 
2π W1 / g  δr 
where:
T1 = 0.13 seconds
Vy1= 398 kips
W1 1133
= 1.0W = kips
g g
W
Γ1 = 1 = 1.0
W
0.414
δr = = 0.00173
(20)(12)
Determine PGV
S a= = 0.4 ( 386 in / sec

(1) 0.4g = 2
) 154in / sec2 from site hazard data.
Sv (1)
=
1S a (1)
=
(154in / =
sec ) 2
24.5in / sec
2π 2π
Sv (1) 24.5in / sec
PGV
= = = 14.5in / sec
1.65 1.65

For i = 1: For the first floor or grade level, peak total floor velocity equals
the peak ground velocity.
vs1 = PGV = 14.5 in/sec
For i = 2:
Δ1 =0.00178
0.13  398   0.00178 
=vs2 14.5in / sec + ( 0.3)   (1.0 )  
2π  1133 / 386   0.00173 
vs2 = 15.3 in / sec
For i = 3:
Δ3 = 0.00167
0.13  398   0.00167 
=vs2 14.5in / sec + ( 0.3)   (1.0 )  
2π  1133 / 386   0.00173 
vs2 = 15.3 in / sec
Step 2. Obtain estimates of median peak total floor velocity by correcting

for inelastic behavior and higher modes using Volume 1, Equation 5-16:
=vi* Hvi ( S , T , hi , H ) × vsi i = 2 to N + 1
where Hvi is the velocity modification factor for floor i, given by Volume 1,
Equation 5-17:
2 3
hi h  h 
ln ( H vi ) =a0 + a1T1 + a2 S + a3 + a4  i  + a5  i  , S ≥ 1, i =2 to N + 1
H H h
with S, T1, hi, and H as previously defined and S >1.
a4, and a5 are found in Volume 1, Table 5-4 as:
a0 = -0.033, a1 = -0.085, a2 = 0.055, a3 = -0.52, a4 = 0.47, a5 = 0
For i = 1, v1* = PGV = 14.5 in/sec
For i = 2, h2 = 10 feet
2
 10   10 
ln( H v2 ) =
−0.033 − 0.085(.13) + 0.055(2.85) − 0.52   + 0.47  
 20   20 
ln( H v2 ) = −0.0298
H v2 = 0.971
=vs2* (=
0.971)15.3in/sec 14.9in/sec
For i = 3, hroof = H = 20 feet

ln( H v3 ) = 0.0627

H v3 = 1.065
= *
vs3 (1.065
= )15.3in/sec 16.3in/sec
Step 3. Total dispersion, βFV, associated with peak total floor velocity is
given by Volume 1 Equation 5-21:
β FV = β av2 + β m2
Values for βav, βm, and βFV are presented in Table 3-6. The first four columns
of this table are excerpted from Volume 1, Table 5-6, and the last column is
calculated with the above equation. For T1 ≤ 0.2s and S = 2.85, interpolation
yields βFV = 0.55.
Table 3-6 Default Dispersions for Total Peak Floor Velocity
Sa ( T1 ) W
T1 S=
(sec) Vy1 βav βm βFV
≤1.00 0.50 0.25 0.56
2 0.51 0.25 0.57
0.20 4 0.40 0.35 0.53
6 0.37 0.50 0.62
≥8 0.24 0.50 0.55
≤1.00 0.32 0.25 0.41
2 0.38 0.25 0.45
0.35 4 0.43 0.35 0.55
6 0.38 0.50 0.63
≥8 0.34 0.50 0.60
≤1.00 0.31 0.25 0.40
2 0.35 0.25 0.43
0.50 4 0.41 0.35 0.54
6 0.36 0.50 0.62
≥8 0.32 0.50 0.59
≤1.00 0.30 0.25 0.39
2 0.33 0.25 0.41
0.75 4 0.39 0.35 0.52
6 0.35 0.50 0.61
≥8 0.30 0.50 0.58
≤1.00 0.27 0.25 0.37
2 0.29 0.25 0.38
1.00 4 0.37 0.35 0.51
6 0.36 0.50 0.62
≥8 0.34 0.50 0.60

Table 3-6 Default Dispersions for Total Peak Floor Velocity
(continued)
Sa ( T1 ) W
T1 S=
(sec) Vy1 βav βm βFV
≤1.00 0.25 0.25 0.35
2 0.26 0.25 0.36
1.50 4 0.33 0.35 0.48
6 0.34 0.50 0.60
≥8 0.33 0.50 0.60
≤1.00 0.28 0.25 0.38
2 0.21 0.25 0.33
2.00+ 4 0.25 0.35 0.43
6 0.26 0.50 0.56
≥8 0.26 0.50 0.56
3.3.2.2.5 Estimate Median Residual Story Drift Ratio and Dispersion
The approach for estimating median residual story drift ratio presented in
Volume 1, Section 5.4 is summarized in Table 3-7 for different structural
systems and illustrated below.
Table 3-7 Procedures for Estimating Median Story Drift Ratio at Yield
Structural System Procedure
Steel concentrically Calculate story drift ratio at which the story shear equals the
braced frames and low expected yield strength of the braces or shear strength of
aspect ratio wall systems the walls
Moment frames Calculate story drift ratio which produces a story shear that
results in:
• beams or columns reach plastic moment capacity
(include axial forces)
• beam-column joint reaches its expected yield strength in
shear
Slender shear walls Calculate the story drift ratio at which the walls attain their
(governed by flexural expected flexural strength using a lateral force distribution
yielding) that reflects the dynamic response of the building
Step 1. Estimate of the median residual drift ratio is generated as follows:

=∆r* 0 ∆i* ≤ ∆y
∆
=*
r 0.3 ( ∆i* − ∆y ) ∆y* < ∆i* < 4∆y
( ∆i* − 3∆y )
∆r* = ∆i* ≥ 4∆y

where ∆i* is the median transient drift ratio and Δy is the median story drift
ratio calculated at yield.
Following the guidance provided in Table 3-7 for a shear wall system
(governed by flexural yielding), Δy can be estimated with ∆1* = 0.00148 ,
∆2* = 0.00110 , and Vy1 = 398 kips, where Vy1 equals the base shear at which
the base of the wall is expected to reach its flexural yield capacity and the
lateral force distribution reflects the dynamic response of the building.
Using the linear model, the building displacements can be scaled for a base
shear of 398 kips as follows:
0.414(398)
=δ ry = 0.074 inches
2226
0.214(398)
=δ 1y = 0.038 inches
2226
0.038
∆1y
= = 0.00032
10(12)
∆* 0.00148
=1
= 4.63
∆y 0.00032
Residual drift ratios are thus predicted as follows:
∆ r1* =
0.00148 − 3(0.00032) =
0.00052
∆ r2* =
0.00110 − 3(0.00032) =
0.00014
Note that the simplified analysis illustrated in this example was developed
for application to structures with peak transient drift demands less than or
equal to 4 times the yield drift. Beyond this level, simplified analysis may
become less valid. If the building is experiencing such large inelastic
demand, the nonlinear analysis method should be considered in order to
obtain more valid results.
Step 2. The total dispersion associated with median values of residual drift
ratio should be taken as βRD = 0.8. This value includes both βa and βm.
Simplified building response analyses produce estimated median values of

peak story drift ratios, ∆i* , peak total floor accelerations, ai* , peak total floor
velocities, v si* , and residual drift ratios in each principal building response
direction. PACT uses this information to make damage state assessments for
all building components contained in the building model. All demands

except residual drift are input via the Structural Analysis Results tab, as
shown in Figure 3-8.
Figure 3-8 PACT Structural Analysis Results tab, intensity-based assessment.
Radio buttons and fields on the upper portion of the tab should be modified
by the user to reflect:
• Assessment Type: Intensity
• Analysis Type: Simplified (Linear)
• Typical Number of Demand Vectors: This number indicates to PACT
the total number of demand vectors that may be input for each intensity
for which an assessment will be performed. The actual number of
vectors input in the lower portion of the screen can be less. Each
intensity included in the assessment can have a different number of
demand vectors.
• Number of Realizations: PACT uses a Monte Carlo process to
determine the probability of incurring consequences. If an insufficiently
large number of realizations are performed for each demand vector, the
indicated probability distribution may be quite inaccurate and can change
significantly each time PACT is run with the same data. The PACT
default setting is 200 realizations. This number of realizations has
generally produced stable results for repair cost and repair time. At least
500 realizations should be used for obtaining estimates for casualties and
unsafe placards. Use of larger numbers of realizations will increase

PACT execution time slightly, however, both storage time and space can
increase significantly.
• Non-Directional Conversion Factor: This factor sets the estimated
non-directional value of demand (story drift ratio, acceleration) as a
multiple of the maximum value of either of the directions. It is set at a
default value of 1.2, which is believed appropriate for most structures.
Under the Identify Intensity section, the user will need to depress the + Add
New Intensity button to prompt the setup for inputting intensity analysis
results. The user can then input the Intensity ID to identify the intensity in
the output files.
The spectral acceleration value at the fundamental building period in each

direction is entered to enable PACT to determine the probability of collapse.
When nonlinear response history analyses are performed, demand vectors are
first input in the Number of Demand Vectors field, then the building
Direction and the Demand Type are identified. The number of demand
vectors includes the results to be input for drift, floor acceleration, and floor
velocity in each building direction. The user can also input other demand
parameters assignable to direction 1, direction 2, or non-directional. For the
simplified analysis approach, there will only be one demand vector
containing drifts, floor accelerations and floor velocities for each building
direction and each floor or story. Results should be input for all demand
types and directions that are used by the fragility specifications selected to
define damageable building components.
For each demand vector, the associated Dispersion is input to the right of the
median values. Results are input from the top of the building to the bottom.
Table 3-8 summarizes the results obtained for median and dispersion values
of peak story drift ratios for a 2-story reinforced masonry structure in Section
3.3.2.2.2. The values for direction 1 are utilized in Figure 3-8.
Table 3-8 Drift Ratio Vectors

Direction 1 Direction 2
∆* ∆*
Floor Story θ β θ β
2 to 3 2nd 0.00110 0.47 0.00110 0.47
1 to 2 1rst 0.00148 0.47 0.00148 0.47
The maximum residual drift value and dispersion are input for each intensity
in the bottom row of the Residual Drift tab, as shown in Figure 3-9. This

portion of the Residual Drift tab is not available for data entry until the
structural analysis results have been input on the Structural Analysis
Results tab.
3.3.3 Review Results
After completing the performance model and analysis results entry, the
building model is saved, and PACT can execute the performance assessment.
This process is initiated by selecting the Evaluate Performance button
(Figure 3-10) on the PACT Control Panel. This will bring up the PACT
Engine window (Figure 3-11).
The File menu in this window is used to open the building project file, and
the Run Evaluation button is pressed. Upon completion, the user can save
the results for review. The results can be reviewed by pressing the Examine
Results button in the PACT Control Panel (See Figure 3-10).
Figure 3-12 and Figure 3-13 display the PACT output representative of
laboratory occupancy with intensity-based assessment. Other performance
measures are available from the tabs at the top of the results screen
(Casualties, Repair Time, and Unsafe Placards). More detailed results can
be viewed in the Data Drill Down and Exports tab.

PACT provides a range of options for viewing assessment results on the
screen or printing out hard copy reports. In addition to providing composite
results considering all performance groups and demand input from both
building directions, PACT can parse results by performance group, direction,
story level, and realization. Please refer to Appendix C for additional
information regarding the use of PACT.
Figure 3-10 PACT Control Panel.
Figure 3-11 PACT Engine window.

Figure 3-12 PACT Repair Cost tab.
The upper portion of the Repair Cost tab (Figure 3-12) illustrates the
contribution to repair cost attributable to each performance group. By
selecting the Realizations button, a graphic representation of performance
group contributors to the repair cost for each realization can be viewed.
Figure 3-13 illustrates this output format.
Figure 3-13 indicates that collapse occurred for realizations 160 through 200.
Realizations 0 through 159 did not result in collapse and the individual
performance group contributions are as shown in the upper half of Figure
3-12. Figure 3-14 illustrates the same result shown in the upper half of
Figure 3-12 with the components grouped into the major fragility
specification categories. For this intensity, chilled water components are the
primary contributor to repair costs for the realization at which median loss
occurs. The contribution to loss for other realizations can be found by
changing the Fraction field at the lower left-hand corner of the screen to any
value between 0 and 1. Options for restructuring performance group viewing
can be found under the Tools menu at the top of the screen.

Figure 3-13 PACT Realizations tab.
Figure 3-14 PACT Repair Cost tab showing repair cost by performance group.
The lower half of Figure 3-12 presents the repair cost probability
distribution. The median value is reported as $777,142.86. Any repair cost

(or non-exceedance probability) can be entered into the boxes below the loss
curve and the corresponding non-exceedance probability (or repair cost) will
be reported upon pressing the “Enter” key. Note that as the costs and non-
exceedance probabilities are changed, the performance group contributions
may change as well.
Under the Data Drill Down and Exports tab, numerous options exist for
viewing and exporting data. Figure 3-15 shows this tab with data for a
representative 4-story laboratory building. By pressing the View Drilldown
Page button, the user can see the results for each realization, for each floor,
each direction, and each performance group.
Figure 3-15 PACT Data Drill Down and Exports tab.
The same output and viewing options are available for assessments
performed using nonlinear response history analysis.
3.4 Scenario-Based Assessment
Scenario-based assessment enables calculation of performance for user-

specified combinations of earthquake magnitude and site-to-source distance.
The selected earthquake magnitude and site-to-source distance are converted
into a shaking intensity distribution by a ground motion prediction equation
(or attenuation relationship) that is developed by performing regression
analyses on the values of intensity parameters obtained from strong motion
recordings of past earthquakes against distance, magnitude, and other
parameters. The estimate may include consideration of many variables
including fault type, soil type, and other factors.

Most ground motion prediction models provide a median estimate of
intensity and a measure of uncertainty. Once intensity is defined, the
performance assessment process is completed in a manner nearly identical to
that for intensity-based assessment.
There can be considerable variation in the predictions among attenuation

relationships depending upon the site conditions and the target scenario.
Volume 1, Chapter 4 and Volume 1, Appendix B provide discussion and
guidance for selecting an attenuation relationship for scenario-based
assessment.
This example utilizes the Open Seismic Hazard Analysis website

(http://www.opensha.org, last accessed March 15, 2018) that provides a
downloadable attenuation relationship plotter which can be used to generate
median spectra and dispersions for the following attenuation relationships for
any magnitude-distance pair selected:
• Boore and Atkinson
• Campbell and Bozorgnia
• Chiou and Youngs
It is noted that use of the Open Seismic Hazard Analysis (OpenSHA) tool
may require a revision to computer security settings to provide the results.
Alternatively, U.S. Geological Survey (USGS) has developed the Unified
Hazard Tool that can be used to develop seismic hazard curves in most
situations. The USGS tool is available at https://earthquake.usgs.gov
/hazards/interactive/ (last accessed March 15, 2018).
In 2014, a new set of ground motion prediction equations (Next Generation

Attenuation Relationships for Western U.S., NGA-West2) were developed
and used in the development of the seismic design maps in ASCE/SEI 7-16.
These new equations can be downloaded from the PEER Ground Motion
Database, available at http://peer.berkeley.edu/ngawest2/ (last accessed
March 15, 2018).
Some decision makers will be interested in a building’s performance if a

repeat occurrence of a historic event occurs. Parameters defining many
historic events are available from numerous sources including searchable
catalogs at https://earthquake.usgs.gov/earthquakes/search/ (last accessed
March 15, 2018).
Some decision makers may desire performance assessments for worst case
scenarios. The USGS posts information on best estimate maximum
magnitudes for active faults in the Western United States and the Inner

Mountain seismic region at http://earthquake.usgs.gov/hazards/qfaults/ (last
accessed March 15, 2018).
Volume 1, Chapter 4 presents a procedure for selecting and scaling ground

motions to a median spectrum obtained from the ground motion prediction
equation. The prediction model uncertainty obtained is assigned as an
additional component of total dispersion. This procedure is illustrated here
and involves the following steps:
Step 1. Determine the building’s fundamental translational periods in each

of two principal directions, T1X , T1Y .
Step 2. Determine the average first mode period T as the average of T1X and
T1Y if the mode shapes are uncoupled and as the larger of T1X or T1Y if the
mode shapes are coupled.
Step 3. Estimate the average wave velocity, Vs30, in the upper 30 meters (100
ft.) of soil at the site.
Step 4. Select an appropriate attenuation relationship that best matches the

fault characteristics, regional geology, and site conditions for the scenario
event and building site. Downloadable attenuation functions are available at
http://www.opensha.org and at https://ngawest2.berkeley.edu/.
Step 5. Determine the values of variables used in the selected attenuation

relationship.
Step 6. Use the attenuation relationship to generate the median value of

spectral acceleration and associated period-dependent dispersion over the full
range of potential structural periods (0.01 to 5 seconds).
Step 7. If the attenuation relationship does not include consideration of site

characteristics and class, adjust the spectrum to account for these effects,
using site response analysis, site class coefficients contained in
ASCE/SEI 7-16, or other suitable adjustment factors.
Step 8. Calculate the structural period range of interest for spectral scaling
as:
Tmin = 0.2T
Tmax = 2.0T

Step 9. Select n ground motion pairs with 5% damped geomean spectral
shape similar to the spectrum generated in Steps 6 and 7 over the period
range Tmin to Tmax.
Step 10. Amplitude scale the suite of records to match the target spectrum.
Volume 1, Appendix B presents a second, alternative procedure with which

ground motions are selected to represent the dispersion in intensity obtained
from the ground motion prediction equation.
Per Section 3.4.1.1, n analyses using the n scaled ground motion pairs are
performed. Determination of median values and dispersions for peak
transient drift ratio, peak floor acceleration, and residual drift ratio are
illustrated in the following subsections.
3.4.1.2.1 Determine Peak Transient Drift Ratios, Peak Floor

Accelerations, Peak Total Floor Velocities, and Residual
Drifts
Step 1. Execute an analysis of the structure for each ground motion pair.
Step 2. Extract the peak transient drift ratios, peak total floor accelerations
and velocities from each analysis for each principal building direction.
Note that many computer programs commonly used to perform nonlinear

dynamic analysis do not provide total floor acceleration or velocity. Floor
velocity (relative to the ground) can be obtained as the first derivative of
floor displacement, and can be computed with the following equation for
each time step:
xi , j − xi , j −1
vi , j =
∆t
where vi,j is the floor velocity at floor i, at time step j, and xi,j and xi,j-1 are the
displacements of floor i at time steps j and (j-1), respectively and Δt is the
time step.
If a velocity history for the ground motion is available, total velocity can be
obtained as the vector sum of the ground velocity and floor velocity at each
time step. If a velocity series for the ground motion is not available,
maximum floor velocity can be approximated from the equation:
(v )
2
V jmax
= jmax + PGV 2

where Vjmax is the approximate maximum total floor velocity at floor j, vjmax is
the maximum of the relative floor velocities at floor j, and PGV is the peak
ground velocity.
If the computer software used to perform dynamic analyses does not provide
total acceleration, it can be derived by taking the vector sum of floor
acceleration relative to the ground and ground acceleration at each time step.
Floor acceleration relative to the ground can be obtained by taking the
second derivative of the floor displacement history from the formula:
vi , j − vi , j −1
ai , j =
∆t
where vi,j and vi,j-1 are the floor velocities at floor i, for steps j and j-1.
Step 3. Estimate residual drift ratio in each direction from the peak transient
drift ratio as follows:
∆r = 0 for ∆ ≤ ∆y
∆r 0.3 ( ∆ − ∆y )
= for ∆y < ∆ < 4∆y
∆=
r ( ∆ − 3∆ )
y for ∆ ≥ 4∆y
where
Δ = values of median story drift ratio determined in Step 2
Δy = median story drift ratio calculated at yield
As noted in Section 3.3.1, residual drifts should not be directly extracted

from the response history simulation results.
3.4.1.2.2 Generate Simulated Response Dispersions
Assign demand dispersion values for βc and βq from Table 3-1 and Table 3-2.
Modeling dispersion, βm, is calculated as β c2 + β q2 . For scenario
assessments, an additional uncertainty is introduced to reflect the variability
associated with the predictive ability of the attenuation relationship. The
dispersions associated with the attenuation relationship can be used directly
for βgm, or if these are not evident, the values of Table 3-9 can be used.
The plot in Figure 3-16, obtained from the Attenuation Relationship Plotter
application available on http://www.opensha.org, illustrates the period-
dependent dispersion for the example application in Section 3.4.2.1. For the
example short-period structure ( T = 0.13s), the corresponding dispersion is
0.57. This value compares favorably with the default value provided in
Table 3-10 for Western North America, 0.60.

Table 3-9 Default Dispersions Ground Motion Variability, βgm
Central and
Western North Eastern United
T1 (sec) America States Pacific Northwest
<0.20 0.60 0.53 0.80
0.35 0.60 0.55 0.80
0.5 0.61 0.55 0.80
0.75 0.64 0.58 0.80
1.0 0.65 0.60 0.80
1.50 0.67 0.60 0.85
≥2.0 0.70 0.60 0.90
Figure 3-16 Standard deviation of Sa vs. T attenuation plot from

http://www.opensha.org.
Next, the ground motion dispersion, βgm, is combined with βc and βq to

produce βm for use in scenario assessments as:
βm = β c2 + β q2 + β gm
2
If alternate procedures for scaling ground motion records are used for
analysis, βgm need not be included in the total assigned dispersion since its
effects have been captured directly in the calculation process.

Structural analysis results are entered into PACT in a manner similar to that
for intensity-based assessment. Figure 3-17 illustrates the input for direction
1 drift values from analyses performed with 7 ground motion pairs scaled to
the scenario target spectrum. The modeling dispersion, βm, includes the βgm
term.
Figure 3-17 PACT Structural Analysis Results tab, scenario-based assessment,

nonlinear analysis.
The simplified analysis method requires spectral response accelerations

obtained from a 5% damped elastic response spectrum consistent with the
selected magnitude and distance scenario. This requires selection and use of
a ground motion prediction equation to define the scenario response
spectrum. Volume 1, Chapter 4 and Volume 1, Appendix B provide
information on selection and use of these models. The basic steps are
illustrated below:
The 2-story reinforced masonry wall building previously used to illustrate the
simplified analysis approach for intensity-based assessment is used in this
example. Assumed conditions are:
• Site soil class D
• Distance to fault (closest horizontal distance to the vertical projection of
rupture), rjb = 20 km
• Earthquake magnitude, Mw = 6.5
• Strike-slip fault

Step 1. Estimate the average shear wave velocity in the upper 30 meters of
site soils:
Vs30 = 800 ft/s = 243.8 m/s
Step 2. Select an appropriate attenuation relationship that best matches the

fault characteristics, regional geology, and site conditions of the scenario
event and the reference building site.
Boore & Atkinson 2008 function is selected on the Attenuation Relationship

Plotter application available at http://www.opensha.org, and the above data
are input. Figure 3-16 and Figure 3-18 indicate the standard deviation and
median prediction for Sa, respectively.
Figure 3-18 Median Sa vs. T attenuation plot from http://www.opensha.org.
Step 3. Define the value of variables used in the selected attenuation

relationship.
From previous analysis, T1X = T1Y = 0.13 seconds.
Step 4. Use the attenuation relationship to generate the median value of
spectral acceleration and associated period dependent dispersion over the full

range of potential structural periods (0.01 to 5 seconds), to derive the
estimated median response spectrum.
From Figure 3-18, Sa(0.13s) = 0.177g for both X and Y directions. The
corresponding dispersion is 0.57.
Step 5. If the attenuation relationship does not include consideration of site

characteristics and class, adjust the spectrum to account for these effects,
using site response analysis, site class coefficients contained in ASCE/SEI 7
or other suitable adjustment factors.
Attenuation relationships generally return spectral acceleration values that
reflect a geomean, S gm , of horizontal components (Sax, Say), where:
S gm = Sax Say
Simplified analysis requires application of a spectral acceleration ordinate

along each principal building axis. To convert the geomean spectrum to a
pair of spectral ordinates oriented along each principal building axis, the
process is as follows:
• Determine periods T1X and T1Y
• Using the response spectrum derived as described above, determine
Sa (T1X ) and S a (T1Y )
For simplified analysis both the value of spectral acceleration at the
structure’s effective first mode period, Sa(0.13s), and the peak ground
acceleration, Sa(0s), are needed. These values can be obtained from the plot
shown in Figure 3-18 or by reexecuting the Attenuation Relationship Plotter
application for peak ground acceleration as an intensity measure in lieu of Sa.
For predictions of peak total floor velocity, a median estimate of the peak
ground velocity is needed. The Attenuation Relationship Plotter application
can be reexecuted to return median estimates and dispersions for this
intensity measure in a fashion similar to that for peak ground acceleration.
The objective of the analysis is to produce estimated median values of peak

transient story drift ratios, peak floor accelerations, peak total floor velocities
and residual drift. In general, the process is the same as that for intensity-
based assessment. An adjustment to the process for assigning dispersion is
needed, to incorporate the variability associated with the ground motion
prediction, βgm, which can be assigned from Table 3-10 or can be taken
directly from the selected attenuation relationship.

3.4.2.2.1 Estimate Median Story Drift Ratio and Dispersion
Estimate the median story drift ratio according to Section 3.3.2.2.2. The
equation to calculate story drift ratio dispersion, βSD, is modified as follows:
β SD= β a2∆ + β m2 + β gm
2
where βgm is the dispersion associated with the ground motion.
3.4.2.2.2 Estimate Median Peak Floor Acceleration and Dispersion
Estimate the median peak floor acceleration at each level according to

Section 3.3.2.2.3. The equation to calculate floor acceleration dispersion,
βFA, is modified as follows:
βFA= βaa2 + βm2 + βgm

2
3.4.2.2.3 Estimate Median Peak Total Floor Velocity and Dispersion
Estimate the median peak floor velocity according to Section 3.3.2.2.4. The
equation used to calculate floor velocity dispersion, βFV, is modified as
follows:
βFV = βav2 + βm2 + βgm

2
3.4.2.2.4 Estimate Median Residual Story Drift Ratio and Dispersion
Calculate the median residual story drift ratio according to Section 3.3.2.2.5.
A value of 0.8 is to be used for the dispersion of residual story drift, βRD, for
intensity- or scenario-based assessments.
Demand vectors and dispersions are organized and entered into PACT in a
similar fashion as that described for intensity assessments in Section 3.3.2.3.
Note that βgm is incorporated into the total dispersion assigned to the median
demand values. Figure 3-19 illustrates the Structural Analysis Results tab
for the simplified analysis approach in scenario-based assessment.

Figure 3-19 PACT Structural Analysis Results tab, scenario-based assessment,
simplified analysis.
The PACT assessment results for scenario-based assessment are similar to

those produced for intensity-based assessment as illustrated in Section 3.3.
3.5 Time-Based Assessment
Time-based assessments express the probability of incurring impacts over a

period of time, considering all earthquakes that occur in that time and the
probability that they could occur. A site seismic hazard curve is used to
express the annual frequency of exceedance for the values of a selected
ground motion parameter. Ground motion parameters of interest for the
assessment methodology include spectral acceleration at the average
fundamental period of the building, peak ground acceleration, and peak
ground velocity.
The time-based assessment process consists of a sequence of intensity-based

assessments that span the range of potential seismic hazards from the lowest
seismic intensity that can result in meaningful loss to the upper limits of
intensity where total loss is highly probable. The process is performed by
dividing the seismic hazard curve into a number of equal segments (eight
segments are recommended). From each segment, a central value of spectral
acceleration is selected and the building is analyzed for each intensity. The
resulting curves express the annual frequency of exceedance for various
performance measures. The sections that follow detail the steps in this
process.

For time-based assessments, nonlinear response history analyses are

conducted for m sets of n ground motion pairs each, scaled to appropriate
intensity values using the procedures of Volume 1, Chapter 4. A value of 8
is recommended for m. For very weak buildings, m can be taken smaller
than 8. For buildings with high resistance, it may be necessary to increase
the number of segments or increase the range of segments to obtain stable
results.
Ground motion records for time-based assessments are generated as follows:
Step 1. Determine the building’s fundamental translational periods in two

orthogonal directions ( T1X and T1Y ).
Step 2. Determine the average first mode period T as the average of T1X and
T1Y if the mode shapes are uncoupled and as the larger of T1X , and T1Y if the
mode shapes are coupled.
Step 3. Generate a seismic hazard curve for the site for Sa (T ) . This can be
done by performing a site specific seismic hazard analysis or using hazard
data compiled by the USGS.
Step 4. Define values of Samin and Samax that span the range of interest for the
building.
Samin (T ) should reflect the onset of non-trivial impacts, and can be taken as
follows:
=Samin 0.05g for T ≤ 1.0s
0.05
=Samin g for T > 1.0s
T
Samax should reflect a maximum value for the selected performance measures,
but need not exceed Sâ (T ) , the building’s median collapse capacity. Section
2.6 provides guidance on estimating median collapse capacity.
Step 5. Divide the seismic hazard curve into an appropriate number of

intensity segments that span the range from Samin to Samax .
Step 6. For each segment defined in Step 5, identify the Sai (T ) value and the
mean annual frequency of exceedance at the midpoint of the segment. Note
that the difference between mean annual frequency of exceedance values at
the segment ends defines the segment annual occurrence frequency Δλi.

Step 7. Develop an appropriate response spectrum for each of the Sai (T ) and
λi value. Volume 1, Appendix B provides information on generation of
various types of spectra. For events with a relatively large frequency, the
uniform hazard and conditional mean spectra should have a similar shape and
are both appropriate for use. For small frequency (infrequent) events, use of
conditional mean or conditional spectra should improve estimates of drifts
and accelerations over those generated using uniform hazard spectra.
Step 8. Select and scale ground motion pairs to the amplitude of the target
spectrum at the Sa (T ) value similar to the process for intensity-based
assessment. Ground motion record pairs can be found at: https://ngawest2
.berkeley.edu/ and www.cosmos-eq.org. A minimum of 7 record pairs with
geomean spectral shape similar to the target spectrum should be used. If
records with appropriate spectral shape cannot be found, at least 11 record
pairs should be used.
Step 9. Repeat Step 8 for each hazard curve segment to obtain scaled
records for the full range of the segments.
Section 3.5.2.1 illustrates Steps 1 through 6. Chapter 5 illustrates execution

of Step 1 through Step 9 for a 4-story reinforced concrete frame structure.
Analyze the structure with the selected records for each spectral intensity
segment. Extract the peak transient drift ratios, floor accelerations, and peak
total floor velocities for each of the orthogonal building directions. Calculate
residual drift ratio and demand dispersions in the same manner as done for
intensity-based assessment in Section 3.3.1.2.
For spectral intensities that approach the structure’s median collapse capacity
Sâ (T ) , some ground motion pairs may produce predictions of collapse.
Discard the results for analyses that produce collapse. In this case, additional
analyses may be need. Section 3.3.1.2 and Section 5.7 provide further
guidance on this.
For time-based assessments, PACT requires input of n demand vectors for

each intensity assessment. Nonlinear response history analysis may produce
a different number of demand vectors for each spectral segment due to
occurrence of collapse for some records. Response vectors should include
values for each demand parameter utilized by the selected component
fragility specifications.

The structural analysis results for each intensity segment are input similarly.
Information defining the mean annual frequency values of each intensity
segment is input on the PACT Hazard Curve tab.
Ground shaking effects are characterized by spectral ordinates extracted from

a seismic hazard curve that spans the anticipated range of meaningful
building response. Volume 1, Chapter 4 sets forth a procedure which is
illustrated below.
The example uses the 2-story reinforced masonry wall structure previously
defined.
Latitude: 37.98699
Longitude: -122.61644
T1X = T1Y = 0.13 seconds
Site Class: B
Step 1. Determine the fundamental translational periods in two orthogonal

directions ( T1X and T1Y ).
T1X = T1Y = 0.13 seconds
Step 2. Determine the averaged first mode period T as the average of T1X
and T1Y if the mode shapes are uncoupled and as the larger of T1X and T1Y if
the mode shapes are coupled.
T = 0.13 seconds
Step 3. Define seismic hazard curve for site
The hazard curve can be constructed by using the following tools provided
by the USGS: (1) the Unified Hazard Tool (https://earthquake.usgs.gov
/hazards/interactive/); and (2) the Java Ground Motion Parameter
Calculator. This example uses the Java Calculator to construct the hazard
curve. The basic information necessary to generate the seismic hazard curve
using this approach includes the site latitude and longitude (accurate to three
decimal places), the average period, T , and the site class.

The USGS application develops seismic hazard curves at the following
discrete building periods: 0, 0.1, 0.2, 0.3, 0.5, 1.0, and 2.0 seconds. Figure
3-20 illustrates the spectral acceleration values at a period of 0.13 seconds
interpolated from the USGS values and their corresponding mean annual
frequency of exceedance. For a site class other than B, an adjustment of the
spectral values is required.
Step 4. Define Samin and Samax that span the building’s range of response
interest. Samin should reflect the onset of non-trivial impacts, and for repair
costs and repair time, can be taken as follows:
=Samin 0.05g for T ≤ 1.0s
0.05
=Samin g for T > 1.0s
T
1.00E+00
1 2 3 4 5 6 7 8
1.00E-01
Mean Annual Frequency of Exceedance
1.00E-02
1.00E-03
1.00E-04
1.00E-05
0 0.5 1 1.5 2
Spectral acceleration in g at T= 0.13s
Figure 3-20 Seismic hazard plot showing segments.
Similarly, Samax should reflect demand that will produce a maximum value
for the selected performance measures and can generally be taken as:
Samax = 2Sa (T )
where Sa (T ) is evaluated at a mean annual frequency of exceedance, λ, of
0.0004 but need not exceed Sa (T ) , the building’s median value of collapse
fragility. Section 2.6 provides guidance on estimating building median
collapse capacity.

For λ = 0.0004
Sa (T ) = 1.76g
2 × Samax = 2 × 1.76 = 3.52g > Sâ (T ) = 1.80 g
Set Samax = 1.80g
Step 5. Divide the hazard curve into 8 equal segments:

For the Sa (T ) range of 0.05g to 1.80g, the spectral segment interval is
1.80 − 0.05
= 0.21875 g
8
Figure 3-20 illustrates the 8 equal width segments spanning from 0.05g to
1.80g.
Step 6. For each segment defined in Step 5, identify the Sai (T ) value at the
segment midpoint and the associated mean annual frequency of exceedance,
at each segment end, λi, λj, and midpoint, respectively. Hazard input into
PACT uses segment midpoint Sai (T ) and λ values.
Table 3-10 summarizes the segment midpoint S a (T ) values and associated
mean annual frequency of exceedance values for each of the 8 spectral
segments.
Table 3-10 Segmented Seismic Hazard Values

Sai ( T ) λ
Segment at midpoint λi λj at midpoint
1 0.1594 0.1372 0.0215 0.0395
2 0.3781 0.0215 0.0095 0.0139
3 0.5969 0.0095 0.0059 0.0067
4 0.8156 0.0059 0.0026 0.0037
5 1.0344 0.0026 0.0016 0.0019
6 1.2531 0.0016 0.0009 0.0012
7 1.4719 0.0009 0.0005 0.0006
8 1.6906 0.0005 0.0004 0.0005
Step 7. The peak ground acceleration hazard curve is obtained using the
Java Ground Motion Parameter Calculator. This seismic hazard curve in
Figure 3-21 expresses the variation in mean annual frequency of exceedance
for peak ground acceleration (PGA).

1.00E+00
1.00E-01
Mean Annual Frequency of Exceedance

1.00E-02
1.00E-03
1.00E-04
1.00E-05
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Peak Ground Acceleration in g
Figure 3-21 Variation of mean annual frequency of exceedance for peak ground
acceleration in g.
Using the spectral segments defined in Table 3-10, Table 3-11 presents the
corresponding mid-point peak ground acceleration values.
Table 3-11 Segmented Peak Ground

Acceleration Values
Segment PGA
1 0.029
2 0.134
3 0.244
4 0.322
5 0.481
6 0.559
7 0.696
8 0.774
Analysis is performed for each intensity segment in the same fashion as for
intensity-based assessment (See Section 3.3.2.2). The simplified analysis
procedure is used to generate peak transient drifts, peak floor accelerations,
peak total floor velocities, and residual drifts. The median demand estimates
are assigned dispersions. Results include a vector of demands for each of the
m intensity segments. The illustration below produces results for Segment 1.
Median estimates for peak transient drift ratio, peak floor acceleration, peak
total floor velocity and residual drift are calculated similarly for each spectral
segment. The associated dispersions are similarly generated.

For Segment 1 as identified in Section 3.5.2.1:
S a (T ) = 0.159g
PGA = 0.05g
T = 0.13s, W = W1 = 1133k, Vy = 398 kips

1
Step 1. Generate median story drift ratio estimates
According to Volume 1, Equation 5-6:
Sa (T )W 0.159 (1133)
=S = = 0.45
Vy1 398
Since 0.45 ≤ 1.0 , according to Volume 1, Equation 5-7:

C1 = 1.0
C2 = 1.0
The load in each direction is computed using Volume 1, Equation 5-3:
V = C 1C 2 S a ( T ) W 1
V (1.0
= )(1.0 )( 0.159 )(1133) 180.1 kips
Using the results presented in Section 3.3.2.2, the resulting structural

displacements and drifts are scaled proportionately using the base shear ratio.
From previous analysis of the linear model:
V = 2226 kips, δr = 0.414 inches, δ2 = 0.214 inches
For Segment 1: Sa = 0.159,
180.1
=δr = ( 0.414 ) 0.034 inches
2226
and
180.1
=δ2 = ( 0.214 ) 0.017
2226
0.034
=∆1 = 0.000283
(10 )(12 )
0.034 − 0.017
=∆2 = 0.000142
(10 )(12 )

To obtain estimates of median story drift ratio, Δi*, at each story by
correcting for inelastic behavior and higher modes Volume 1, Equation 5-10
is used:
=∆i* H ∆i ( S , T1 , hi , H ) × ∆i i = 1 to N
2 3
h h  h 
ln ( H ∆i ) =a0 + a1T1 + a2 S + a3 i +1 + a4  i +1  + a5  i +1  , S ≥1, i =1 to N
H  H   H 
with T1, S, hi and H as previously defined.
a4, and a5 are determined from Volume 1, Table 5-4 as:
a0 = 0.92, a1 = -0.036, a2 = -0.058, a3 = -2.56, a4 = 1.39, a5 = 0
For i = 1, h2 = 10 feet, H = 20 feet:

2
 10   10 
0.92 0.036 ( 0.13) − 0.058 ( 0.45 ) − 2.56   + 1.39  
ln H ∆1 =−
 20   20 
ln H ∆1 = −0.043
H ∆1 = 0.957
∆1* = 0.957 ( 0.000283) =0.00027
For i = 2, hroof =H = 20 feet

2
 20   20 
ln H ∆2 0.92 0.036 ( 0.13) − 0.058 ( 0.45 ) − 2.56   + 1.39  
=−
 20   20 
ln H ∆2 = −0.281
H ∆2 = 0.755
∆2* = 0.755 ( 0.000142 ) =0.000107
From Table 3-4 for T = 0.13 seconds, S = 0.45:
βSD = 0.26
Step 2. Generate median floor acceleration estimates
Obtain estimates of median peak floor acceleration, ai*, at each story above
the building base by correcting peak ground accelerations using Volume 1,
Equation 5-12:

For this example, PGA = 0.05g.
Hai is the acceleration correction factor for floor i given by Volume 1,

Equation 5-13:
2 3
hi h  h 
H H H
with T1, S, hi , and H as previously defined.
a4, and a5 are found in Volume 1, Table 5-4 as:
a0 = 0.66, a1 = -0.15, a2 = -0.084, a3 = -0.26, a4 = 0.57, a5 = 0
For i = 1 (at grade):
a1* pga
= = 0.05
For i = 2, h2 = 10 ft, H = 20ft

2
 10   10 
0.66 0.15 ( 0.13) − 0.084 ( 0.45 ) − 0.26   + 0.57  
ln H a2 =−
 20   20 
ln H a2 = 0.6152
H a2 = 1.85
=a2* 1.85
= ( 0.05) 0.093g
For i = 3, hroof = H = 20ft
2
 20   20 
0.66 0.15 ( 0.13) − 0.084 ( 0.45 ) − 0.26   + 0.57  
ln H a3 =−
 20   20 
ln H a3 = 0.9127
H a3 = 2.49
=a2* 2.49
= ( 0.05) 0.125 g
From Table 3-5 for T = 0.13 seconds, S = 0.45:
βFA = 0.27
Step 3. Generate residual drifts.
From previous analysis:

∆2y 0.000283 =
= > ∆2* 0.00027
∆1y 0.000142 =
= > ∆1* 0.000107
Since peak transient drifts do not exceed the yield drifts, residual drift is zero
for Segment 1.
Analysis results are input into PACT in a manner similar to that for intensity-
based assessments. The total number of intensity data sets should encompass
all m hazard segments for all demand parameters in each principal building
direction. Once the demand vectors are input, the hazard curve data are input
into PACT using the Hazard Curve tab. The mean annual occurrence
frequencies associated with each of the 8 spectral intensities are input.
Figure 3-22 illustrates input of the spectral intensity information from Table
3-10 into PACT.
Figure 3-22 PACT Hazard Curve tab.
PACT results for time-based assessments can be reviewed in terms of annual

probabilities of non-exceedance for each performance measure. Chapter 5
illustrates sample results output for repair cost. Results can also be reviewed
by intensity segment. For each intensity segment results can be reviewed in
a similar fashion to those produced for intensity-based assessment as
presented in Section 3.3.3.

Chapter 4
Example Application:
Intensity-Based Performance
Assessment Using Simplified
Analysis
4.1 Introduction
This chapter provides an example application of the intensity-based

performance assessment methodology using simplified analysis in the
Performance Assessment Calculation Tool (PACT) with provided fragility
and consequence function data.
The work necessary to obtain a median repair cost estimate for earthquake
ground shaking with an average return period of 475 years (10% probability
of exceedance in 50 years) for an office building is presented here as follows:
• Obtain site and building description (Section 4.2)
• Select assessment type and performance measure (Section 4.3)
• Assemble building performance model (Section 4.4)
• Select analysis method and construct analytical building model (Section
4.5)
• Define earthquake hazards (Section 4.6)
• Analyze building response (Section 4.7)
• Input response data and calculate performance (Section 4.8)
• Review results for selected performance measures (Section 4.9)
4.2 Obtain Site and Building Description
The building site for the example building is located in Los Angeles,
California. The soil profile at the site is categorized as ASCE/SEI 7-10,
Minimum Design Loads for Buildings and Other Structures, (ASCE, 2010)
Site Class D, having a shear wave velocity of 259 m/s in the upper 30 meters.
The structure has plan dimensions of 120 feet by 180 feet. Floor-to-floor
height is 15 feet at the lower story and 13 feet at other stories. The structure
FEMA P-58-2 4: Example Application: Intensity-Based Performance 4-1

Second Edition Assessment Using Simplified Analysis
has reinforced concrete special moment frames around the building
perimeter. The floors and roof are two-way post tensioned flat slabs
(8 inches thick) supported by the perimeter moment frame and interior 18
inch square reinforced concrete columns on a 30 foot by 30 foot grid. Figure
4-1 and Figure 4-2 illustrate plan and elevation views. When entering into
PACT, direction 1 is arbitrarily aligned with the North-South axis and
direction 2 is aligned with the East-West building axis.
Figure 4-1 Plan view of example 4-story office building.
Figure 4-2 Typical frame elevation for example 4-story office building.
4-2 4: Example Application: Intensity-Based Performance FEMA P-58-2

Assessment Using Simplified Analysis Second Edition
The building is designed and detailed according to the requirements of
ASCE/SEI 7-05, Minimum Design Loads for Buildings and Other Structures,
(ASCE, 2005a), ACI 318-02, Building Code Requirements for Structural
Concrete and Commentary, (ACI, 2002), and the 2007 California Building
Code (ICC, 2007). Vulnerable nonstructural building features include
exterior glazing, gypsum board partitions, suspended acoustical ceilings, fire
sprinkler system, traction elevator, concrete roof tiles on a perimeter
mansard, hot and cold water piping, and HVAC ducting. Tenant furnishings
and equipment including office partitions and computers are not included in
the assessment.
4.3 Select Assessment Type and Performance Measure
Intensity-based assessment will be performed with the building performance

expressed in terms of repair costs as a percentage of total replacement cost.
4.4 Assemble Building Performance Model
The building performance model will be constructed in PACT by following

the sequence:
1. Provide project information
2. Provide building characteristics
3. Select fragility specifications and performance groups
4. Identify collapse fragility and collapse modes
5. Provide residual drift fragility
The following subsections illustrate the implementation of this sequence.
4.4.1 Project Information
The first set of PACT data are input on the Project Information tab, as
shown in Figure 4-3.
For this example, the regional and date cost multipliers are set equal to 1.
For any particular building, these values should be set to reflect the actual
building location and time value of escalation or devaluation effects. The
PACT default consequence functions reflect 2011 costs for the Northern
California region, which for this example, are deemed similar to those at the
building’s location in Southern California.
The Solver Random Seed Value is set to a non-zero value, 5, so that the
assessment model will produce the same answer for multiple executions of
the problem using the same input data.

Figure 4-3 PACT Project Information tab.
4.4.2 Building Information
At the top of the Building Information tab, the following information is

input, as shown in Figure 4-4:
• Number of Stories: 4.
• Total Replacement Cost: Estimated as $250/sf × 86,400 sf or
$21,600,000.
• Replacement Time: Estimated as 550 days.
• Core and Shell Replacement Cost: Estimated as $100/sf × 86,400 sf or
$8,460,000.
• Maximum Workers per Square Foot: Default value of 0.001 is used.
• Total Loss Threshold (as Ratio of Total Replacement Cost): Default
value of 1.0 is used.
• Carbon Emissions Replacement (kg): Estimated as: [$21, 600,000
(total replacement cost) × 0.414 kg CO2e/$ (environmental impact
factor) / 0.88 (inflation factor from 2013 to 2011, based on U.S. Bureau
of Economic Analysis)] or 10,161,818 kg CO2e.
• Embodied Energy Replacement (MJ): Estimated as: [$21, 600,000
(total replacement cost) × 5.729 MJ/$ (environmental impact factor) /
0.88 (inflation factor from 2013 to 2011, based on U.S. Bureau of
Economic Analysis)] or 140,620,909 MJ.
In the Most Typical Defaults section, the following fields are input:
• Floor Area: 21,600 square feet
• Floor Height: 13 feet

Variation in floor height is input via the Floor Number drop down selector,
which also permits input of non-typical floor areas.
Figure 4-4 PACT Building Information tab.
The Height Factor, which adjusts repair cost and effort for the difficulty of
performing work on upper floors of tall buildings, is set to 1 to reflect the
lack of substantial cost differential for work performed on different floors of
a low-rise building. Since the building is recently constructed, no lead-based
paints or asbestos containing materials are present and thus the Hazmat
Factor is set to 1. In this example, it is assumed that that if the building is
damaged, it can be vacated for repairs, so that increase of the cost and effort
of repair in an occupied space is not expected, therefore, the Occupancy
Factor is set equal to 1.
4.4.3 Population Distribution
Population distribution within the building is not an input required for the
estimation of repair costs.
4.4.4 Structural Components
Fragility specifications provided in PACT are utilized for both structural and
nonstructural components.
4.4.4.1 Structural Component Fragility Specifications
Structural components are input based on the basic building characteristics

previously described. To input structural fragility groups, the dropdown

menu of selections in the Component Fragilities tab, shown in Figure 4-5,
are utilized. These selections list the most typical specifications for the
building and will be used by PACT to generate a floor-by-floor distribution
of specifications. The selection process proceeds progressively from
foundations through super structure. The following information summarizes
the structural components included in the performance assessment model.
Figure 4-5 PACT input screen for beam/column joint fragility.
A10: Foundations. Foundations on this site will be considered rugged and

thus not vulnerable to earthquake damage. No fragility group for
foundations will be selected or input.
A20: Basement Construction. The example building has no basement. No

fragility group is selected.
B10: Superstructure.
B101: Floor Construction. Elevated floors (8 inch post-tensioned flat slabs)

will be considered rugged for in-plane (diaphragm) deformations.
Vulnerability to slab-column joint rotations will be included under reinforced
concrete elements (B104).
B102: Roof Construction. Roof construction (8 inch post-tensioned flat

slabs) will be considered rugged for in-plane (diaphragm) behavior, similar

to the floor construction. Vulnerability to slab-column joint rotations will be
included under reinforced concrete elements.
B104: Reinforced Concrete Elements. Appendix A provides a compilation of

available structural fragility specifications for reinforced concrete moment
frames. The following steps illustrate the selection of the appropriate
fragility specifications for the example building:
• Selections range from classification number B1041.001a through
B1049.133b, depending upon the framing system attributes.
• The building’s primary lateral-force-resisting frames are detailed in
accordance with a recent edition of ACI 318-02, Building Code
Requirements for Structural Concrete and Commentary, (ACI, 2002).
Therefore, the fragility specifications entitled “ACI 318 Special Moment
Frames,” ranging from B1041.001a through B1041.013b, are
appropriate.
• Detailing is such that joint failure is not expected, and thus the selection
range is narrowed to B1041.001a through B1041.003b. Beams occur on
one side of the columns at grids A-1, A-2, A-6, A-7, E-1, E-2, E-6, and
E-7 (fragility specifications B1041.001a, B1041.002a, and B1041.003a
have beams on one side). Beams occur on two sides of the columns at
grids A-3, A-4, A-5, B-1, B-7, C-1, C-7, D-1, D-7, E-3, E-4, and E-5
(fragility specifications B1041.001b, B1041.002b, and B1041.003b have
beams on both sides).
• Beam sizes for all framing bays are 34 inches wide by 30 inches deep.
As noted in Section 2.4.2, special moment frame fragility specifications
with varying beams sizes are provided (B1041.001 for 24×24,
B1041.002 for 24×36, and B1041.003 for 36×36).
Based on the criteria listed above, a reasonable selection of fragility

specifications to estimate the consequences for four 34×30 inch beams in the
example building is illustrated in Figure 4-6 and listed as the following:
• B1041.002a for the one-sided beam/column joints with a 24×36 inch size
beam
• B1041.003b for the beam/column joints with 36×36 inch beams framing
into the column from each side

Figure 4-6 Illustration of reinforced concrete element fragility specification
selections.
Although users can modify the provided fragility specification to reflect the
actual 34×30 inch beam size, this level of precision is typically not justified.
Table 4-1 summarizes the fragility group selections.
Table 4-1 Fragility Group Selections for the Beam/Column Components

Fragility Fragility
Joint Classification Input Joint Classification Input
Location Number Direction Location Number Direction
A-1 B1041.002a 1 A-5 B1041.003b 2
A-2 B1041.002a 2 B-1 B1041.003b 1
A-6 B1041.002a 2 B-7 B1041.003b 1
A-7 B1041.002a 1 C-1 B1041.003b 1
E-1 B1041.002a 1 C-7 B1041.003b 1
E-2 B1041.002a 2 D-1 B1041.003b 1
E-6 B1041.002a 2 D-7 B1041.003b 1
E-7 B1041.002a 1 E-3 B1041.003b 2
A-3 B1041.003b 2 E-4 B1041.003b 2
A-4 B1041.003b 2 E-5 B1041.003b 2
Other vulnerable reinforced concrete elements include the post-tensioned

slab-column joints. Post-tensioned reinforced concrete slab systems are
listed as fragility groups B1049.021a through B1049.032. In the example
building, slab shear reinforcement has been provided at columns and the
gravity shear to shear capacity ratio is less than 0.4. Thus, B1049.031 is the
most appropriate selection. This fragility specification is defined with a non-
directional demand parameter since the joints are subject to damage when
story drifts occur in any direction. As shown in Figure 4-5, the non-
directional fragility is signified by a solid block in the columns for direction
1 and direction 2, compared to a check mark symbol for a directional
fragility.
Figure 4-5 illustrates selection of the above fragility groups in the upper half
of the Component Fragilities tab. The lower half of Figure 4-5 illustrates

the fragility groups selected for the first floor. The user must edit this half of
the window to indicate the fragility specifications that apply at each floor and
for each direction. If components identified in the upper half of the page do
not occur on all floors or in both directions, they should be deleted from the
floors where they do not occur, on the bottom half of the page.
4.4.4.2 Structural Component Performance Groups
Quantities are assigned and distributed among the building floors using the
Performance Groups tab. This tab will be imprinted with the previously
selected structural fragility specifications on the Component Fragilities tab.
For each floor, the number of special moment frame beam-column joints
vulnerable to story drift in each building direction are entered for each of the
pre-selected specifications. Table 4-2 reorganizes Table 4-1 for use in
defining performance groups in PACT:
Table 4-2 Performance Group Quantities for Reinforced

Concrete Elements
Fragility Classification Number Direction Number Per Floor
B1041.002a 1 4
B1041.002a 2 4
B1041.003b 1 6
B1041.003b 2 6
Note that for these components, no dispersion is identified with the quantity
since it is known precisely. Note also that for this example, components
within the performance group are designated as correlated. With this option
selected, PACT will execute realizations for the entire group and assign a
damage state distribution among the members of the group that reflect the
result of the realization. If the components are designated as uncorrelated,
PACT will determine unique damage states for each component within the
group for each realization. Since only a median estimate of loss is desired in
this example, the assumption of fragility correlation is appropriate. If
statistics on the possible range of loss are desired, or the probability of
obtaining an unsafe placard is needed, the component performance groups
should be designated as being uncorrelated.
Input of the post-tensioned slab/column joint information is similarly inserted

at each floor; however, these fragilities are input as non-directional. There
are 15 joints per floor.
The performance group definition process is repeated for each floor and for
each direction (including non-directional). Figure 4-7 through Figure 4-9

illustrate data input for direction 1, direction 2, and non-directional fragility
specifications on the first floor.
Figure 4-7 PACT entries for 1st floor structural performance groups,
direction 1.
Figure 4-8 PACT entries for 1st floor structural performance groups,
direction 2.
Figure 4-9 PACT entries for 1st floor structural performance groups, non-
directional.
4.4.5 Nonstructural Components
4.4.5.1 Nonstructural Component Fragility Specifications
The process of identifying and selecting the type and distribution of the
nonstructural components can be greatly simplified by the use of the
Normative Quantity Estimation Tool, provided in Volume 3. This tool can
be used to generate a floor-by-floor listing of nonstructural components with
estimates of their performance group quantities with the simple input of
building floor areas and occupancies. The tool is an aid for estimating
normative quantities and careful review by the user is necessary. Also note
that the tool only identifies typical quantities of components present in a
building without distinguishing their vulnerability to damage. For example,

the tool will identify the typical square footage of suspended ceilings in a
building without discrimination as to the extent of seismic bracing present.
Users must determine this aspect of performance group selection by
themselves.
To use the Normative Quantity Estimation Tool, the building floors and
occupancies are entered into the Building Definition Table in the
Normative Quantity Estimation tab. Figure 4-10 illustrates input of this
information for the example building.
Figure 4-10 Normative Quantity Estimation Tool, Building Definition Table.
After this information is entered, pressing the Compile button at the top of
the worksheet executes the tool and produces a list of nonstructural
components and their representative quantity at each floor for each
occupancy type. Figure 4-11 shows inventory generated for floor 3 of the
example building. Similar information is also generated for other levels.
The nonstructural components listed in Figure 4-11 represent the most

probable distribution found in typical buildings with commercial office
occupancy and having the same size as the building under evaluation. This
distribution is based on expert judgment. Every building will have somewhat
different types and quantities of components than those identified by the tool
and this will also vary over time. If desired, a building specific inventory can
be performed to capture a more accurate distribution of these nonstructural
components at the time the evaluation is performed.
A median estimate of the number of units is provided for each fragility

specification listed in Figure 4-11. The total number of units is identified for
both directional and non-directional specifications. The user must distribute
the total number of directional units between direction 1 and direction 2. It
should be noted that for distributed systems, such as piping or partitions,
units of measure for components are typically hundreds of lineal or square
feet. As a result, the quantity of units calculated will often be a decimal
fraction. If it is assumed that the fragility of all components within a

performance group is correlated, these decimal fractions can be used directly.
However, if uncorrelated behavior is assumed, the units should be entered as
integers.
Figure 4-11 Normative Quantity Estimation tool, Component Summary Matrix showing nonstructural
inventory.
Similar to the structural fragility specifications, the user must transfer this
information onto the Component Fragilities and Performance Groups tabs
in PACT. The user should review the fragility specifications and quantities
provided by the Normative Quantity Estimation Tool for their
appropriateness at each floor and modify quantities and add (or delete)
fragility selections as appropriate to best reflect the damageable nonstructural
components present at each floor. Items that occur only once in the building
are kept at the appropriate floor and deleted from other floors. For example,
roof tiles only occur at the roof level.
4.4.5.2 Nonstructural Component Performance Groups
After completion of the floor-by-floor input of fragility specification data,

nonstructural performance groups can be identified and input using the
Performance Groups tab. When the tab is initiated PACT will provide a
warning at the bottom of the window if fragility groups that require user-
defined information, such as anchorage capacity, are selected. Figure 4-12
illustrates PACT input of first floor data for non-directional performance
groups respectively.

The performance group quantities identified with the Normative Quantity
Estimation Tool are median estimates. For distributed systems (piping,
ducting, ceilings, etc.), the median estimates are accompanied by dispersions
that reflect the variability in component quantity for the selected occupancy.
For this example, the dispersion input to PACT has been set to 0 to reflect a
precise understanding of the amounts of components. In most instances, the
user should select the recommended dispersion values from the tool to reflect
the variability of the component, both as presently deployed and as may be
modified over time. If the user is confident in the quantity of the component,
the dispersion can be set to 0. For discrete items such as boilers and cooling
towers, a dispersion value of 0 will generally be a more appropriate selection
since the quantity of these components varies by whole integers and the
Normative Quantity Estimation Tool will round off its median estimate to the
nearest whole integer.
Figure 4-12 PACT Performance Groups tab for floor 1, non-directional.
The following information summarizes the nonstructural components

included in the performance assessment model based on the results obtained
from the Normative Quantity Estimation Tool. The normative and actual
quantities are not rounded for illustration purposes.
B20: Exterior Enclosure
B201: Exterior Nonstructural Walls. The building’s exterior skin is a glazed

curtain wall consisting of glazing (category B202) and aluminum panels.
The aluminum panels are considered rugged and therefore no fragility group
is selected.

B202: Exterior Windows. The building includes 5 foot by 6 foot panels (6:5
aspect ratio) of insulated single pane annealed insulated glazing units with no
additional information. According to guidance provided in Section 2.5.2 and
Table B-2, these components are identified as B2022.001. The Normative
Quantity Estimation Tool indicates that 216 units of 30 square foot of glazing
will be present on each floor of this building in total, in both directions.
Since each panel is 5 foot by 6 foot in size, this corresponds to 216 panels.
Based on the building dimensions given in Figure 4-1, the glazing quantities
can be distributed in a 2 to 3 ratio between direction 1 and direction 2.
B203: Exterior Doors. Exterior doors are considered to be rugged items for
the subject structure. No fragility group is selected.
B30: Roofing
B301: Roof Coverings. The subject structure has a system of concrete roof
tiles attached to the mansard roof in compliance with the 2007 California
Building Code. For this assessment, the attachment is deemed similar to the
requirements of the 1994 UBC, Uniform Building Code, (ICC, 1994).
Accordingly, specification B3011.011 best defines the roof system
damageability.
C10: Interior Construction
C101: Partitions. Interior partitions consist of gypsum wallboard over steel

studs that are fixed at top and bottom. Fragility specification C1011.001a is
selected to best characterize the performance of the building partitions. Wall
finishes are identified under category C301 below.
C20: Stairs
The specification that describes prefabricated steel stairs with seismic joints,
C1011.001, is selected. The Normative Quantity Estimation Tool indicates a
quantity of 3 stairs within the building. Each stair is oriented in direction 1.
C30: Interior Finishes
C301: Wall Finishes. Ceramic tile finish has been included (C3011.002c) at
each floor of the building. Finish on partition walls are measured in units of
100 lineal feet, with an estimated height of 9 feet. The Normative Quantity
Estimation Tool indicates a quantity of 1.63 units (163 lineal feet) in total for
directions 1 and 2.
C303: Ceilings. The structure has suspended acoustical ceilings,

C3032.003b. The Normative Quantity Estimation Tool indicates that 32.40

units of suspended ceilings, measured in units of 600 square feet, will be
present. This fragility is identified as non-directional.
Specification C3034.002 is selected for suspended pendulum lighting with

independent support wires. The Normative Quantity Estimation Tool
estimates that 324 units are present per floor for this non-directional
specification.
D10: Conveying
D101: Elevators and Lifts. The structure includes one traction elevator,
D1014.011. This non-directional component is located only at floor 1.
D20: Plumbing
Table 4-3 summarizes the selection of plumbing components and quantities

indicated by the Normative Quantity Estimation Tool.
Table 4-3 Performance Group Quantities for Plumbing Components

Quantity Unit of Actual
General Description Fragility ID per Floor Measure Quantity
Domestic Cold Water
Piping – (greater than D2021.013a 0.32 1000 LF 324 feet
2 ½ Φ)
Domestic Hot Water
Piping – (less than 2 ½ D2022.013a 1.81 1000 LF 1814.4 feet
inch Φ)
Domestic Hot Water
Piping – (greater than 2 D2022.023a 0.65 1000 LF 648 feet
½ inch Φ)
Sanitary Waste Piping,
Cast Iron with Flexible D2031.013b 1.23 1000 LF 1231.2 feet
Couplings
D30: HVAC
D304: HVAC Distribution Systems. This category includes fans, drops,

diffusers, and variable air volume (VAV) boxes. Table 4-4 summarizes the
selection of HVAC distribution components, system fragility specifications
and performance groups indicated using the Normative Quantity Estimation
Tool.
D306: Packaged Air Handling Units. The Normative Quantity Estimation

Tool recommends 16 units of D3052.013b with a capacity of 4,000 CF each.
These units are non-directional and located in the first floor.

Table 4-4 Performance Group Quantities for Distributed HVAC Components
Quantity per Unit of
General Description Fragility ID Floor Measure Actual Quantity
HVAC stainless ducting < 6 ft.2 D3041.021c 0.43 1000 LF 432 feet
HVAC drops/diffusers without 19.44 10 each 194.4

D3041.032c
ceilings
Variable Air Volume boxes D3041.041b 10.8 10 each 108
D40: Fire Protection
D401: Fire Sprinkler. Fragility specification D4011.033a was selected to

represent suppression sprinklers dropping into unbraced soft lay-in tile
ceilings. It is further assumed that since the building is of recent
construction, the ceiling bracing conforms to Seismic Design Category D, E,
or F to appropriately reflect the seismic hazard of the Los Angeles site. The
fragility specification for D4011.033a identifies the units of measure for fire
sprinkler drops as 100 each. The Normative Quantity Estimation Tool
indicates 1.94 units, thus 194 drops at each floor.
D50: Electrical
The fragility specification D5012.023d was selected for low voltage

switchgear with a unit of measure of 400 amps. The Normative Quantity
Estimation Tool indicates one unit per floor is present.
Equipment and furnishings and fragilities for special construction were not
considered.
4.4.6 Collapse Fragility and Collapse Modes
4.4.6.1 Collapse Fragility
Volume 1, Chapter 6 discusses three approaches for development of collapse

fragility. This example demonstrates the use of the nonlinear static approach
as summarized in Section 2.6.2.
Step 1. Construct a mathematical model of the building for nonlinear static

analysis in each building direction. For the example structure, each building
direction has identical strength and stiffness characteristics so only one
analysis is needed.
Step 2. Nonlinear static analysis was performed to identify the sequence of

component yielding to a deformation at which collapse is judged to occur.
Step 3. Figure 4-13 illustrates the results of the pushover analysis for both
building direction 1 and direction 2, which are identical. Collapse is judged

to occur at a roof drift of 2.7 feet, which corresponds to a roof drift ratio of
5%.
Figure 4-13 Pushover curve developed by analysis.
Step 4. The coordinates of the pushover curve in Figure 4-15 are input into
the SPO2IDA Tool provided in Volume 3 along with the building height (54
feet), building weight (8365.6 kips) and fundamental building period
(1.13seconds). Four control points are used to approximate the pushover
curve as illustrated in Figure 4-14.
Figure 4-14 SPO2IDA Tool, SPO tab.
Step 5. Figure 4-15 presents the results of the SPO2IDA evaluation. The
value of Sâ (T ) is estimated as 2.23g. The collapse fragility is thus defined
as having a median value Sâ (T ) of 2.23g and a dispersion of 0.6.

Figure 4-15 SPO2IDA Tool, IDA Results tab.
Step 6. The median collapse capacity and a dispersion of 0.6 are input into
PACT using the Collapse Fragility tab, as shown in Figure 4-16.
Figure 4-16 PACT Collapse Fragility tab.

4.4.6.2 Collapse Modes
PACT collapse mode input includes the number of independent collapse

modes that can occur and the probability of each. This is difficult to predict
analytically. To determine these data, the user must use judgment based on
upon building type, structural system, past experience, and analytical
inferences. When using the simplified analysis approach, limited analytical
information regarding potential collapse modes is available. For this
example, only a single mode of collapse is considered. Additional
information gained from numerous response history analyses can provide
additional insight into potential collapse modes.
PACT input also includes the fraction of each floor subject to collapse-
associated debris and the resulting fatality and injury rates. These must also
be determined judgmentally. Figure 4-16 shows the input values for this
example.
4.4.7 Residual Drift Fragility
The PACT default residual drift fragility function is modified for this
assessment. The median residual drift ratio at which damage will be deemed
irreparable is increased to 0.0125. The estimated median building residual
drift ratio demand and dispersion are input into the PACT Residual Drift tab
after the analysis is completed.
4.5 Select Analysis Method and Construct Building

Analytical Model
Simplified analysis is used to generate median estimates of peak transient

drift, peak floor accelerations and residual drifts. Associated dispersions are
generated using simplified analysis. Peak total floor velocities are not
generated since none of the vulnerable building components use this demand
parameter. A linear building model is constructed using the modeling
criteria of ASCE/SEI 41-13 (ASCE, 2014), linear static procedure.
4.6 Define Earthquake Hazards
The target intensity is given as earthquake ground shaking with an average

return period of 475 years. In order to estimate median values of story drift
ratios, peak floor accelerations, and residual drift ratios, the following ground
motion parameters will be required: Sa( T1X ), Sa( T1Y ), and Sa(T = 0). These
parameters are obtained by using the steps outlined in Section 3.3.2.1 and
Section 3.5.2.1.

Step 1. In order to determine the building’s fundamental translational
periods in two orthogonal directions, modal analysis is performed. For this
doubly symmetric building T1X = T1Y = 1.13 seconds.
Step 2. The first mode average period is T = 1.13 seconds since T1X and T1Y
are identical.
Step 3. Obtain a seismic hazard curve for S a (T ) .
The following tools provided by the USGS that can be used to obtain a
seismic hazard curve at the site: (1) the Unified Hazard Tool
(https://earthquake.usgs.gov/hazards/interactive/, last accessed March 15,
2018); and (2) the Java Ground Motion Parameter Calculator. This
example uses the Java Calculator to construct the hazard curve. The basic
information necessary to generate the seismic hazard curve using this
approach includes the site latitude and longitude (accurate to three decimal
places), the average period, T , and the site class as defined in ASCE/SEI
7-10.
The example building is located in the Los Angeles area with: Latitude:
33.996°, Longitude: -118.162°, and Site Class D.
The Java Ground Motion Calculator provides spectral acceleration hazard

curves for sites conforming to ASCE/SEI 7-10 Site Class B at the following
building periods: 0, 0.1, 0.2, 0.3, 0.5, 1.0, and 2.0 seconds at the building
location. For intermediate periods, the value of Sa can be determined by
constructing the spectrum using the above periods and graphically
determining the value of Sa(T). For structures in the period range of 0.7 to
2.0 seconds, it is generally considered sufficiently accurate to obtain Sa(T)
from the formula:
S a (1.0 )
S a (T ) =
T
Figure 4-17 illustrates the Java Ground Motion Calculator input and tabular
output for spectral acceleration values at 1.0 second period and their mean
annual frequency of exceedance at the building site, assuming Site Class B
conditions.

Figure 4-17 USGS hazard data for 1.0 second period at Site Class B.
Table 4.5 shows the spectral acceleration values (at 1.0s) and mean annual frequency of exceedance, λ, values
obtained from the ground motion calculator, and values of Sa(1.13s) derived using the above formula with T =
1.13s. The table also provides values of Fv obtained from ASCE/SEI 7-10 at each λ and the corresponding
Sa(1.13s) values at Site Class D, using Fv. Figure 4-18 is a plot of these data evaluated at the building
average period at Site Class D and presented in log-linear form.

Table 4-5 USGS Hazard Data, Adjusted for Building Period
and Site Class
Sa(T=1.00s) Sa(T=1.13s) Sa(T = 1.13s)
Site Class B λ Site Class B Fv Site Class D
0.002 0.75855 0.0018 2.4 0.00432
0.004 0.64518 0.0035 2.4 0.0084
0.006 0.52414 0.0052 2.4 0.01272
0.008 0.40407 0.0071 2.4 0.01704
0.013 0.30233 0.012 2.4 0.0288
0.019 0.22048 0.017 2.4 0.0408
0.029 0.15411 0.026 2.4 0.0624
0.043 0.10169 0.038 2.4 0.0912
0.064 0.06226 0.057 2.4 0.1368
0.096 0.034926 0.084 2.4 0.204
0.144 0.017497 0.127 2.2 0.2921
0.216 0.007576 0.191 2.0 0.382
0.324 0.002775 0.287 1.8 0.5166
0.487 0.000849 0.431 1.5 0.6896
0.730 0.00022 0.646 1.5 0.969
1.090 0.000048 0.965 1.5 1.4475
1.640 0.000008 1.451 1.5 2.1765
2.460 0.000001 2.177 1.5 3.2654
Figure 4-18 Seismic hazard curve for Sa(T = 1.13s), Site Class D.

The mean annual frequency of exceedance for ground shaking intensity with a mean return period of 475
years is 1/475 = 0.002105. From Table 4-5, the Sa (T = 1.13s) value at Site Class B corresponding to this λ, is
0.381g. Adjusting this value for Site Class B using the site coefficient, Fv (value of 1.7 for S1 between 0.3
and 0.4 in from ASCE/SEI 7-10 Table 11.4-2), yields Sa (1.13s) = 1.7 × 0.381g = 0.648g.
The Java Ground Motion Calculator can also be used to generate the peak ground acceleration at the site for a
return period of 475 years, as shown in Figure 4-19. The value of the peak ground acceleration is identified
as 0.419g.
Figure 4-19 USGS hazard data for peak ground acceleration.
Using ASCE/SEI 7-10 Table 11.4-1 which provides a conversion factor, Fa, for ranges of Ss (the Sa value at
0.2 seconds), an appropriate value is selected to adjust the peak ground acceleation value to reflect site class.
Peak ground acceleration can be related to Ss for this purpose by factoring peak ground acceleration by 2.5,
yielding a value of 1.1g.

Returning to ASCE/SEI 7-10 Table 11.4-1 with Ss = 1.1, a value of Fa = 1.1
is selected.
The resulting site adjusted peak ground acceleration is: 1.1 × 0.419g =
0.461g.
4.7 Analyze Building Response
Following the steps outlined in Section 3.3.2.2, median estimates and

dispersions of peak transient drift ratio, peak floor acceleration, and residual
drift ratio are generated using the simplified analysis approach.
4.7.1 Estimate Median Story Drift Ratio and Dispersion
Step 1. Determine pseudo lateral force.
V = C1C2Sa(T1)W1
T1 = 1.13 seconds; Sa (1.13s) = 0.648g; W = 8365.6 kips

for T 1 ≥ 1.0 seconds , C 1 = 1.0
C2 =
Step 2. Distribute forces vertically, as shown in Table 4-6.
Table 4-6 Lumped Weight Distribution
wh
Story W h (in.) ∑ wh
4 2091.4 648 0.391
3 2091.4 492 0.297
2 2091.4 336 0.203
1 2091.4 180 0.109
Step 3: Calculate displacements and drift ratios.

Using a linear modeling representation of the building consistent with
ASCE/SEI 41-13, the following displacements are calculated, see Figure
4-20:

Figure 4-20 Results from linear analysis.
The corresponding drift ratios are:

3.334
Δ2 = = 0.01852
180
6.270 − 3.334
Δ3 = = 0.01882
156
8.780 − 6.270
Δ4 = = 0.0161
156
10.254 − 8.780
Δr = = 0.00945
156
Step 3. Obtain estimates of median story drift ratio, Δi*, at each story by
correcting for inelastic behavior and higher modes using Volume 1, Equation
5-10:
=∆i* H ∆i ( S , T1 , hi , H ) × ∆i i = 1 to N
2 3
hi +1  hi +1   hi +1 
ln ( H ∆i ) =a0 + a1T1 + a2 S + a3 + a4   + a5  H  , S ≥1, i =1 to N
H  H   
with T1 = 1.13s, H = 648 inches
Values for coefficients a0 through a5 for buildings 9 stories or less in height

are provided in Volume 1, Table 5-4. For a 4-story moment frame structure,
the coefficients are as follows:

a0 = 0.75, a1 = -0.044, a2 = -0.010, a3 = -2.58, a4 = 2.30, a5 = 0.0
S a (T1 ) W
S=
Vy1
S a (T1 ) 0.648g,
= = W 8365.6kips
The value of Vy1 is taken from the pushover analysis used to estimate the
collapse fragility (see SPO2IDA input, Figure 4-14, Elastic Segment
endpoint) as 1066 kips.
0.648 ( 8365.6 )
=S = 5.09
1066
Table 4-7 summarizes the calculation for converting calculated peak story
drift ratio into a median estimate of peak story drift ratio.
Table 4-7 Median Story Drift Ratio Estimates

hi + 1
Story Δi H ln HΔi HΔi Δi*
1 0.01852 0.28 0.1073 1.113 0.0206
2 0.01882 0.52 -0.0707 0.9317 0.0175
3 0.01610 0.76 0.01708 1.0172 0.0164
4 0.00945 1.00 0.3694 1.4469 0.0137
Step 4. Assign dispersions.
From Table 3-4 with T1 = 1.13s and S = 5.09 βSD is obtained by interpolation
to be 0.59.
4.7.2 Estimate Median Peak Floor Acceleration and Dispersion
Step 1. Generate median peak floor accelerations.

S = 5.09, T1 = 1.13s, H = 648 inches, PGA = 0.461g

2 3
hi h  h 
H H H
where the values for coefficients a0 through a5 for a moment frame structure
are given in Volume 1, Table 5-4 are as follows:

a0 = 0.66, a1 = -0.25, a2 = -0.080, a3 = -0.039, a4 = 0.0, a5 = 0.0
Table 4-8 summarizes the results of application of the above expressions
to generate median estimates of peak floor acceleration.
Table 4-8 Median Floor Acceleration Estimates
hi
Floor PGA H ln HAi HAi a i*
1 0.461 - - - 0.461
2 0.461 0.28 -0.04062 0.9602 0.443
3 0.461 0.52 -0.05000 0.9512 0.439
4 0.461 0.76 -0.05934 0.9424 0.434
Roof 0.461 1.00 -0.06870 0.9336 0.430
Step 2. Assign dispersions.

From Table 3-5 with T1 = 1.13s and S = 5.09 βFA is obtained by
interpolation to be 0.48.
4.7.3 Estimate Median Residual Story Drift Ratio and Dispersion
Step 1. Generate median residual drift ratio.
Residual drift ratio is determined by the difference between peak transient

drift ratio and median story drift ratio at yield. The peak transient drift ratios
were estimated in Section 4.7.1. The yield drift ratio is obtained from the
pushover analysis used to generate the story shear at yield. From Figure
4-15, the roof displacement at yield is 0.189 feet or 2.268 inches. Thus, the
yield drift ratio is
2.268
∆y
= = 0.00350
648
The maximum transient drift ratio for the building occurs at the first story
(0.0206). The maximum residual drift ratio is therefore:
∆r* ( max ) =
0.0206 − 3 ( 0.0035 ) =
0.0101
Note that in this example, ∆i* is substantially greater than 4 Δy. This
significantly exceeds the recommended limitation on the degree of inelastic
behavior for which the linear approach is appropriate. At higher levels of
nonlinearity, the effects of component degradation and P-Δ may result in an
underestimation of median demand.
The median estimate of residual drift obtained from the simplified analysis
method is assigned a dispersion of 0.8.

4.8 Input Response Data and Calculate Performance
The median demand estimates for peak transient drift ratio, peak floor
acceleration, and residual drift ratio are input to PACT for direction 1 and
direction 2 in the Structural Analysis Results tab. Figure 4-21 illustrates
the input of peak transient drift ratio values. Residual drift ratio is input
using the Residual Drift tab. Figure 4-22 illustrates input of the peak
residual drift ratio and its associated dispersion on the lowest row of the tab.
Figure 4-21 PACT peak transient drift ratio input on Structural

Analysis Results tab.

The assessment can be executed by saving and exiting the model. After
returning to the PACT Control Panel, pressing the Evaluate Performance
button and identifying the assessment model will run the tool.
4.9 Review Results
Figure 4-23 provides illustrative results from PACT analysis. At the bottom
of Figure 4-23, the estimated median repair cost is shown as $5,126,667,
which corresponds to 23.7% of the example building’s total replacement
cost. The top half of the figure can be used to identify the contribution of
different fragility groups to the repair cost total. If one type of component is
identified as having a large contribution, e.g., partitions fixed at top and
bottom, replacement with an alternate configuration of the component, e.g.,
slip track partitions, could reduce the report cost total. After such
consideration the performance assessment model can be revised and rerun
with PACT.

Figure 4-23 PACT Repair Cost tab.
Performance results from PACT can be examined in numerous ways.

Viewing results by realization reveals that residual drift plays a more
significant role than collapse (see Figure 4-24). For approximately 186 of
the 500 realizations, residual drift resulted in the damage being judged
irreparable.

Figure 4-24 PACT Repair Cost tab with realizations.

Chapter 5
Example Application:
Time-Based Performance
Assessment Using Nonlinear
Response History Analysis
5.1 Introduction
This chapter provides an example application of the time-based performance

assessment methodology using nonlinear response history analysis in the
Performance Assessment Calculation Tool (PACT) with provided fragility
and consequence function data.
The work necessary to obtain a probable maximum loss value that reflects
the repair cost, expressed as percentage of building replacement cost, having
a 10% probability of exceedance over a 30-year loan period for an office
building is presented here as follows:
• Obtain site and building description (Section 5.2)
• Select assessment type and performance measure (Section 5.3)
• Assemble building performance model (Section 5.4)
• Select analysis method and construct analytical building model (Section
5.5)
• Define earthquake hazards (Section 5.6)
• Analyze building response (Section 5.7)
• Input response data and calculate performance (Section 5.8)
• Review results for selected performance measures (Section 5.9)
5.2 Obtain Site and Building Description
The example uses the same building and performance assessment model as
Chapter 4 with some modifications. Section 4.2, provides a detailed building
description. In summary, the building is a 4-story reinforced concrete
structure with commercial office occupancy located in Los Angeles,
California.
FEMA P-58-2 5: Example Application: Time-Based Performance Assessment 5-1

Second Edition Using Nonlinear Response History Analysis
5.3 Select Assessment Type and Performance Measures
Time-based assessment will be performed with building performance

expressed in terms of average annual repair cost in dollars. The resulting
cost distribution is used to determine the loss associated with a 10%
probability of exceedance over a 30-year period of time and converted from
dollars to percentage of replacement cost.
5.4 Assemble Performance Assessment Model
Section 4.4 describes performance assessment model assembly in detail.

However, the following additions and changes to the assessment model are
made for this assessment:
• Performance groups are identified with uncorrelated components to
obtain an improved estimate of the performance variability.
• The collapse modes identified and input to the assessment model are
expanded to include partial collapse. As a result of information obtained
from response history analyses used to generate analysis input, a partial
collapse mode is added to the previously defined total collapse mode.
The partial collapse mode consists of sidesway failure of the top floor.
This mode does not lead to collapse of the lower floors. Collapse
mode 1 from the previously developed assessment model is revised to
reflect an occurrence probability (given that collapse occurs) of 0.5.
Collapse mode 2 is similarly assigned a probability of 0.5. Table 5-1
summarizes the salient collapse characteristics.
• Structural analysis results are input for 8 ground shaking intensities and 7
ground motions for each intensity level. Each demand vector, associated
with one ground motion includes input for peak story drift and floor
acceleration at each level in each direction and maximum residual story
drift.
Table 5-1 Collapse Mode Consequences

Collapse Debris Fatality Rate with Injury Rate with
Floor Fraction COV = 0.5 COV = 0.5
Mode 4 3 2 1 4 3 2 1 4 3 2 1
1 1 1 1 1 1 1 1 1 0 0 0 0
2 1 0 0 0 1 0 0 0 0 0 0 0
5-2 5: Example Application: Time-Based Performance Assessment FEMA P-58-2

Using Nonlinear Response History Analysis Second Edition
5.5 Select Analysis Method and Construct Building
Analytical Model
The structure is analyzed using nonlinear response history analysis. A three

dimensional analytical model is assembled as described in Section 3.2.2 and
Volume 1, Chapter 5.
5.6 Define Earthquake Hazards
Time-based assessment requires execution of the procedure presented in

Section 3.5.1. The nine steps described in that section are illustrated below
for the example building.
Step 1 through Step 3 were illustrated in Section 4.6 and the results are
repeated here.
Step 1. The building’s fundamental translational periods in two orthogonal

directions, are T1X = T1Y = 1.13 seconds.
Step 2. The first mode average period is T = 1.13 seconds since T1X and T1Y
are identical.
Step 3. Obtain a seismic hazard curve for S a (T ) .
Although site-specific seismic hazard analysis can be used, for this example,
the USGS Java Ground Motion Calculator is used, as previously illustrated
in Figure 4-17.
For intermediate periods, the value of Sa can be determined by constructing

the spectrum using the above periods and graphically determining the value
of Sa(T). For structures in the period range of 0.7 to 2.0 seconds, it is
generally considered sufficiently accurate to obtain Sa(T) from the formula:
S a (1.0 )
S a (T ) =
T
Table 5-2 shows the spectral acceleration values (at 1.0s) and mean annual
frequency of exceedance, λ, values obtained from the ground motion
calculator, and values of Sa(1.13s) derived using the above formula with T =
1.13s. The table also provides values of Fv obtained from ASCE/SEI 7-10 at
each λ and the corresponding Sa(1.13s) values at Site Class D, using Fv.
Figure 5-1 is a plot of these data evaluated at the building average period at
Site Class D and presented in log-linear form.

Table 5-2 Hazard Data, Adjusted for Building Period and Site
Class
Sa(T=1.00s) Sa(T=1.13s) Sa(T = 1.13s)
Site Class B λ Site Class B Fv Site Class D
0.002 0.75855 0.0018 2.4 0.00432
0.004 0.64518 0.0035 2.4 0.0084
0.006 0.52414 0.0052 2.4 0.01272
0.008 0.40407 0.0071 2.4 0.01704
0.013 0.30233 0.012 2.4 0.0288
0.019 0.22048 0.017 2.4 0.0408
0.029 0.15411 0.026 2.4 0.0624
0.043 0.10169 0.038 2.4 0.0912
0.064 0.06226 0.057 2.4 0.1368
0.096 0.034926 0.084 2.4 0.204
0.144 0.017497 0.127 2.2 0.2921
0.216 0.007576 0.191 2.0 0.382
0.324 0.002775 0.287 1.8 0.5166
0.487 0.000849 0.431 1.5 0.6896
0.730 0.00022 0.646 1.5 0.969
1.090 0.000048 0.965 1.5 1.4475
1.640 0.000008 1.451 1.5 2.1765
2.460 0.000001 2.177 1.5 3.2654
Figure 5-1 Seismic hazard curve for Sa(T = 1.13s), Site Class D.

Step 4. Define Samin (T ) and Samax (T ) that span between building response
that produces negligible damage to response that produces significant
probability of collapse. For T =1.13s, S amin (T ) can be taken as:
0.05 0.05
(T )
S amin = = = 0.044g
T 1.13
S amax (T ) need not be taken larger than the median collapse capacity or
alternatively as 2 × S a (T ) with a λ of 0.0004.
From Section 4.4.5.1 the structure’s median collapse capacity Sâ (T ) = 2.23g .
From Figure 5-1 the Sa (T ) value is interpolated as 0.935g. Twice this value
is 1.87g, which is taken as S amax (T ) .
Step 5. Divide the seismic hazard curve into 8 segments that span the range
of Sˆ (T ) from S min (T ) to S max (T ) .
a a a
Figure 5-2 illustrates the striping of the seismic hazard curve of Figure 5-2
producing 8 equal Sa-width segments spanning the range of Sâ (T ) from
( )
0.044g to 1.87g Samin (T ) to Samax (T ) . Each segment has a width
1.87 − 0.044
= 0.228g
8
Figure 5-2 Segmented seismic hazard curve to be used for time-based

assessment.
Step 6. The central value of Sâ (T ) at the midpoint of each segment is

determined together with its mean annual frequency of exceedance. Table
5-3 summarizes these data.

Table 5-3 Intensity Segment Values
Spectral Sa
Segment λ
1 0.158 0.05364
2 0.387 0.00740
3 0.615 0.00170
4 0.843 0.00050
5 1.071 0.00018
6 1.299 0.000101
7 1.528 0.0000435
8 1.756 0.0000311
For Sa (T )min = 0.044, the corresponding λ value is 0.2106. For Sa (T )max = 1.87g,
the corresponding λ value is 0.00002495.
Step 7. For each spectral segment, a response spectrum is constructed that is used
for the ground motion scaling of Step 8. Uniform Hazard Spectra (UHS),
Conditional Mean Spectra (CMS) or Conditional Spectra (CS) can be used for this
purpose. For events with a relatively large frequency both the UHS and CMS should
have similar shapes and are appropriate for use. For small frequency (infrequent)
events, use of the CMS or CS spectra should improve estimates of drift and
accelerations over those generated using UHS spectra. For this example, UHS
spectra will be used for segments 1 and 2, which have mean annual return periods of
11.2 and 152.1 years respectively. For segments 3 through 8, CMS spectra will be
used for selecting and scaling ground motions.
Construction of a UHS for segments 1 and 2 can be accomplished using the Java
Ground Motion Parameter Calculator with the Probabilistic Uniform Hazard
Response Spectra analysis option, as illustrated in Figure 5-3. The spectral shape can
be considered constant for λ segments up to 0.002105. For less frequent hazard
intensities, see ASCE/SEI 41-06, Section 1.6 for UHS spectral adjustments.

Figure 5-3 Uniform Hazard Spectrum.
Table 5-4 summarizes deaggregation results for segments 3 through 8. Figure 5-4 illustrates a sample
deaggregation for segment 3. This tool can also be used to directly generate conditional mean spectra
from the deaggregation information.
Table 5-4 Deaggregation Summary
Segment Sa λ M R ε
3 0.615 0.0016 6.9 18 0.9
4 0.843 0.00064 6.9 18 1.5
5 1.071 0.00032 6.9 13 1.6
6 1.299 0.00018 6.8 10 1.7
7 1.528 0.00010 6.8 8 1.7
8 1.756 0.00004 6.8 7 1.8
These data can be used to construct a conditional mean spectrum for each intensity segment. Figure 5-6
illustrates the conditional mean spectrum generated for segment 3 by the USGS calculator.

Figure 5-4 Deaggregation of segment 3.
Figure 5-5 Conditional mean spectrum for segment 3.

Step 8. Select and scale the geomean of ground motion record pairs to the
amplitude of the target spectrum at the Sa (T ) as illustrated in Section
3.3.1.1 for intensity-based assessment. Ground motions for selection and
scaling can be found at https://ngawest2.berkeley.edu/, www.cosmos-eq.org
(last accessed March 15, 2018) and other websites. The PEER website also
provides software tools for spectral shape matching and scaling of ground
motion records, and a tool that can be used to select and scale record pairs to
any user defined spectrum.
An alternate site with tools to assist in selection and scaling include can be
found at http://www.stanford.edu/~bakerjw/. This site provides a tool for
selecting and scaling record pairs from the PEER database to match a
conditional mean target spectrum. A minimum of 7 well matched pairs of
records is recommended. For further discussion of the number of records
needed to produce statistically valid analysis results, see Volume 1.
Step 9. Repeat Step 8 for each spectral segment.
5.7 Analyze Building Response
Seven ground motions were selected and scaled for each of the eight
intensities. Table 5-5 summarizes structural response data obtained from
response history analysis at each of the eight intensities. Note that the
selected records for this example came from NGA West database that has
now been superseded by the NGA-West2 database.
As noted in Section 3.3.1.2 and Volume 1, Chapter 5, the analytically

generated median drift and acceleration estimates are assigned dispersions to
reflect response variability. The assigned dispersion, βm, is composed of two
components, βc and βq. Values for each are provided in Volume 1, Table 5-1
and Volume 1, Table 5-2 and are reproduced in Volume 2 as Table 3-1 and
Table 3-2.

Table 5-5 Summary of Structural Analysis Results
Intensity 1 Sa ( T ) =0.158 g Intensity 2 Sa ( T ) =0.387 g
Demand
Parameter Median Parameter Median
Δ4 0.002 Δ4 0.003
Δ3 0.004 Δ3 0.008
Δ2 0.005 Δ2 0.0134
Δ1 0.005 Δ1 0.010
a5 0.24 a5 0.34
a4 0.18 a4 0.30
a3 0.19 a3 0.30
a2 0.16 a2 0.30
a1 0.19 a1 0.22
Demand
Δ4 0.006 Δ4 0.008
Δ3 0.013 Δ3 0.019
Δ2 0.017 Δ2 0.024
Δ1 0.015 Δ1 0.022
a5 0.43 a5 0.48
a4 0.36 a4 0.41
a3 0.38 a3 0.48
a2 0.43 a2 0.53
a1 0.36 a1 0.48
Demand
Δ4 0.012 Δ4 0.011
Δ3 0.024 Δ3 0.025
Δ2 0.030 Δ2 0.039
Δ1 0.029 Δ1 0.037
a5 0.49 a5 0.55
a4 0.44 a4 0.45
a3 0.48 a3 0.52
a2 0.59 a2 0.69
a1 0.52 a1 0.68
Δi: Maximum drift ratio at story i
ai: Peak floor acceleration at floor i

Table 5-5 Summary of Structural Analysis Results (continued)
Demand Demand
Δ4 0.017 Δ4 0.019
Δ3 0.029 Δ3 0.041
Δ2 0.042 Δ2 0.069
Δ1 0.040 Δ1 0.067
a5 0.61 a5 0.700
a4 0.58 a4 0.800
a3 0.58 a3 0.610
a2 0.87 a2 1.057
a1 0.89 a1 1.000
Δi: Maximum drift ratio at story i
ai: Peak floor acceleration at floor i
For this example:
βc = 0.25 and βq = 0.40
( 0.25) + ( 0.40 ) =
2 2
βm = 0.47
Residual drift is determined from the peak transient drift associated with each
demand vector for a given intensity. The drift with the largest ratio of
transient drift to yield drift for direction 1 and direction 2 for each demand
vector (EQ1, EQ2, EQ3, etc.) is selected. The residual drift ratio is
calculated using the values of transient story drift, Δmax, and the yield story
drift, Δy, as follows:
Δr = 0 for Δmax < Δy
Δr = 0.3(Δmax – Δy) for Δy < Δmax < 4Δy
Δr = Δmax – 3(Δy) for Δmax ≥ 4Δy
The value of yield drift, Δy, was determined in Chapter 4 for the example
building as Δy = 0.00350.
Table 5-6 displays maximum and residual drift values from each analysis for
intensity 1. For this intensity, peak drifts do not exceed four times the yield
drift (4 × 0.0035 = 0.014) therefore Δr = 0.3(Δmax – Δy).

Table 5-6 Residual Drift, Intensity 1
Intensity 1: Sa = 0.158g
1 2 3 4 5 6 7
Δmax 0.007 0.005 0.005 0.005 0.006 0.005 0.005
Δr 0.00105 0.00045 0.00045 0.00045 0.00075 0.00045 0.00045
Table 5-7 illustrates the maximum story drifts and residual drifts calculated
for intensity 3. For this intensity, peak drifts exceed 4Δy and therefore
Δr = Δmax – 3(Δy)
Table 5-7 Residual Drift, Intensity 3

Intensity 3: Sa = 0.615g
1 2 3 4 5 6 7
Δmax 0.023 0.022 0.022 0.018 0.030 0.023 0.020
Δr 0.0125 0.015 0.015 0.0075 0.0195 0.0125 0.0095
Residual drift estimates are assigned a dispersion of 0.8.
The analysis results are used by PACT to indicate a statistical distribution of

response states given that the building does not collapse. Initially, 7 scaled
ground motion record pairs were selected, scaled, and analyzed for each
intensity level. Collapse did not occur during any of the analyses for
intensities 1, 2, 3, 4, or 5. However, for intensities 6, 7, and 8, collapse,
indicated by lack of analysis convergence, did occur for some of the
analyses.
For intensity 6, collapse occurred for one of the analyses and the results of
that analysis were discarded. As discussed in Section 3.3.1.2 and Volume 1,
Section 5.2.2, when the remaining number of ground motion pairs for which
collapse did not occur is less than 7 and the probability of collapse for the
( )
spectral segment does not exceed 0.5 Sa (T )i ≤ Sâ (T ) , the remaining sets
of analyses (n-q) may not produce a valid estimate of demands when collapse
does not occur. Since Sa (T ) = 1.299g is less than the median collapse
( )
estimate for the structure Sâ (T ) = 2.23g , an additional analysis was
performed with a new scaled ground motion pair to provide the
recommended suite of at least 7 analyses in which no collapse occurred.
For intensity 7, analyses did not converge for two of the 7 ground motion
pairs. Two additional ground motion pairs were selected, scaled, and used to
reanalyze the structure. One of these additional analyses also failed to
(
converge. Since the spectral intensity of the segment Sa (T ) = 1.528 is less )
than Sâ (T ) and fewer than 7 analyses with no collapse exist, an additional

ground motion pair was selected, scaled, and used to analyze the structure.
Analysis with this pair did not produce collapse.
For intensity 8, three of the 7 ground motion record pairs produced collapse.
As with intensities 6 and 7, the probability of collapse for this intensity
segment is less than 0.5, therefore additional analyses are needed to produce
a total of 7.
5.8 Input Response Data and Calculate Performance

Structural analysis results for each of the 8 intensity levels are input to PACT
on the Structural Analysis Results tab. Figure 5-6 illustrates the drift input
values for intensity 3, direction 1. Input includes spectral acceleration (Sa =
0.615g), dispersion (βm = 0.47) and includes input for all 7 demand vectors.
Peak floor accelerations are input similarly.
Figure 5-6 PACT Structural Analysis Results tab with drift input for
intensity 3.
Peak residual drift values are input via the Residual Drift tab. Figure 5-7
illustrates this input for intensity 3. The residual drift fragility information in
the upper portion of the screen remains the same as in the previous
assessment model developed in Chapter 4 (median value of 0.0125 and a
dispersion of 0.3).
The mean annual frequency of exceedance values corresponding to each of

the intensities is input via the Hazard Curve tab.

Figure 5-7 PACT Residual Drift tab input.
5.9 Review Results
PACT displays assessment results on the Time Based Results tab as shown
in Figure 5-8, which shows annual probability of exceedance for repair cost
of different amounts as well as average annual loss values. In the illustration,
the average annual probability of collapse is shown in the box at the upper
left hand corner of the figure as 0.000157. This is equal to a mean return
period for collapse of 6,370 years. Similarly, the box below this indicates an
average annual probability of 0.0051 for damage being so severe that an
unsafe placard is posted on the building, equal to a mean return period of
approximately 200 years.
To determine the repair cost that has a 10% probability of non-exceedance

over the 30 year loan period, it is necessary to calculate the corresponding
return period for such a loss. The mean return period, PR, can be calculated
as a function of the number of years in the period of interest, Y, and the
desired probability of exceedance, PEY, using the formula:
−Y
PR =
ln (1 − PEY )
For this example: Y = 30 years, PEY = 10% = 0.10. Substituting and solving
−30
=PR = 285 years
ln (1 − 0.1)

The 285-year return period for the desired loss equates to an annual
frequency of exceedance of 1/285 years or 0.0035 per year. For the case of
Figure 5-8, with this annual probability of non-exceedance, the associated
repair cost is found as $3.8M. This represents 17.6% of the building’s total
replacement cost.
Figure 5-8 PACT Time Based Results tab showing annualized repair cost.

Chapter 6
Structural Fragility Calculation
6.1 Introduction
Volume 1, Section 3.8 presents procedures for calculation of fragility

parameters for component damage states associated with limit states that can
be predicted by calculation. This chapter illustrates those procedures using a
sample tilt-up building comprised of a plywood diaphragm, wood-framed
roofing, and reinforced concrete walls. The examples illustrate application
of the procedures for ductility-limited and strength-limited damage states
(ductile and brittle). The examples provide insight into the process of
selecting damage states and demand parameters to characterize component
performance. The following structural components and systems are
addressed:
• Plywood diaphragms
• Tilt-up concrete walls
• In-plane wall to roof attachment
• Out-of-plane wall to roof attachments
By applying the procedures illustrated here, users can apply the methodology
to include structures that have components for which there are no provided
fragility specifications. Section 6.4 describes the procedure for entering
user-defined fragility specifications into Performance Assessment
Calculation Tool (PACT) provided in Volume 3.
Chapter 7 provides guidance on the development of nonstructural component

fragility functions by calculation. Chapter 8 provides guidance on the
development of consequence functions to complete the fragility specification.
6.2 Building Description
An example structure is used to illustrate the calculation of various

component fragility functions. The structure is of conventional tilt-up
concrete construction with a plywood roof diaphragm. Figure 6-1 is a roof
plan for the structure and Figure 6-2 a cross-sectional elevation. Table 6-1
defines basic structural data.
FEMA P-58-2 6: Structural Fragility Calculation 6-1

Second Edition
Figure 6-1 Roof plan.
Figure 6-2 Cross-sectional elevation.
Table 6-1 Basic Building Information

Building Components Details of Construction
Plywood diaphragm ½” structural I plywood
Blocked and chorded
Uniformly nailed
10d @ 6” panel edges
Tilt-up reinforced 6” thick
concrete walls #5 @ 12” EW ASTM A615 Grade 60 reinforcement
f’c = 4000psi (normal weight)
Chords: 2 #5 rebar
In-plane wall to roof 10d @ 6” plywood to ledger nailing
attachment 5/8” Φ anchor bolt @ 2’ o.c, with 4 ½” embedment
into concrete wall
Out-of-plane wall to 2 holdowns (12 GA) with 2 - 5/8” Φ thru bolts (purlin
roof attachment attachment) and 2 5/8” Φ anchor bolts (or threaded
rods) with 4 ½” embedment into concrete wall, at 8
feet on center
The building’s basic lateral force-resisting components include the horizontal

plywood diaphragm, which distributes seismically induced inertial loads to
the walls of the building parallel to the direction of seismic action. The
seismic inertial loads predominantly reflect the inertial mass of the perimeter
walls normal to the direction of seismic action. The diaphragm resists
seismically induced deformation by spanning horizontally to the vertical
shear resisting elements (tilt-up walls) at the building perimeter. The
6-2 6: Structural Fragility Calculation FEMA P-58-2

Second Edition
diaphragm exhibits fundamentally the same behavior modes in each direction
of seismic demand.
Reinforced concrete walls act as shear walls to provide restraint to the

diaphragm for seismic forces acting parallel to the walls. For seismic forces
acting perpendicular to the walls, the walls transmit inertial reactions to the
roof diaphragm and to the foundation. The attachments of the walls to the
roof transfer the diaphragm end forces along the length of the in-plane walls
and transfer the out-of-plane inertial wall loads into the horizontal diaphragm
at discrete 8-foot intervals where tension ties (2 holdown devices) are
located.
Damageable components of the building’s lateral force-resisting system

include:
• Plywood roof diaphragm
o Transverse
o Longitudinal
• Tilt-up reinforced concrete walls
o In-plane
o Out-of-plane
• Wall to roof attachments
o In-plane
o Out-of-plane
The walls and roof are attached to transfer in-plane and out-of-plane wall and
diaphragm demands. The in-plane attachment (Figure 6-3) consists of a
diaphragm ledger (3×8, DF#1, timber) and steel anchor bolts that are
embedded into the concrete wall (5/8 inch diameter at 2 feet on center). The
out-of-plane attachment consists of holdown hardware (see Figure 6-4 and
Figure 6-5) attached to diaphragm sub-framing and bolted to the concrete
wall in a fashion similar to that of the in-plane connection. The sub-framing
is attached to the plywood diaphragm via nailing that provides a mechanism
for engaging the diaphragm to resist inertial wall demands. The sub-framing
includes splice detailing to maintain transfer continuity over a distance
sufficient to engage an adequate number of nails between the sub-framing
and the plywood.

Second Edition
Figure 6-3 In-plane attachment.
Figure 6-4 Out-of-plane attachment.
Figure 6-5 Horizontal holdown (out-of-plane)

attachment.

Second Edition
6.3 Development of Structural Component and System
Fragilities by Calculation
Volume 1, Section 3.8 presents procedures for calculating component

fragility functions based on characteristics of the anticipated component limit
state. The following four limit state categories are identified:
• Ductility-limited damage states
• Strength-limited damage states
• Displacement-limited damage states
• Code-based capacity damage states
Each limit state assumes a mathematical relationship between a demand

parameter and the occurrence of a damage state. For each category, an
estimate of the median value of the demand, θ, at which the damage state is
likely to occur is analytically estimated. It is assumed that the probability of
incurring damage at different values of demand is lognormally distributed
with dispersion, β. The value of the dispersion is selected to reflect the
variability in the following: component material strength, βM; the predictive
equations used to derive the median demand estimate, βD; and the degree to
which the actual construction is likely to match the construction quality, βC,
implicit in the median demand calculation. These procedures are illustrated
with the generation of fragility functions for the components and systems of
the example building.
6.3.1 Plywood Roof Diaphragm
The plywood roof diaphragm is considered to include composition roofing

and flashing, plywood sheathing, nailing attachment to the sub-framing, and
chord and collector elements.
The diaphragm can sustain damage in the following ways:

• Nail pullout at plywood panel edges or buckling of plywood panels
• Chord or collector buckling
• Chord or collector splice failure in tension
In this example, there are no collectors because the walls and diaphragm are
continuously attached. The chords consist of reinforcing steel embedded in
the concrete perimeter walls and are unlikely to buckle or fail given the
relative vulnerability of the wood framing components. Therefore, potential
damage states associated with chords or collectors are not addressed in this
example.

Second Edition
Testing and observation of the performance of plywood diaphragm systems
(Tissel, 1967) suggests damage states of substance include nail pullout and
panel buckling (Figure 6-6 and Figure 6-7).
Figure 6-6 Plywood diaphragm

damage (Tissel, 1967).
Figure 6-7 Plywood diaphragm

damage (Tissel, 1967).
A fragility function for plywood diaphragm damage states consisting of nail

pullout and panel buckling damage mechanisms is generated by calculation
using the procedure provided in Volume 1, Section 3.8.
Anticipated horizontal diaphragm performance can be characterized using

the generalized force-deformation “backbone” relationship as defined in
ASCE/SEI 41-06, Seismic Rehabilitation of Existing Buildings, (ASCE,
2007) and as shown in Figure 6-8 below:

Second Edition
Q/Qy
b
a
1.0 C
B
A D E c
θ or ∆
Figure 6-8 Generalized force-deformation relationship
adapted from ASCE/SEI 41-06 (ATC, 2010).
Response modeling parameters a, b, and c (illustrated in Figure 6-8) are

determined, according to ASCE/SEI 41-06 Table 8-4, since the subject
diaphragm has a span-to-depth ratio of approximately 3 to 1 and is blocked
and chorded. ASCE/SEI 41-06 also provides response values that are judged
to reflect three levels of component performance: Immediate Occupancy,
Life Safety, and Collapse Prevention. ASCE/SEI 41-06 Table C1-4 provides
a qualitative description of the anticipated degree of damage associated with
each of these response levels. For Immediate Occupancy performance,
observable loosening, withdrawal of fasteners, and splitting of sheathing or
framing are not anticipated. For Life Safety performance, observable
withdrawal of fasteners is expected, as well as splitting of framing,
sheathing, and connections. For Collapse Prevention performance, large
permanent distortion of the diaphragm is expected, accompanied by
extensive splitting of framing, connections, and sheathing in addition to
partial nail withdrawal.
For the diaphragm system in this example, three damage states of

consequence are considered. In Damage State 1 (DS1), no damage to the
plywood, nailing, or framing is anticipated, but damage to the roof flashing at
the wall to diaphragm interface is expected. This damage will require
removal and reinstallation of the flashing. Additionally, tearing of roofing
materials at this interface can be expected.
In the backbone curve:

Δ = Total elastic and plastic deformation
Δy = Displacement at effective yield strength
The ratio of total displacement to yield displacement is frequently

characterized as the displacement ductility ratio, μ. This characterization of

Second Edition
component performance corresponds to the ductility-limited damage state
category previously discussed.
Using the ASCE/SEI 41-06 backbone curve and the Immediate Occupancy
(IO) and Life Safety (LS) acceptability limits of Δ/Δy equal to 1.8 and 3.3,
respectively, Damage State 1 (DS1) can be inferred to occur after the IO
acceptability limit has been exceeded but before the LS acceptability limit is
reached. Thus, DS1 will be assumed to occur at a Δ/Δy = 2.5, as shown in
Figure 6-9.
Figure 6-9 Determining plywood diaphragm damage states.
As diaphragm deformations increase, damage requiring repair of nailing,

sheathing, and framing will occur. The onset of this damage will be defined
as Damage State 2 (DS2) and be assumed to occur at a deformation level of
Δ/ Δy = 3.5 (slightly greater than the LS acceptability limit).
As diaphragm deformations continue further, a level of damage will be

reached at which complete resheathing and reframing over approximately
25% of the diaphragm span nearest to the in-plane walls will be required.
This damage state is defined as Damage State 3 (DS3) and is assumed to
occur at Δ/Δy = 4.25. This corresponds to a deformation slightly beyond the
Collapse Prevention acceptability limit yet short of the complete loss of
residual component strength.
The five-step procedure presented in Volume 1, Section 3.8 characterizing

ductility-limited damage states is illustrated below.

Second Edition
Step 1. Calculate the median value, θ, and dispersion, β, for the onset of
yield of the element using Volume 1 Equation 3-1:
θ = Cq e(2.054 β )φ RN
1
The yield strength of a plywood diaphragm can be determined using the load
resistance factor design (LRFD) procedures contained in ANSI/AF&PA
SDPWS-2005, Special Design Provisions for Wind and Seismic (AF&PA,
2005b). For a ½-inch thick, Structural I plywood, blocked diaphragm with
10d common nails spaced at 6 inches along diaphragm boundaries and at all
panel edges (with 2 inch wide sub-framing), Table 4.2A of ANSI/AF&PA
SDPWS-2005 provides a nominal unit shear capacity, vs, of 640 plf. For
LRFD, a resistance factor, φD , with a value of 0.8 is applied to this value for
seismic demands.
Thus φ RN = 0.8(640) = 512 plf
Cq is a coefficient to adjust the median value based on sensitivity of the

behavior mode to construction quality. For this example, Cq will be taken as
1.0.
The dispersion parameter, β1, is composed of three parts intended to reflect

uncertainty associated with the onset of yield and is calculated as follows:
β1 = β D2 + β M2 + βC2
where:
βD = variability of the predictive equation used to calculate component
yield
βM = variability in material properties
βC = variability associated construction quality and care
For this example, βD is taken as 0.1 to reflect the relatively complex behavior
of plywood, nails, and sub-framing, βM, is taken as 0.2 (Volume 1, Table 3-4)
to reflect the variability of visually graded lumber products and βC will be
taken as 0.1 to reflect an average degree of construction quality control and a
moderate sensitivity to construction quality. Therefore,
β1 = (0.1) 2 + (0.2) 2 + (0.1) 2 = 0.24
The median value, θ, for the onset of yield is calculated to be:
θ = (1.0 ) e( 2.054×0.24) ( 512 )
= 1.64 (512) = 838 plf

Second Edition
Step 2. Calculate the median deformation of the element, δy, at the onset of
yield, using elastic deformation analysis under the applied median yield load.
A median estimate of the midspan deformation, δy, at yield can be calculated

using ANSI/AF&PA SDPWS-2005 Equation 4.2.1 as follows:
5v ye L3 0.25v ye L ∆c X
δy = + +∑
8 EAW 1000Ga 2W
where:
vye = median estimate of diaphragm strength at yield: 838 plf
L = diaphragm span (ft)
E = diaphragm chord modulus of elasticity (psi)
A = diaphragm chord cross sectional area (in2)
W = diaphragm width (ft)
Ga = diaphragm effective shear stiffness including nail slip and panel
shear deformation (kips/inch)
Δc = diaphragm chord slip at yield
X = distance from chord splice to nearest support (ft)
For the diaphragm transverse direction:

L = 120 ft
E = 29,000,000 psi (2- #5 rebar)
A = 0.62 in2
W = 40 ft
vye = 838 plf
Ga = 24.0 kips/inch
Δc X = Assumed to be zero for this example
Therefore,
( 5)(838)(120 ) ( 0.25)(838)(120 ) + 0
3
=δy +
(8)( 29,000,000 )( 0.62 )( 40 ) 1000 ( 24 )
δ y = 1.26 + 1.05 + 0
δ y = 2.3 inches

Second Edition
Note that the equation used to estimate diaphragm mid-span deflection at
yield is based on a uniformly distributed loading pattern over the span
length. Such a distribution may not be a good approximation of the likely
pattern of diaphragm loading for a flexible diaphragm system (ASCE/SEI
41-06, Commentary to Section 3.2.4). To account for the uncertainty
associated with the use of this selected deformation formulation, an
adjustment to the yield dispersion, β1, is made in Step 5 below.
Step 3. For DS1, μ = Δ/Δy = 2.5 and therefore this damage state behavior
mode is classified as “moderate ductility.” For DS2, μ = 3.5 and is also
classified as “moderate ductility.” For DS3, μ = 4.25 which corresponds to
“high ductility.”
Step 4. The median displacement estimates for each damage state are
calculated with θ = µδ y .
Step 5. The dispersion associated with each of the damage states is given
by Volume 1, Equation 3-4:
β = β12 + 0.16 ≤ 0.6
As noted in Step 2, the formulation used to estimate the median value of

diaphragm deflection at yield does not reflect a good approximation of the
distribution of inertial loading. The added uncertainty of using this
formulation will be addressed by modification of the dispersion. The
previously calculated β1 (see Step 1) is recalculated using a βD value of 0.2
in lieu of 0.1.
The resulting value of β1 is
( 0.2 ) + ( 0.2 ) + ( 0.1) = 0.3

2 2 2
β1 =
using the revised β1 value,
( 0.3)
2
=β + 0.16 = 0.5
Table 6-2 summarizes the damage states and median mid-span

displacement estimates determined by calculation for the transverse span
of the plywood diaphragm governed by nail pullout and panel buckling
behavior limiting mechanisms. Figure 6-10 plots the resulting fragility
functions for these three damage states, assuming lognormal distributions.

Second Edition
Table 6-2 Damage State Descriptions for Plywood Roof Diaphragm - Transverse
Damage Displacement
State Damage Description Repair Description (in) Dispersion
DS1 Tearing of roofing materials Remove existing flashing along wall. 5.8 0.5
and flashing at wall supports. Remove 4-foot width of roofing at
No damage to plywood and wall. Install new roofing and flashing.
framing below.
DS2 In addition to DS1 damage, Extend DS1 repairs to an 8-foot width, 8.1 0.5
nail pull out, framing remove damaged nails and install new
splitting and plywood tearing nails where no splitting of framing
occurs. occurs (25% of nails over 8 foot width
of diaphragm along wall). Remove
and replace plywood and framing
including ledgers at split framing or
damaged plywood (25% of 8 foot
width of diaphragm along wall).
DS3 In addition to damage Remove and replace roofing, flashing, 9.8 0.5
described for DS1 and DS2, plywood, framing (girder not replaced)
plywood panels pull free of over 25% of diaphragm span and full
sub-framing and framing depth of wall at each diaphragm
splits over approximately support.
25% of the diaphragm span
at each diaphragm support.
0.9
0.8
0.7
Probability of Exceedance
0.6
DS1
0.5 DS2
DS3
0.4
0.3
0.2
0.1
0
0 5 10 15 20 25 30
Displacement (inches)
Figure 6-10 Fragility functions for plywood diaphragm in transverse direction.

Second Edition
The demand parameter used with these fragilities is the mid-span diaphragm
displacement, Δ. Values for this parameter should be calculated from a
response history analysis using a nonlinear characterization of the diaphragm
construction.
6.3.2 Tilt-Up Walls
The six-inch thick, reinforced concrete walls are subjected to in-plane and
out-of-plane seismic demands, which induce fundamentally different types of
performance. In-plane performance can be adequately represented by the
provided fragility specifications for low aspect ratio (h/l < 1.0) concrete walls
(B1044.001 through B1044.083). These specifications reflect the
performance characteristics of low aspect ratio concrete walls for which the
relative capacity of the wall-to-foundation connections are adequate to
develop the shear and overturning capacity of the wall without a pre-emptive
failure of the connection. For this example, it has been assumed that the
panel base anchorages are adequate to provide this condition. For buildings
where this is not the case, fragility specifications for wall anchorages should
be developed using the procedures of Volume 1, Section 3.8.
Out-of-plane seismic demands on slender concrete walls can also produce

damage. When walls are subjected to out-of-plane accelerations that exceed
the wall’s flexural and shear resistance, cracking will occur. Initially, the
cracking will require cosmetic repairs to interior and exterior finishes. At
higher levels of acceleration, flexural reinforcement will yield and larger
cracks will result, requiring patching or epoxy injection to reconstitute the
wall’s weather protection. At yet larger accelerations, flexural yielding will
lead to spalling and degradation that when coupled with the effects of the
supporting weight of the wall above can result in wall collapse.
In addition to the effects of acceleration on out-of-plane wall behavior, large

deformations of the roof system can induce significant flexural demands in
the attached walls. These demands produce wall flexure due to the
eccentricity of the wall weight from its vertical support at grade. These
effects are generally described as P-Δ effects.
Damage will initiate in the wall at the location where flexural demand is a
maximum. Maximum flexural demand can be determined by analysis, by
modeling representative wall panels as nonlinear multi-segment beam
elements, extending from the ground to the roof diaphragm, and by
distributing the wall panel mass along these elements. Response history
analysis will compute the moment demand along the wall at each time step

Second Edition
and permit prediction of a maximum value of moment demand for each
analysis.
The moment curvature relationship for the tilt-up wall subject to out-of-plane
flexure can be idealized as illustrated in Figure 6-11.
Figure 6-11 Idealized moment-rotation relationship.
In order to construct fragility curves following the procedures of Volume 1,

Section 3.8, the median value of strength and dispersion have to be
calculated. The following example illustrates this calculation.
In order to fulfill Volume 1, Equation 3-1, the component design strength

needs to be calculated in accordance with an appropriate industry standard.
To accomplish this, the moment values associated with the control points of
the curve can be approximated as follows:
Step 1. Establish moment values associated with the control points A, B,

and C, in the moment-rotation relationship of Figure 6-11.
Point A: Cracking Moment

ACI 318-05, Building Code Requirements for Structural Concrete and
Commentary, (ACI, 2005) provides the following formula for cracking
moment:
fr I g
M cr =
yt
where:
fr = modulus of rupture and for normal weight concrete can be taken as
7.5 f c′
Ig = gross moment of inertia of section (neglecting reinforcing)
yt = distance from centroidal axis of section to tension face (neglecting
reinforcing)

Second Edition
For the tilt-up wall section
= fr 7.5
= 4000 474.3 psi
12(6)3
=Ig = 216 in4
12
yt = 3.0 in
Thus,
(474.3)(216)
=M cr = 34149.6 lb-in
3.0
or
M cr = 2846 lb-ft
Point B: Yield Moment
The yield moment can be calculated using ACI 318-05. The yield moment
can be calculated as
M y = As au d
where:
As = Area of longitudinal reinforcement
 1 − 0.59q 
au = f y  
 12000 
As f y
q =
bd fc′
d = distance from extreme compression fiber to centroid of longitudinal steel
fy = reinforcement yield stress
f c′ = concrete strength
b = width of section
Substituting
As = 0.31 in2 (#5@12 inches), fy = 60,000 psi, f c′ = 4,000 psi, b = 12 in, d = 3 in.
60,000 0.31(60,000)
au = (1 − (0.59)( )
12,000 (12)(3)(4,000)
au = 4.62
My = (0.31)(4.62)(3) = 4.30k – 1 or My = 4297
Point C: Peak Moment
The peak moment is estimated to occur at a rotation that is approximately 3

to 4 times that which produced My. The peak moment can be assumed to be
120% of the yield moment, MB = 5,156 lb-ft.

Second Edition
Step 2. Establish damage state descriptions.
The following damage states are considered:

• DS1: Onset of flexural cracking requiring only cosmetic repair (minor
patching and painting)
• DS2: Post-yield flexural cracking including cracking and spalling
requiring patching, possibly epoxy injection, and painting. Figure 6-12
illustrates a damage state similar to DS2.
• DS3: Significant spalling and degradation leading to partial wall collapse
requiring removal and reconstruction.
Figure 6-12 Photo showing DS2 level cracking (Selna, 1981).
Step 3. Calculate median strength and dispersion for Damage State 1
For DS1, the demand will be taken as the moment which is 10% of the
difference between points A and B or
MDS1 = 2,846 + 0.1(4,297 – 2,846)

MDS1 = 2,991 lb-ft
Define this as a strength-limited damage state characterized by ductile

performance (flexure).
Determine the median strength as
θ = Cq e(2.054 β )φ Rn
1

Second Edition
From ACI 318-05:
φ = 0.9 for tension controlled actions
φ Rn φ=
= M DS 1 0.9(2,991)
= 2,692 lb-ft
β1 = β D2 + β M2 + βC2
For the DS1 estimate (cracking strength), select
=βD 0.10,
= βM 0.12,
= βC 0.10
β1 = (0.1)2 + (0.12)2 + (0.1)2 = 0.185 or β1 = 0.2
Cq = 1.0
θ = (1.0)e 2.054(0.2) (2,692)
θ = 4,060 lb-ft
Assign a dispersion
=β β12 + 0.16 where β1 is defined above
β= (0.2) 2 + 0.16= 0.45 or β = 0.5
For DS1, θ = 4,060 lb-ft, β = 0.5
For DS2, the demand will be taken as 25% of the difference between points
B and C’ or
M DS 2 =4, 297 + 0.25 ( 5,156 − 4, 297 ) =4,511 lb-ft
Similar to DS1, this will be defined as a strength-limited damage state

characterized by ductile performance (flexure).
Determine the median strength as

θ = Cq e(2.054 β1 )φ Rn
M DS 2 0.9 ( 4,511lb-ft
φ Rn φ=
= = ) 4,060 lb-ft
For the DS2 strength estimate (post yield), select
=βD 0.05,
= βM 0.12,
= βC 0.10
β1 = (0.05) 2 + (0.12) 2 + (0.10) 2 = 0.164 or β1 = 0.2
Cq = 1.0
θ (1.0
= ) e( 2.054)( 0.2) ( 4,060 ) 6,122 lb-ft
Assign a dispersion:
β= (0.2) 2 + 0.16= 0.45 or β= 0.5

Second Edition
For DS2: θ= 6,122 lb-ft, β = 0.5
The occurrence of DS3 will be considered as a form of partial collapse and

will have a collapse fragility associated with it, based on incremental
dynamic analysis. This procedure is not illustrated here.
Table 6-3 summarizes the out-of-plane damage states and demand

parameters for the longitudinal tilt-up concrete walls. The walls are
subjected to out-of-plane acceleration and displacement effects that produce
a wall flexure behavior mechanism. Demands, in the form of peak bending
moment, are predicted by modeling individual wall panels as strings of
nonlinear beam elements with distributed mass, extending from the building
base to the roof diaphragm. Figure 6-13 presents a plot of the fragility
functions for damage states DS1 and DS2.
Table 6-3 Damage State Descriptions for Tilt-Up Walls—Out-of-Plane

Bending
Damage Moment
State Damage Description Repair Description (lb-ft) Dispersion
DS1 Onset of flexural Over 4 foot distance at mid-height of wall grout 4,060 0.5
cracking requiring patch surface and repaint wall. Work occurs over
cosmetic repair middle third of diaphragm span.
DS2 Post yield flexural In addition to work done for DS1, over 6-foot 6,122 0.5
cracking with light distance at mid-height of wall, patch spalls, epoxy
spalling, requiring inject 2 cracks (20 feet long each) and paint wall.
patching, possible Work occurs over middle third of diaphragm span.
epoxy injection, and
painting.
DS3 Significant spalling and Determined using incremental dynamic analysis and - -
degradation leading to predicted by a building collapse fragility function.
partial wall collapse.

Second Edition
1
0.9
0.8
0.7
Probability of Damage State Occurence
0.6
DS1
0.5
DS2
0.4
0.3
0.2
0.1
0
0 10000 20000 30000
Moment (lb-ft)
Figure 6-13 Fragility functions for out-of-plane wall flexure.
6.3.3 Wall to Roof Attachments
As previously described, the wall and roof are attached to transfer in-plane
and out-of-plane wall and diaphragm demands. Separate attachments serve
to resist demands acting in and out of the plane of the perimeter wall.
Damage can occur to each type of these attachments. Fragility functions
which were derived analytically will be developed for each of these
attachments in the sections to follow.
6.3.3.1 In-Plane Attachment – Failure at Ledger Bolting
Figure 6-3 illustrates the in-plane wall attachment detail. Attachment

components include plywood nailing to the ledger and ledger bolting to the
concrete wall. Nailing performance was previously addressed as part of the
fragility specification for the plywood diaphragm. The ledger bolting system
has four basic behavior modes that can result in damage. As identified in

Second Edition
ANSI AF&PA National Design Specification for Wood Construction
(AF&PA, 2005a) the four potential limit states are:
• Fastener bearing failure in main or side member
• Fastener rotation failure without bending
• Fastener plastic hinge failure in main or side member
• Fastener plastic hinge failure in main and side member
The National Design Specification for Wood Construction provides LRFD

design equations for each of these modes. The lowest value obtained from
the yield limit equations is listed in National Design Specification for Wood
Construction Table 11E.
These four limit states assume that system behavior is governed by the wood
or bolt components. This requires adequate anchorage of the bolt into the
concrete to preclude a preemptive failure of the concrete anchorage. This
requires at least six inches of anchor embedment. Recognizing that only 4.5
inches of embedment has been provided in the example building, additional
limit states associated with bolt and concrete behavior need to be considered.
This will be done after investigating the potential limit states for the
wood/bolt interface.
In this example, the wood/bolt capacity is limited by fastener plastic hinge

failure in main and side member. This capacity is given by the formula:
D2 2 Fem Fyb
Z=
Rd 3(1 + Re )
where:
D = 0.625 inch
Fem = Dowel bearing strength in concrete = 7500 psi
Fyb = Dowel bending yield strength = 45,000 psi
Fes = Dowel bearing strength in wood = 5,600 psi (G = 0.5)
Rd = 3.2 Kθ
Fem
Re =
Fes
θ 
Kθ = 1 + 0.25  
 90 
θ = angle between direction of grain and application of load (for this
example, loading is parallel to the grain and therefore 0)

Second Edition
Substituting the numerical values yields:
( 0.625) 2 ( 7500 )( 45,000 )
2
Z= = 1197 pounds
3.2 3 (1 + 1.34 )
Damage in this mode consists of localized crushing of the ledger at the bolt
head, bolt bending displacement and the onset of wood crushing at the
bolt/concrete interface.
This force value can be converted to an LRFD capacity using National
Design Specification for Wood Construction Table N1 and Table N2.
K F Z = φ Rn
where KF is given in National Design Specification for Wood Construction
Table N1 as (2.16 /ϕ) for all connections and ϕ is given in Table N2 as 0.65
for all connections. Yielding:
2.16
=φ Rn (= )(1200) 3,988 lb
0.65
Damage associated with this limit state is severe lateral distortion of the bolt
coupled with significant crushing and widening of the hole in the wood
ledger through which the bolt passes. Repair of this damage requires
installation of new bolts (post-installed) in new holes. In some locations,
replacement of the ledger may be necessary due to ledger splitting.
Development of a fragility function to reflect this damage state is illustrated
in the following steps:
Step 1. Define this as a strength-limited damage state characterized by

ductile performance.
Determine the median strength as:

θ = Cq e(2.054β )φ Rn
1
where φ Rn = 3,988 lb as determined above.
β1 = β D 2 + β M 2 + βC 2
For the dispersions, select,
=β D 0.10,
= β M 0.20,
= βC 0.10
β1 = (0.10) 2 + (0.20) 2 + (0.10) 2 = 0.24

Cq = 1.0
Therefore, θ 1.0
= = e( 2.054)( 0.24) ( 3,988 lb ) 6,549 lb

Second Edition
Step 2. Assign a dispersion:
=β ( β1 ) 2 + 0.16
β
= (0.24) 2 + 0.16 β 0.5
= 0.47 or =
Table 6-4 summarizes the fragility data for in-plane damage to the
ledger/bolt components. Figure 6-14 plots this fragility function.
Table 6-4 Damage State Description for Ledger to Bolt Wall Attachment—In-Plane
Damage Bolt Force
State Damage Description Repair Description (lbs) Dispersion
DS1 Severe lateral bolt distortion and wood Install new epoxy anchors into ledger 6,549 0.5
crushing at bolt with some ledger & concrete wall. Replace 20% of
splitting ledgers.
1
0.9
Probability of Damage State Initiation
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Bolt Force (Thousands lbs)
Figure 6-14 Fragility function for in plane ledger to wall connection.
6.3.3.2 In-Plane Attachment – Failure at Bolt/Concrete

Interface
The preceding fragility function assumes that damage only occurs to the
wood ledger and bolt system. Damage may also occur to the bolt/concrete
interface due to limited anchor bolt embedment. ACI 318-05 Appendix D
specifies design equations for this behavior. The following strength limit
states are considered:
• Steel strength of anchor in shear
• Concrete breakout strength of anchor in shear
• Concrete pryout strength of anchor in shear

Second Edition
The median estimate of strength associated with each of these potential limit
states will be evaluated and the lowest value will be selected for fragility
function development.
Steel Strength of Anchor in Shear. For a cast-in-place headed stud anchor,

ACI 318-05 Appendix D provides the following capacity:
Vsa = nAse futa
where n is the number of anchors in a group, taken as 1 for this example, Ase
is the effective cross section of the anchor, 0.307 in2 for a 5/8 inch diameter
stud anchor, and futa is the tensile strength of anchor, taken as 58 ksi.
Yielding:
=Vsa (1)(0.307)(58,000)
= 17,806 lbs
Concrete Breakout Strength of Anchor in Shear. If it is assumed that

anchor bolts will be placed no closer than 12 inches from the edge of the tilt-
up panel, ACI 318-05 Appendix D provides the following capacity:
A
Vcb = vc (ψ ed )v (ψ c )v Vb
Avco
where:
Avc = (2)(1.5)(Cai)ha
with:
Cai = edge distance = 12 inches
ha = wall thickness = 6 inches
Avc = (2)(1.5)(12)(6) = 216 in2
Avco = 4.5(Cai)2 = 4.5(12)2 = 648
(ψ ed )ν = 1.0 for Ca2 > 1.5 Ca, Ca2 = 24 + 5 = 29
(ψ c )ν = 1.4 for no cracking at service load levels

le 0.2
Vb = 7( ) ( d0 )( f c′ )(Cai )1.5
d0
with:
le = heff = 4.5 - 0.31 = 4.2 inches
d0 = anchor diameter = 0.625 inches
4.2 0.2
Vb 7(
= ) ( 0.625)( 4,000)(12)1.5 12,047 lbs
0.625

Second Edition
216
Therefore Vcb (=
= )(1.0)(1.4)(12,047) 5,622 lbs
648
Concrete Pryout Strength of Anchor in Shear. ACI 318-05 Appendix D,
provides the following pryout capacity:
Vcb = kcp N cb
where:
kcp = 2.0 for heff ≥ 2.5
ANc
N cb = (ψ ed ) N (ψ c ) N (ψ cp ) N N b
ANco
with:
ANc A=
= Nc 0 ) 2 9(4.2)
9(heff= = 2
158.8 in2
(ψ ed ) N = 1.0 for C a ≥ 1.5heff
(ψ c ) N = 1.25 for cast-in-place anchors
(ψ cp ) N = 1.0 for cast-in-place anchors
N b = k c f c ' (heff )1.5

kc = 24 for cast-in-place anchors
=N b 24
= 4, 000(4.2)1.5 13, 065 lbs
158.8
=N cb = (1.0)(1.25)(1.0)(13,065) 16,331 lbs
158.8
=Vcp 2.0(16,331)
= 32,662 lbs
To determine the governing strength, the median strength estimate for each
potential limit state must be determined and compared. This can be done
using Volume 1, Equation 3-1 for ductile modes and Volume 1, Equation
3-2 for brittle modes.
θ ductile = Cq e(2.054 β )φ Rn
θbrittle = Cq e(2.81β )φ Rn
Note that ACI 318-05 Appendix D requires that in regions of high seismic
risk the shear strength shall be taken as 0.75φVn , thus
φ Rn = ( 0.75 )( 0.75 )Vn . Table 6-5 summarizes this calculation and the
corresponding median strength estimates for Cq = 1.0 and β1 = 0.38

Second Edition
Table 6-5 Median In-Plane Limit State Strengths for Anchor Bolt
Strength
Limit State φ Rn (lbs) Failure Category (lbs)
Steel strength of anchor in shear 10,016 Ductile 18,548
Concrete breakout strength of 3,162 Brittle 7,346
anchor in shear
Concrete pryout strength of anchor 18,372 Brittle 42,684
in shear
The limit state with the lowest estimated median strength is the concrete
breakout mode. The damage associated with this limit state is the breakout and
spalling of the concrete at the ends of the panels.
Note that for bolts located away from the panel edges, the strength of the
anchor in shear will produce the lowest estimate of median strength among the
three limit states. Its estimated capacity is sufficiently greater than that of the
strength associated with the limit state for the wood/bolt components (18,548
lbs vs. 6,549 lbs), thus it can be judged unlikely to occur.
Comparison of the median strength estimate of the in-plane wall attachment

(ledger/bolt) of 6,549 lbs to the median strength estimate of bolts embedded at
the panel edges (7,346 lbs), suggests that there is uncertainty as to which mode
will actually occur. It is possible to treat these as two separate sequential
damage states, or alternatively, as mutually exclusive damage states with a
common median value, taken as the average of 6,549 lbs and 7,346 lbs.
To utilize the in-plane anchorage fragility, it is necessary to extract the anchor

bolt force from the analysis. This can be done by dividing the diaphragm edge
shear by the number of bolts.
6.3.3.3 Out-of-Plane Attachment
Figure 6-3 and Figure 6-4 illustrate the out-of-plane attachment of the roof to
the wall. This attachment consists of two holdown devices fabricated from
12-gauge steel spaced at 8 feet on center. The devices are bolted together
through a purlin at the transverse walls and through a doubled subpurlin
along the longitudinal walls. Each device is also attached to a threaded rod
that is embedded into the perimeter concrete wall with a washer and nut to
resist pullout. The purlins and subpurlins are nailed to the plywood
diaphragm to transfer the out-of-plane wall forces into the horizontal
diaphragm system. The wall spans horizontally 8 feet between holdown
devices resisting out-of-plane forces. The attachment provides resistance to

Second Edition
the inertial out-of-plane wall forces and provides restraint against P-Δ wall
forces at the roof level.
The attachment may reach its limiting capacity in one of the following ways:
• Purlin to diaphragm nailing failure
• Holdown device bolt failure at purlin or subpurlin
• Holdown device yielding or fracture
• Bolt yielding in tension
• Bolt failure from concrete breakout
Using the example configuration, a capacity will be estimated for each of the
mechanisms outlined above. The mechanism that has the lowest median
capacity estimate will be used to develop a fragility function for the
assembly.
Purlin to Diaphragm Nailing Failure. Diaphragm nailing consists of 10d

nails at 6 inches on center over an 8-foot length of purlin or subpurlins. The
capacity can be taken from AF&PA/ASCE 16-95, Standard Load and
Resistance Factor Design (LRFD) for Engineered Wood Construction,
(ASCE, 1996) Table C4.2.2.D as 260 pounds per nail. The total capacity is
8320 pounds.
Holdown Device Bolt Failure at Purlin or Subpurlin. The bolting consists
of 2 5/8” Ф machine bolts through the 4×12 purlin or two 2×4 subpurlins,
with 12 gage sideplates.
For the subpurlins (3 inches thick), National Design Specification for Wood
Construction Table 11G yields a Z11 value of 2,270 lbs. For 2 bolts, the
capacity is
2.16
C=
DB 4,540 lbs × = 15,087 lbs = ϕRn
0.65
Holdown Device Yielding or Fracture. The Allowable Stress Design
(ASD) capacity of the double holdown device is given in the manufacturer’s
catalog as 2 × 1,956 = 3,913 lbs. Converting to an LRFD design strength
yields:
2.16
C=
Dd 3,913 lbs × = 13,002 lbs = ϕRn
0.65
Bolt Yielding in Tension. ACI 318-05 Appendix D, provides the following
tensile capacity:
2
= C DAB (0.75)
= (58)(.226)(2) 14,747 lbs = ϕRn

Second Edition
Bolt Failure from Concrete Breakout. ACI 318-05 Appendix D provides the
following design breakout strength:
A
Ncb = NC (ψ ec )N (ψ ed )N (ψ c )N (ψ cp ) Nb
ANC 0
where:
N b = kc f c (hef )1.5
=Nb 24
= 4,000(4.25)1.5 13, 299 lbs
= = 302.4 in2
ANc (14.75)(20.5)
2
ANC 0 9(4.25)
= = 162.6 in2
(ψ ec ) N = 1.0 (no eccentricity)
(ψ ed ) N = 1.0 (Camin > 6.375)
(ψ c ) N = 1.25 (cast-in-place)
ψ cp
= 1.0
Substituting:
302.4
Ncb = (1.0)(1.0)(1.25)(1.0)(13, 299) 30,916 lbs
162.6
=CDBr (0.70)(0.75)(30,916)
= 16, 231 lbs = ϕRn
Table 6-6 summarizes the design strengths and estimated median strength
values, θ, for each of these limit states. For all failure modes, a Cq value of 1.0
and β value of 0.24 has been used.
Table 6-6 Limit State Strengths Out-Of-Plane Attachment

Limit State φ Rn (lbs) Failure Category θ (lbs)
Diaphragm nailing failure 8,320 lbs Ductile 13,621
Holdown device bolt failure 15,087 lbs Ductile 24,700
Holdown device yielding 13,002 lbs Ductile 21,286
Bolt yielding in tension 14,747 lb Ductile 24,143
Bolt failure at concrete breakout 16,231 lbs Brittle 31,859
The lowest median capacity estimate is associated with diaphragm nailing.

This estimate is significantly lower than that for other limit states, suggesting
that it is the most likely to occur. The damage that can be anticipated from the
occurrence of this damage state consists of splitting of the purlin and subpurlin
and tear through of the nails through the diaphragm plywood. Replacement of
the purlin and plywood and renailing are required to restore the original
integrity.

Second Edition
Development of the fragility function for this limit state is as follows:
Step 1. Define this as a strength-limited damage state characterized by ductile performance.
Determine the median strength as:
θ = Cq e(2.054 β )φ Rn
1
where φ Rn = 8,320 lbs

β1 = βD2 + βM 2 + βC2
For the dispersions, assign
=βD 0.10,
= βM 0.20,
= βC 0.10
β1 = (0.10)2 + (0.20)2 + (0.10)2 = 0.24
Cq = 1.0
Therefore, θ (1.0)
= = e (2.054)(0.24) (8,320) 13,621 lbs
Step 2. Assign a dispersion
=β ( β1 )2 + 0.16
β
= (0.24) 2 + 0.16
= 0.54 or β = 0.5
Table 6-7 summarizes the fragility function for out-of-plane wall-to-roof

anchorage. Figure 6-15 plots this fragility function.
Table 6-7 Damage State Description for Wall Attachment (Purlin Nailing) – Out-of-Plane
Damage Median Strength
State Damage Description Repair Description (lbs) Dispersion
DS1 Purlin/subpurlin splitting, nail Shorings as needed, purlin, subpurlin 13,621 0.5
tear out through ply, roofing replacement, plywood reinstallation
tearing and distortion. and nailing, reroofing. Damage
occurs over middle 1/3 of diaphragm
span.

Second Edition
Figure 6-15 Fragility function for out-of-plane wall-to-roof nailing.
If the analytical model is configured as previously described, with wall

panels modeled as beam elements with distributed mass spanning between
the ground and roof, then the out-of-plane anchorage force demand can be
directly determined from the analyses.
6.4 User-Defined Fragilities in PACT
User defined fragility specifications can be input into PACT using the
Fragility Specification Manager. This utility is accessed via the Control
Panel. Additional information regarding this utility can be found in
Appendix C.
New fragility specifications are input by first entering a new fragility

specification ID (Uniformat numbering used for components, see Chapter 2).
This action will insert a blank specification form into the sequence displayed
in the overview screen. Figure 6-16 illustrates the identification of a new
fragility specification (C9994.002) into the library. Figure 6-17 illustrates
the screen presented when the fragility specification is selected and the
Fragility Specification Details tab at the top of the screen is clicked. The
user can input the fragility specification Name, Description, and Basic
Units of measure. The Demand Parameter can be drift, acceleration, or any
user defined parameter, such as out-of-plane anchor force.

Second Edition
Figure 6-16 PACT Fragility Specification Manager showing new
specification.
Figure 6-17 Fragility specification data input form.

Second Edition
Figure 6-18 shows the Damage State tab where new damage state data
including median values and dispersions can be assigned. For each damage
state, a consequence function must be entered. This is done via the Add
Consequence tab at the left of the screen. For information on developing
and inputting consequence function data, see Chapter 8. Multiple damage
states can be inserted into the fragility specification by repeating the process.
Figure 6-18 New fragility damage state definition form.

Second Edition
Chapter 7
Nonstructural Fragility
Calculation
7.1 Introduction
Volume 1, Section 3.8 presents procedures for calculation of fragility

parameters for component damage states associated with limit states that can
be predicted by calculation. This chapter illustrates those procedures for
nonstructural components for damage states that can be predicted on the
basis of the following calculated structural limit states:
• Strength-limited damage states when the occurrence of damage can be
predicted by a demand exceeding a calculable force capacity
• Ductility-limited damage states when the occurrence of damage can be
predicted by a demand exceeding a calculable amount of ductility
• Displacement-limited damage states when the occurrence of damage can
be predicted by a displacement demand exceeding a calculable value
• Code-based limit states when damage can be predicted as a function of a
code-specified demand parameter
• Overturning limit states when predicting overturning failure of
unanchored components
• Sliding limit states when predicting sliding of unanchored components
Note that this methodology is generally not applicable to damage states

associated with loss of electrical or mechanical function as these types of
failures cannot typically be predicted by structural calculation. Shake table
testing or earthquake experience data are generally necessary to develop
fragility functions associated with loss of mechanical or electrical function,
unless the loss of functions can be directly tied to a structural limit state.
7.2 Unanchored Components
Unanchored components are vulnerable to damage associated with

overturning and sliding. Their vulnerability is dependent on the static
coefficient of friction between the component and its resting surface, μs,
listed in Table 7-1, the amount of distance the component can slide before
experiencing damage, and the component’s aspect ratio and weight
FEMA P-58-2 7: Nonstructural Fragility Calculation 7-1

Second Edition
distribution. When sliding occurs, the amount of sliding displacement is
controlled by the kinetic coefficient of friction. Volume 1, Section 3.8
presents a methodology for calculating fragility parameters for unanchored
components based on predictive equations presented in ASCE/SEI 43-05,
Seismic Design Criteria for Structures, Systems and Components in Nuclear
Facilities, (ASCE, 2005b). This section illustrates the application of those
procedures.
Table 7-1 Static Coefficient of Friction for Common Surface

Combinations
Contact Surfaces μs
Carpet-steel 0.45
Wood-steel 0.40
Poly/tetra/fluoro/ethylene - steel 0.27
The demand parameter used to determine the onset of overturning or sliding

is peak total floor velocity, VPT. VPT is defined as the peak value of the vector
sum of the ground velocity and the relative velocity between the floor and the
ground.
7.2.1 Overturning
To illustrate the procedure, overturning fragility parameters for a fully

loaded, 5-shelf metal bookcase is calculated. Figure 7-1 defines the basic
geometric parameters including the tipping angle, α, the minimum horizontal
projected distance from the center of gravity (cg) to the edge of the
component’s base, b, and the height of the component’s center of gravity
above the base, h.
Figure 7-1 Overturning parameters for unanchored objects.
The goal of the procedure is to estimate the median peak total floor velocity
at which overturning initiates and a dispersion to reflect the uncertainty.
7-2 7: Nonstructural Fragility Calculation FEMA P-58-2

Second Edition
The bookcase is 71 inches high and has a narrow dimension of 12-5/8 inches.
It rests on hardwood floor. From Table 7-1, the static coefficient of friction
between the bottom of the bookcase and the hardwood floor is 0.40.
h = 71/2 = 35.5 inches
b = 12.625 / 2 = 6.31 inches
The units for the calculation will be in inches, seconds and radians.
Step 1. Determine the critical tipping angle, α (in units of radians) at which
the object is on the verge of overturning from Volume 1, Equation 3-5:
b
α = arctan  
h 
b/h = 6.31/35.5 = 0.178
=α arctan(0.178)
= 0.176 radians
Step 2. Determine the horizontal spectral acceleration overturning capacity

of the component, SaO, in units of the acceleration due to gravity, g, from the
critical tip angle α and the component’s dimensions with Volume 1, Equation
3-6:
2B
SaO =
α
where:
b 6.31
=B cos (α ) + sin (α=
) − 1 cos ( 0.176 ) + sin ( 0.176=
) − 1 0.016
h 35
2 B 2 ( 0.016 )
S=
aO = = 0.18 g
α 0.176
Step 3. Comparison of the value of SaO (0.18) with the static coefficient of
friction (0.4) reveals that the object will likely overturn rather than slide.
Step 4. Determine rocking frequency, fr, with Volume 1, Equation 3-10 and
damping ratio, βr, with Volume 1, Equation 3-11:
1 2gB
fr =
2π C3α 2 h
γ
βr =
4π 2 + γ 2

Second Edition
where:
C3 is given by Volume 1, Equation 3-12:
4   b   4   6.31  
2 2
C3 =1 +    =1 +   = 1.38

3   h   3   35.5  
γ is given by Volume 1, Equation 3-13:

γ = -2ln(CR)
CR is given by Volume 1, Equation 3-14:
2 (b / h) 2 (.178 )
2 2
1
CR =
− 1−
= 0.954
=
C3 1.37
γ = -2ln(0.954) = 0.0942
1 2 gB 1 2 ( 386 )( 0.016 )
=fr = = 0.45 hz
2π C3α h 2π 1.38 ( 0.176 )2 ( 35.5 )
2
γ 0.0942
=βr = = 0.015
4π 2 + γ 2 4π 2 + 0.09422
Step 5. Determine the estimated median peak total floor velocity for
overturning with Volume 1, Equation 3-8:
S aO g
VPT =
2π f r Damp
where Damp is given by Volume 1, Equation 3-9:
Damp = 2.31 − 0.41 ln (1.5 ) =

2.31 − 0.41 ln (100 β r ) = 2.14
S aO g 0.178(386)
=VˆPT = = 11.4 in / sec
2π f r Damp 2π ( 0.45 ) 2.14
The median overturning value of VPT is 11.4 inches/second. A dispersion of

0.50 is recommended to reflect the variability of the prediction. Figure 7-2
plots this fragility function.

Second Edition
Figure 7-2 Fragility function for overturning of 5-shelf metal bookcase.
This procedure has been used to develop overturning fragility functions for a
range of common unanchored office furniture. Table 7-2 and Table 7-3
present the calculated fragility values and provide some basic information
used in the derivation of these parameters for metal bookcases and filing
cabinets, respectively. Dispersion, β, is taken as 0.5 for both. The static
coefficient of friction μs listed in these tables should be confirmed by the
user.
Table 7-2 Calculated Fragility Values for Metal Bookcases

Number of Depth Height
Shelves (inches) (inches) μs VˆPT (in/sec)
2 12.625 29 0.442 28.5
3 12.625 41 0.312 19.4
4 12.625 56 0.229 14.2
5 12.625 71 0.178 11.4
6 12.625 81.25 0.152 9.9
Table 7-3 Calculated Fragility Values for Filing Cabinets

Narrow
Number of Plan Dim Height
Drawers/Type (inches) (inches) μs VˆPT (in/sec)
2* Drawer/Vertical 15 24 0.64* 48.4
4 Drawer/Vertical 15 52 0.29 19.6
4 Drawer/Lateral 18.63 52.5 036 27.1
5 Drawer/Lateral 18.63 67.75 0.28 20.7
* Sliding likely to occur for common resting surfaces.

Second Edition
7.2.2 Sliding Displacement
Unanchored components mounted on surfaces with static coefficient of

friction, μs, less than the horizontal overturning spectral acceleration
capacity, SaO, as defined in Section 7.2.1, will likely slide rather than
overturn. Sliding components can be damaged by contact with other objects,
can rupture attached utilities, and slide off mounting surfaces. Volume 1,
Section 3.8 presents a methodology to determine the median peak floor
velocity at which sliding displacements are expected to exceed a user-defined
displacement, δ, at which damage is assumed to occur.
The methodology first determines whether the component will slide or

overturn, as described in Steps 1 through 3 for overturning, above.
If it is determined that the component will slide, the median peak floor
velocity, VˆPT , that causes sliding displacements in excess of the damage limit
is calculated with Volume 1, Equation 3-15:
2 µ D gδ
VˆPT =
1.37
Table 7-4 provides mean estimates of μD, the dynamic coefficient of friction
between the component and its supporting surface, for common contact
surfaces.
Table 7-4 Dynamic Coefficient of Friction for Common Surface

Combinations
Contact Surface µD
Carpet - steel 0.40
Wood - steel 0.32
Poly/tetra/fluoro/ethylene - steel 0.20
The median peak total floor velocity value is determined for an unanchored
component where μD, the assumed dynamic friction coefficient between the
component’s base and the bearing surface according to Table 7-4, is 0.40 and
the tolerable sliding displacement, δ, is 2 inches. For this example, it is
assumed that b/h for the component is such that the component slides, rather
than rocks.
Substituting g = 386.2 in/s2,
(2)(0.4)(386)(2)
=VˆPT = 18.14 in/sec
1.37
Under this methodology, a value of β of 0.5 is recommended. Figure 7-3 is a
plot of the resulting fragility function.

Second Edition
Figure 7-3 Fragility function for sliding of 5-shelf metal bookcase.
7.3 Anchored Components
In areas of high seismicity, nonstructural components are usually anchored.

The seismic vulnerability of these components to earthquake ground motions
is influenced by the capacity of the anchorage. The following sections
present two alternative approaches for calculating anchorage fragility
parameters for building components. The first of these approaches assumes
that the component has been anchored in accordance with the building code
provisions in effect at the time of installation. The second approach requires
the user to have sufficient knowledge of the anchorage details to permit
calculation of median demand values that correspond to anchorage failure
modes.
Building component anchorage details are generally unique to each building,

component type, and anchorage location within the building. Consequently,
it is not possible to define generalized fragility parameters for nonstructural
component anchorage.
7.3.1 Code-Based Limit State Determination of Anchorage

Fragility
7.3.1.1 Recent Code Requirements
Seismic design of nonstructural component anchorage typically uses an

equivalent static force method to determine required anchorage strength,
typically identified by the parameter, Fp.

Second Edition
Prior to 1997
Prior to the 1997 edition, the Uniform Building Code (UBC) defined Fp by
the equation:
Fp = ZI p C pW p
where Z is a seismic zone factor reflecting the intensity of anticipated ground

shaking, Cp is a nonstructural component coefficient, with a value assigned
from a tabulation of different component types and corresponding values, Ip
is a component importance factor usually assigned a value of 1.0 except for
components in essential facilities, and Wp is the component weight. For
critical components, the Ip factor could be assigned values of 1.25 or 1.5
depending on the code edition.
The component anchorage force, Fp, was used together with allowable stress
design procedures that permitted a 1/3 increase on anchorage capacities.
Nonstructural components located on the ground floor or in basement levels
were permitted to be designed for 2/3 of the anchorage force obtained from
the Fp equation. Anchorage of flexibly supported (i.e., vibration isolated)
components typically required a twofold increase in the Fp force to account
for dynamic amplification effects. For most nonstructural components
(including non-vibration isolated mechanical and electrical equipment
located above the first floor) in the highest seismic zone, the Fp coefficient
had a value of 0.3 for equipment on elevated floors and 0.2 at the ground
floor. Allowable stress loads can be converted to equivalent strength-level
design forces by applying a factor of 1.4 to the allowable stress load.
1997 Uniform Building Code
The 1997 UBC, Uniform Building Code (ICC, 1997), adopted the following
modified version of the Fp formula;
Ca I p a p
=Fp [1 + 3 hx hr ]W p
Rp
where Ca is a short period ground motion coefficient including soil condition

effects, Ip the component importance factor similar to earlier codes, ap is the
component amplification factor, Rp in the force reduction factor, hr is the
structure’s height above grade, and hx is the height above grade at which the
component is attached to the structure. According to 1997 UBC, the value of
Fp cannot be taken less than 0.7CaIpWp and need not be taken greater than
4.0CaIpWp.

Second Edition
In 1997 UBC, the value of Ca varies between 0 and 0.66 depending on the
location of the site and the site soil conditions. For seismic zone 4 (not in the
near field), the value of Ca is 0.44. Values of the term [1 + 3 hx/hr] can range
from 1 to 4. The value of the ratio of ap / Rp for most listed items varies
between 0.33 and 0.83. For mechanical and electrical equipment, the ratio of
ap /Rp is 0.33, except for unsnubbed vibration isolated equipment for which
the ratio is 1.67. The potential range in anchorage design forces (in high
seismic zones) is quite large, varying from as low as 0.3g to as high as 1.76g,
depending on component type and location within the structure.
For rigidly mounted mechanical and electrical equipment located on soil sites
in seismic zone 4, not in the near field and with Ip = 1.0, the value of Fp /Wp
ranges from 0.308 at grade and lower floors to 0.58 at the roof level. Using
the same seismic zone and Ip parameters, the value of Fp/Wp for vibration
isolated equipment ranges from 0.73 at grade to 1.76 at the roof. When
expansion anchors or shallow anchors (those with an embedment length to
diameter ratio less than 8) are required, anchor strengths are double these
values.
International Building Code and ASCE/SEI 7 after 2000
The 2000 through 2012 editions of the IBC, International Building Code,
(ICC, 2000, 2009, 2012), as well as 2002 through 2010 editions of
ASCE/SEI 7, Minimum Design Loads for Buildings and Other Structures,
(ASCE, 2002, 2005a, 2010) specify the Fp coefficient using the following
equation:
0.4 S DS I p a p
=Fp [1 + 2 z h]W p
Rp
where SDS is a short period ground motion coefficient including site effects, z
is the height above grade at the point of component attachment to the
structure, h is the height of the structure above grade, and other parameters
are as previously defined above. The value of Fp cannot be taken less than
0.3SDSIpWp and need not be taken greater than 1.6SDSIpWp. Forces determined
using this equation are considered to be strength level forces.
The value of SDS varies between 0 and 2.5 depending on site location and
class. The value of the [1 + 2 z/h] term can vary from 1 to 3. The value of
the ratio of ap /Rp for most listed items varies between 0.29 and 1.0. For most
rigidly mounted mechanical and electrical equipment, the ap /Rp ratio ranges
between values of 0.4 and 1.25. For vibration isolation mounted equipment,
required anchor strength is increased by a factor of two, if the gap between
the lateral restraint and equipment exceeds 1/4 inch. The potential range of

Second Edition
required anchorage strength (in high seismic zones) varies from as low as
30% to as high as 176% of equipment weight.
Second Edition Update The nonstructural provisions of ASCE/SEI 7-10 were modified in
Supplement No. 1 (ASCE, 2013) to include an overstrength factor applied to
The discussion has been
updated to consider anchors in concrete and masonry where the force applied to the anchor is not
ASCE/SEI 7-10 Supplement limited by a ductile, yielding element. This provision is intended to prevent
No. 1. brittle failure of the anchors.
7.3.1.2 Procedure
Volume 1, Section 3.8 presents a code-based limit state procedure for the
case where there is insufficient information available to directly calculate the
anchorage capacity assuming that the original design met the requirements of
the applicable building code.
Peak floor acceleration is the demand parameter used to evaluate anchorage

failure. Most nonstructural components can be treated as rigid boxes, and
therefore the equivalent static acceleration can be directly substituted for the
peak floor acceleration. This is the demand parameter specified in the
Performance Assessment Calculation Tool (PACT) provided in Volume 3 for
anchored nonstructural components.
For vibration isolated equipment, peak floor acceleration can be amplified

significantly. In anchorage design, this effect is accounted for by the
specification of higher Fp forces for use in design. To reflect this demand
amplification in the fragility calculation, it is recommended that the value of
θ be divided by 2.5 for isolated components to convert from spectral
acceleration to peak floor acceleration.
Where component anchorage was designed in accordance with a building

code that specifies strength-level forces (1997 UBC, or 2000 or later edition
of the IBC), the design resistance, φRn, of an anchorage system, expressed as
a peak floor acceleration, can be taken as the ratio Fp/Wp, where Fp and Wp
are as described in the previous section. If anchorage was designed under
codes that specified design forces in an allowable stress design format, the
value of φRn can be taken as 1.4 Fp /Wp.
Prior to the adoption of ASCE/SEI 7-10 including Supplement No. 1,

concrete anchorage was typically considered to have brittle failure modes
associated with breakout type failures. The capacity of components designed
after the adoption of Supplement No. 1 are more likely to governed by the
capacity of the component itself or its supports. In either case, the minimum
median failure capacity of a nonstructural component designed to an

Second Edition
equivalent static acceleration coefficient Fp /Wp is given by Volume 1,
Equation 3-2:
θ = Cq e( 2.81β )φ Rn
where:
Cq = 0.5
β = 0.5
θ = the median peak floor acceleration for anchorage failure
Yielding:
θ = 2.04φRn
To obtain θ, the user must identify the building code under which component
installation was designed and determine the equivalent static acceleration
coefficient for the component and the anchorage. This will require
knowledge of the building site (Z, Ca, SDS), location within the building where
the component is supported (hx and hr), and the appropriate Ip, ap, and Rp
coefficients.
The following examples illustrate application of this procedure, first for

rigidly mounted equipment, next for isolation-mounted equipment, and lastly
for out-of-plane anchorage for precast wall panel. The first two designs were
done with allowable stress forces.
Rigidly mounted equipment is located in the building’s basement (or first

floor equipment room) in conformance with the 1994 UBC in Seismic Zone
4.
According to 1994 UBC:
Z = 1.0
Ip = 1.0
Cp = 0.3
Nonstructural components located in basement levels are permitted to be

designed for 2/3 of the anchorage force:
Fp / 3 (1.0 )(1)( 0.3)W p 0.2W p
2=
Thus Fp / Wp = 0.2
Since the 1994 UBC specified design forces in allowable stress design
format,

Second Edition
φRn
= (1.4 ) F=
p / Wp 1.4
= ( 0.2 ) 0.28 g
θ = 2.04 x 0.28 = 0.49 g and
β is taken as 0.5.
Figure 7-4 plots this fragility function.
Figure 7-4 Fragility function for anchorage of rigidly mounted equipment

designed under the 1994 UBC.
Vibration isolated equipment is located in the building’s basement (or first

floor equipment room) in conformance with the 1994 UBC in Seismic Zone
4.
Z = 1.0
Ip = 1.0
Cp = 0.3
Nonstructural components with vibration isolation require a twofold increase

in the force to account for dynamic amplification:
Fp = (2)(1.0)(1.0)(0.3) = 0.6Wp in allowable stress design format.
Fp/Wp = 0.6 in allowable stress design format and thus, in strength format is
given φ Rn is given by:
 Fp 
=φRn 1.4
=  W p  1.4(0.60g) = 0.84g
 
Note however, that because this equipment is flexibly mounted, the value of
θ derived above should be divided by a factor of 2.5 to adjust for
amplification of peak floor acceleration (to convert from spectral acceleration
to acceleration). Thus:

Second Edition
=θ 2.04 ) 0.84
(= 0.68 g
2.5
Again, the dispersion is taken as β = 0.5. Figure 7-5 plots this fragility
function.
Figure 7-5 Fragility function for anchorage of vibration isolated equipment

designed under the 1994 UBC.
A precast wall panel was designed and installed in conformance with

ASCE/SEI 7-05. The wall panel is located between the roof and the floor
just below. From ASCE/SEI 7-05:
SDS = 1.0
Ip = 1.0
ap = 1.0
Rp = 1.25
z/h = 1.0
0.4 × 1.0 × 1.0 × 1.0 (1 + 2 × 1.0 )W p
Fp =
1.25
Fp Fp
0.96 <
= max =
1.6
Wp Wp
Since ASCE/SEI 7-05 specifies design forces in a strength format, φRn is

equal to Fp/Wp.
Fp
φR
=n = 0.96
Wp
=θ 2.04
= ( 0.96 ) 1.96g
The dispersion, β, is taken as 0.5.

Second Edition
7.3.2 Strength-Based Limit State Approach to Anchorage
Fragility Calculation
When details of a nonstructural component’s anchorage design are known,

the more detailed evaluation procedure of Volume 1, Section 3.8 can be used
to calculate the fragility parameters. Note that with the introduction of the
overstrength factor applied to anchors in concrete and masonry in ASCE/SEI
7-10 including Supplement No. 1, the strength capacity of the anchorage may
no longer govern the lateral capacity of the nonstructural component.
The strength capacity of an anchorage embedded in concrete is typically

determined using Appendix D of ACI 318-14, Building Code Requirements
for Structural Concrete and Commentary, (ACI, 2014). This value can be
adjusted using the procedures of Volume 1, Section 3.8 to obtain the median
capacity associated with a failure mechanism (damage state). The median
anchorage capacity can be expressed in terms of a floor acceleration, which
serves as the fragility demand parameter.
This process requires the following steps:
Step 1. Determine the strength capacity of the anchorage detail for each
failure mode using the procedure of Volume 1, Section 3.8 and ACI 318,
Appendix D. Adjust the strength capacity to correspond to the median
failure capacity.
Step 2. Determine the smallest median capacity of the anchorage using

Volume 1, Section 3.8.
Step 3. Determine the forces on each individual anchorage for a given

horizontal nonstructural component force, Fp, using the connection geometry.
Step 4. Solve for Fp that corresponds to the smallest anchorage median

failure capacity.
Step 5. Divide Fp by the weight of the component, Wp, to determine the

corresponding median floor acceleration. Assign a dispersion value
according to Volume 1, Section 3.8.
The procedure is illustrated with an example below.
Assumptions:
• A rigidly mounted equipment item is supported on a foundation pad at
the building’s first floor level.
• The equipment is tall and penetrates the first floor, where is it attached
for seismic and gravity forces.

Second Edition
• The equipment’s center of gravity approximately aligns with the first floor slab,
such that the anchor bolts at the first floor are in shear only.
• There are a total of 4 ¾-inch diameter, da, anchor bolts, using A36 threaded rod
embedded 12 inches, le, into the slab with an 8 inch edge distance, ca1. Concrete
strength is f’c = 3,000psi. The mounting pad ha a depth of 24 inches. No
supplementary reinforcing is present around the anchor bolts.
• The equipment weighs 50 kips.
Step 1. Determine the strength and median failure capacity of each anchor.
Based on calculations according to ACI 318-11, Appendix D, the anchorage capacity

is found to be controlled by the yielding of the steel bolt in shear with:
2
π 0.9743 
φ Rn =
φ Ase,v futa 0.65 ( 0.6 )  da −
=  ( 58 ) =
7.6 kip
4 nt 
Total dispersion is calculated using values given in Table 7-5 and Volume 1, Equation
3-3:
β= β D 2 + β M 2 + βC 2
Table 7-5 Dispersion Values for Failure Mode 1

Dispersion Discussion Value
Design (βD) Ductile behavior, well represented by design formula 0.05
Material (βM) A36 steel, from Volume 1, Table 3-4 0.15
Quality (βC) Low sensitivity to construction quality, from Volume 1, Table 3-5 0
β
= ( 0.05)2 + ( 0.15)2 + ( 0=
)2 0.16
Step 2. Determine the smallest median failure capacity of the anchorage.
Based on calculation, it is found that the failure modes of concrete breakout and
concrete pryout have higher median strengths and thus are less likely to occur than the
bolt shear mode.
Thus, the governing median failure (shear) capacity of an individual anchor is
=θ C=
qe
( 2.054 β )
φ Rn (1.0 ) e=
0.328
7.6 kip 10.6 kip
Step 3. Determine the forces on an individual anchor. For the given component
configuration, the anchorage on an individual anchor is Fp / 4.
Step 4. Solve for the Fp that results in an anchorage force equal to the median fragility
anchorage failure force.

Second Edition
Fp / 4 = 10.6 kip
Fp = 42.4 kip
Step 5. Solve for the median failure acceleration

Fp / W p 42.4
= = / 50 0.85 g
The dispersion, β, is taken as 0.30.
7.4 Displacement-Based Limit State Approach to Define

Calculated Fragilities
Nonstructural components for which displacement based fragilities can be

developed include exterior wall systems, such as precast cladding designed
with slip joints of defined size and slide plate type seats for structural
members crossing expansion joints.
Exterior wall systems are generally constructed with a gap to accommodate

horizontal drift. Using geometric relationships, the size of the gap can be
expressed as a story drift ratio that the wall panel system can accommodate
before contact with adjacent panels occurs (usually at corners and column
covers). Proprietary panel systems feature unique characteristics but
suppliers can generally provide the drift or accommodation gap that inhibits
adjacent panel contact.
The design story drift for the structure should be used for estimating the drift
accommodation ratios of precast cladding systems. If this unknown, Table
7-6 provides guidance regarding the code minimum story drift ratio (often
used by cladding system suppliers as the basis for the design) required by
recent building codes as a function of the Occupancy Category referenced by
the current code. Alternatively, the maximum story drift ratio for the
building lateral system specified in the edition of ASCE/SEI 7 enforced at
the time the structure was designed may be used.
Table 7-6 Suggested Drift Accommodation Ratios for Precast Cladding and
Other Brittle Cladding Systems
Building Occupancy Category According to ASCE/SEI 7)
Building Code II III IV
1994 UBC or earlier 0.012 0.012 0.012
1997 UBC 0.020 0.020 0.020
2000 IBC – 2009 IBC 0.020 0.015 0.010
2012 IBC 0.020 0.019 0.015

Second Edition
The drift ratio at which damage occurs can be taken as the drift
accommodation ratio shown in the table plus an additional value (0.005 is
recommended). Note that in some cases, cladding design can accommodate
greater drift, either because of intentional design conservatism or because use
of convenient dimensions, such as 1 inch gaps, resulted in more tolerance
than required by the applicable code. If drawings depicting the gap system
are available, users should determine if actual drift tolerance can be
determined from these drawings. A dispersion of 0.5 is recommended for
story drift related fragilities.
The example below illustrates the determination of the story drift fragility of
a precast concrete exterior wall panel.
Assume a precast concrete exterior wall panel was installed in conformance

with ASCE/SEI 7-05, SDS = 1.0, Ip = 1.0.
From Table 7-6, the assumed accommodation drift ratio is 0.020
Therefore:
θ = 0.020 + 0.005 = 0.25 with β = 0.5.

Second Edition
Chapter 8
Consequence Function
Development
8.1 Introduction
Consequence functions are used to convert component and building damage

states into performance measures including repair costs, repair times, unsafe
placarding probabilities, and casualties. As with each of the other
performance assessment steps, uncertainties arise in the translation of
damage states to consequences. Consequence functions reflect this
uncertainty in the form of probability distributions of consequences for each
damage state.
Volume 1, Section 3.9 provides guidelines for developing consequence

functions. This chapter illustrates the consequence function development
process and presents procedures that can be used to modify provided
consequence functions to reflect the unique circumstances users may
encounter in individual building assessments.
8.2 General Considerations
Repair cost and repair time consequence functions provided with the
Performance Assessment Calculation Tool (PACT) provided in Volume 3
have been developed by professional cost estimators using component
damage state repair descriptions for a reference location in the United States
(Northern California) at a reference time (2011), neglecting uncertainty in
contractor pricing strategies or construction cost escalation. Repair
descriptions were developed in collaboration with engineering consultants
familiar with repair techniques and cost estimators familiar with collateral
cost and work. Repair cost and repair time estimates were developed to
include all collateral work, such as finish removal and replacement,
scaffolding, and debris removal, associated with each component damage
state repair. Consideration is given to cost and time efficiencies that are
gained by multiple component repairs occurring in a given realization.
General contractor mark-up is included in the cost estimates. The repair time
estimates do not include the time needed to design, permit, and fund the Second Edition Update
repair project, the time required to procure long lead time components that The definition of repair time
need replacement, or the time to mobilize the contractors and close the has been clarified.
FEMA P-58-2 8: Consequence Function Development 8-1

Second Edition
project. The environmental impact estimates were generated directly from
the repair cost estimates in collaboration between construction cost
estimators and environmental life cycle assessment researchers.
For each damage state, a determination is made as to whether loss of life or

serious injuries are possible as a result of the damage. For damage states
deemed to present a potential for loss of life or serious injuries, an affected
area is estimated. The affected area is that occupied floor space adjacent to
the damaged component in which there would be risk of injury. The
likelihood of loss of life or serious injuries in the impacted area is estimated
based on the component’s weight and location and judgment.
The likelihood of unsafe placard posting is determined using engineering

judgment. For most nonstructural components, an unsafe placard posting is
judged unlikely as a consequence of their damaged condition. A few notable
exceptions include fire sprinklers (since their failure affects the post-
earthquake fire life-safety of the building) and suspended ceiling systems
(since their failure can create the impression of significant damage that may
cause a post-earthquake inspector to post the building as unsafe). On the
other hand, most structural components are judged to have the potential for
triggering an unsafe placard posting since damage to these components can
result in a reduction of the gravity or lateral-load carrying capacity of the
building.
8.3 Calculation of Consequence Functions

8.3.1 Repair Cost
A detailed description of repair scope is needed to enable estimation of repair

labor and materials. The repair description must include all work associated
with repair of the actual damage as well as collateral work associated with
removal and replacement of finishes, relocation of utilities, provision of
scaffolding, and similar actions.
The following steps outline the process for converting a component damage
state into a repair cost consequence.
8.3.1.1 Establish Required Repair Actions
Where multiple repair options exist, users must choose which strategy to use,
or alternatively, select multiple strategies and develop consequences for each.
In the latter case, the uncertainty of each method will be calculated by taking
a weighted-average value of the costs associated with each as the median
estimate of repair cost. Accordingly, in order to cover the range of potential
repair options and cost, dispersions should be selected sufficiently large.
8-2 8: Consequence Function Development FEMA P-58-2

Second Edition
Typically, only one repair strategy will be selected, being the lowest-cost
acceptable solution.
8.3.1.2 Establish Repair Boundaries and Interfaces
The repair scope should include all necessary access work, including removal
and replacement of finishes, mechanical/electrical work or scaffolding.
However, a considerable amount of overlap and double-counting of access
work is possible in estimating repair work on a component basis. For
example, repair of a structural element may require removal of
mechanical/electrical system components that prevent access to the damaged
structure. However, the mechanical/electrical systems may be directly
damaged resulting in double-counting of the repair of these components.
For more severe damage states, it may be advisable to exclude removal and
replacement of mechanical/electrical systems associated with a structural
repair, assuming that as more severe damage occurs, these systems will
become damaged and require repair by themselves. Users should consider
the likelihood that these systems will be directly damaged and if so, exclude
them from the user-defined consequence function being developed.
8.3.1.3 Consider Associated Work Items
The following items should generally be included in damage repair

estimates:
• Temporary relocation and protection of building contents
• Removal and reinstallation of covering elements
• Cutting in, patching, and repairing of adjacent building elements
• Clean-up
• Staging and access
• Costs for these associated work items should include subcontractor level
mark-up, including overhead and profit. General contractor mark-up
should not be included since this is separately addressed.
8.3.1.4 Estimate the Repair Cost
Using an appropriate estimating methodology, develop an estimate for the

repair cost. Since most repairs are not standard work found in cost
estimating guides, it is usually necessary to formulate an estimate using local
expertise and judgment regarding the labor hours and materials needed to
execute the specific work required. Allowances should be included for scope
not specifically identified.
FEMA P-58-2 8: Development of User Defined Consequence Functions 8-3

Second Edition
8.3.1.5 Establish Repair Cost Ranges
PACT is structured to use probabilistically defined cost ranges. Input of

values for the 10th, 50th, and 90th percentile repair costs capture this
uncertainty. Median values and either the coefficient of variation (COV) or
dispersion values calculated from the cost data, based on the selection of the
distribution (normal or lognormal) are also input.
Best estimate, lower bound, and upper bound costs should be identified and
then a lognormal or normal distribution should be fit to these bounds. The
selection of the distribution is based on judgment.
In addition to identifying best estimate, lower, and upper bound estimates of

repair costs, an adjustment based on economies of scale is also applied.
Costs per unit of repair at lower bound quantities and upper bound quantities
are indicated in the consequence function. For example, if fewer than three
components are damaged in a realization, the unit repair cost has a median
value of $15,000 while if more than ten components require repair in a
realization, the unit median cost decreases to $8,000 per item. With this
information, the user can enter a lower bound of three, with an associated
median cost of $15,000, and an upper bound of 10, with an associated
median cost of $8,000. For quantities intermediate to the upper and lower
bounds, PACT applies linear interpolation.
8.3.1.6 Assign Dispersion
Dispersion values are based on the uncertainty in values of unit cost and are
assumed to be constant regardless of the quantity of repairs that must be
conducted.
8.3.2 Repair Time
Similar to estimating repair cost, a detailed description of repair scope is

needed to enable estimation of repair time. The repair description must
include all work associated with repair of the actual damage, as well as
collateral work associated with removal and replacement of finishes,
relocation of utilities, provision of scaffolding, and similar actions. The
repair time estimates should not include the time needed to design, permit,
and fund the repair project, long lead times, or the time to mobilize the
contractors and close the project.
The following steps outline the process for converting the occurrence of a
component damage state into a repair time consequence.

Second Edition
8.3.2.1 Establish Baseline Repair Time Estimate
The baseline repair time estimate is based on the total number of trade hours
required to perform the work and can be generated directly from the total
baseline repair cost or from the detailed line items of the cost estimate.
For repair work that is highly labor intensive, the most accurate approach is
to estimate the labor hours for each detailed line item and sum them. This
also provides a validation check on the cost estimate by ensuring that all
complexity of labor has been considered.
A reasonable approximation can be obtained, however, by simply applying a

labor factor to the overall baseline repair cost estimate, and dividing the
result by an effective hourly rate for the likely labor mix. For example, if the
baseline cost is $10,000, and labor is assumed to make up 80% of the total
cost, the labor cost is $8,000. If the labor rate (fully burdened) is assumed to
be $80/hr, that represents 100 hours of work for repair.
8.3.2.2 Establish Repair Time Estimate Ranges
Variability in repair time estimates can be directly assigned as equal to the

repair cost estimate variability. Users can also independently develop an
estimate of variability using the detailed line item elements of the cost
estimate and judgmental considerations on the difficulty of performing the
work item. An important consideration with regard to repair time variability
is that PACT computes the number of days of repair time by dividing the
total labor hours determined for each realization by the number of workers
that can be accommodated in the building during repair. This latter quantity
justifies higher dispersions for repair time than are used for cost.
8.3.3 Unsafe Placards
Use of judgment is necessary to assess if the damage state is likely to result

in unsafe placard posting. Consideration must be given to the appearance of
damage, and the effect it may have on post-earthquake inspectors, as well as
the actual effect the damage will have on the building’s post-earthquake
safety. Damage that significantly increases the probability of building
collapse in aftershocks should trigger unsafe placarding, whether or not the
damage is directly observable to post-earthquake inspectors. Similarly,
damage that could appear substantive to post-earthquake inspectors should be
assumed to be capable of triggering unsafe placarding, even though the
actual reduction in building safety may be small.
state into an unsafe placard consequence.

Second Edition
8.3.3.1 Determine Likelihood of Unsafe Placard
Unsafe placards are posted on buildings by inspectors performing rapid

assessments of many buildings after an earthquake. These inspectors must
apply individual judgment as to whether a building’s safety has been
impaired. These judgments are imperfect. If significant structural damage,
such as shear failure of concrete columns is evident, inspectors are likely to
post the building as unsafe. However, in many types of buildings, significant
structural damage is not readily observable without removal of finishes.
Therefore, when performing post-earthquake assessments, inspectors often
rely on the extent of damage to nonstructural elements as an indicator of
structural damage that is not itself readily visible. This methodology assigns
an unsafe placard under two conditions:
• Actual damage sustained by the structure is such that a life-threatening
condition actually exists either through impairment of required egress or
fire protection systems, or through substantive reduction in structural
resistance; or
• A preponderance of nonstructural damage such that there is the
appearance that substantive structural damage has occurred whether or
not this is actually the case.
For each damage state, users will need to assess whether an unsafe placard
posting can occur as a consequence of the damage. For most nonstructural
components, an unsafe placard posting is not likely to occur unless it affects
the life safety function of the building or is present in such quantity that it
can create a visual appearance that is suggestive of unobservable structural
damage. For structural components, the more severe damage states are more
likely to result in an unsafe placard posting since the resulting damage can
affect the overall safety of the building.
8.3.3.2 Establish the Median Fraction of Total Quantity that will
Trigger an Unsafe Placard
For each component damage state, a median fraction is estimated that reflects
the number of damaged components out of the total number of similar
components at a given story or floor that would result in an unsafe placard
posting. Generally, the fraction of the total quantity that will result in unsafe
placarding reduces with the severity of the damage state given that it takes
fewer occurrences to affect either the actual or apparent effect on building
safety. For example, special concentrically braced frames with wide flange
braces are assigned the following fractions by damage state:
• DS1: Not applicable, unsafe placard posting not probable

Second Edition
• DS2: 60% of total quantity will result in unsafe placard posting
For the most severe structural component damage states, the percentage
could be as low as 1%, when the damage state results in an obvious reduction
in gravity load or seismic load carrying capacity.
A dispersion value of 0.5 is recommended unless more detailed information

is available.
8.3.4 Casualties
Judgment is required to assess if the damage state is likely to result in loss of

life or serious injury. An impacted area and rate of severity must be defined
for estimating casualty and serious injury.
state into casualty consequences.
8.3.4.1 Determine Potential for Non-Collapse Casualties
For each damage state, the possibility of casualties considering the weight
and location of the component should be determined. While the failure of
most components will likely not result in loss of life, serious injury may still
be possible.
8.3.4.2 Determine Affected Occupancy Areas
For those components with damage states identified to result in casualties, an

affected area is estimated. The affected area is that occupied floor space
adjacent to the damaged component in which there would be risk of injury or
casualty. Population models are used to establish the number of persons
likely to be present within the affected area.
For example, the occupied affected area for ceiling systems is estimated as
the floor area directly underneath the ceiling. For HVAC ducting, the area is
estimated as 15 square feet for ducts with less than 6 square feet of cross-
sectional area, assuming a 5 feet length of 3 foot wide ducting falling to the
floor below.
8.3.4.3 Determine Rate of Loss of Life in Affected Area
The likelihood of loss of life occurring from the component damage state is
judged based on the component weight and location. For the vast majority of

Second Edition
components, such as HVAC ducting, the loss of life rate is taken as zero;
however, other provided fragility specifications, such as “heavy” wall
partitions and in-line fans, include non-zero values.
8.3.4.4 Determine Serious Injury Rate in Affected Area
Serious injury is defined as injury requiring hospitalization. Component

damage states are more likely to result in serious injuries than loss of life.
The serious injury rate represents the median percentage of people in the
affected area that will be seriously injured. Generally, this percentage will be
small since people have the opportunity to react to collapsing components
and get out of the affected area and also because components may either fall
towards or away from a person. Minor injuries, including scrapes, bruises,
and even some lacerations not requiring hospitalization are not considered.
For example, the serious injury rate is taken as 5% for HVAC ducting.
A dispersion value of 0.5 should be used unless more detailed information is

available.
8.3.5 Environmental Impact
The environmental impact estimates are directly derived from the

Second Edition Update construction cost estimate. Environmental impacts are estimated using the
Calculation of environmental Economic Input-Output Life Cycle Assessment (EIO LCA) data (Carnegie
impact estimates have been Mellon University Green Design Institute, 2008) that correlate environmental
added.
impacts to the number of dollars spent in each industrial sector. The
industrial sectors are defined by the North American Industry Classification
System (NAICS). Examples of industrial sectors include ready-mix concrete
manufacturing, adhesive manufacturing, and iron, steel pipe and tube
manufacturing from purchased steel.
Two metrics are tracked in the environmental impact assessment: greenhouse

gas emissions (measured in kg CO2e) and energy consumption (measured in
MJ). The scope of material impacts is limited to the ‘cradle-to-gate’ impacts
of manufacturing, and does not include on-site construction impacts.
Routine maintenance and disposal of materials at end-of-life are not
considered. No environmental impact is assigned to labor, as is common in
industry Life Cycle Assessments (LCA), and the impact of using temporary
powered equipment is also excluded.
The following steps outline the process for converting a fragility repair cost
consequence into environmental consequence.

Second Edition
8.3.5.1 Obtain the Repair Cost Estimate
See Section 8.3.1 for information about developing the repair cost estimate.
The more detailed the cost estimate, the more detailed the LCA evaluation
can be.
8.3.5.2 Allocate Costs to Industrial Sectors
The environmental impacts of the fragilities are based on the median value of
the construction cost estimate. In the detailed cost estimate, the cost of each
repair activity must be disaggregated and allocated into the following broad
categories: labor, energy, rental equipment, and industrial sectors. The
appropriate industrial sectors should be selected from the EIO LCA database
for each work activity in the cost estimate. Finally, the cost in each
applicable sector is summed. As previously stated, the scope of this study
assumes zero environmental impact for labor, energy, and rental equipment,
but allocating costs to these categories may assist in developing the overall
allocations.
Cost allocation will require some judgement on how the costs should be
distributed. In this assessment, the costs were allocated using professional
judgement by a construction cost estimator.
8.3.5.3 Obtain Environmental Impact Factors
The EIO LCA database should be queried to obtain the environmental impact
factors for the industrial sectors. Ensure that the 2002 Purchaser Price Model
is selected when extracting the TRACI Impact Assessment and Energy
factors. Before applying the environmental impact factors to the repair costs,
the factors must be adjusted to reflect per dollar costs ($1 instead of $1
million) and inflation. The overall embodied environmental impacts can be
calculated by multiplying environmental factors with the allocated repair
costs determined in the previous step. These calculations can be
consolidated by using matrix multiplication.
Regarding inflation: At the time of this writing, the reference year for the
PACT data is 2011, while the reference year for the EIO-LCA data is 2002.
The resulting inflation factor going from 2002 to 2011 was determined to be
1.45. Since the dollar units are in the denominator, the environmental
impacts / $ spent are divided by 1.45 to convert 2002 to values to 2011
values. From there, the same process can be repeated to convert 2011 values
to present values.

Second Edition
8.3.5.4 Calculate Total Environmental Impacts
The total environmental impacts are calculated by multiplying the sector

costs with the adjusted environmental impact factors. If per dollar values are
desired, the total environmental impact can be divided by the total cost of the
fragility or project. Multiplication by matrices is recommended for this step.
A dispersion value of 0.5 is recommended unless more detailed information

is available.
8.4 Example Application
This section illustrates the development of the consequence function for a

low aspect ratio reinforced concrete shear wall of 8 inches or less thickness
with a single curtain of reinforcement in damage state 2. The fragility and
consequence functions for this component (B1044.001) are provided in the
Fragility Database in Volume 3 which is referenced by PACT.
8.4.1 Estimation of Repair Cost
Damage state descriptions for this component were established through

testing and include shear cracking, concrete spalling, jamb crushing, rebar
buckling, and rebar fracture.
8.4.1.1 Establish the Required Repair Actions
Repair strategies are developed at a level of detail sufficient to approximate

likely costs. Consideration is given to the length of cracks requiring epoxy
grouting, extent of reinforcing requiring removal, and the process for splicing
in place of new reinforcing with existing. For provided fragility groups, this
information is accessible via the PACT Fragility Specification Manager
utility under each damage state description.
The damage is not uniformly distributed across the wall. The majority of
significant damage occurs at wall ends. The basis of measurement selected is
a 25 foot long by 15 foot high wall section, judged to represent a reasonable
panel size and story height. The area of this wall, 375 square feet, is used for
repair cost estimations.
In addition to the unit of measure, it is necessary to identify the wall

thickness and reinforcing steel density. If the consequence function is to
include a range of these parameters, it may be necessary to include multiple
fragility specifications and associated consequence functions to represent the
different conditions.

Second Edition
8.4.1.2 Establish Repair Boundaries and Interfaces
In this example, the estimated repair costs do not include removal and
replacement of any architectural finish.
8.4.1.3 Estimate the Repair Cost including Associated Work
Items
Table 8-1 presents the median repair cost estimate for a B1044.001 wall in
damage state 2.
Table 8-1 Repair Cost Estimate for Wall

Wall repairs Quantity Unit $/ unit Total ($)
Grout injection at wall cracks (330 LF of crack
1,238 LF 8.00 9,904
per 100 SF of wall)
Remove portion of concrete wall for access to
56 SF 75 4,200
reinforcing steel (15 SF per 100 SF of wall)
Remove portion of buckled #8 reinforcing
steel12" long, replace reinforcing steel with
19 EA 200.00 3,800
mechanical splices each end (5 locations per
100 SF of wall)
Install eight #4 ties at end of wall, re-bend
2 EA 1,000.00 2,000
horizontal reinforcing steel bars around new ties
Cast 5000 PSI concrete at wall infill including
56 SF 75.00 4,200
formwork
The information in Table 8-1 is extracted from in the Consequence

Estimation Tool – Structural in Volume 3 in the Standard Wall to 8” tab.
In addition to the direct repair work, the repair work for a reinforced concrete
shear wall may include the following elements:
• Office furniture relocation within the building and their protection
• Floor protection consisting of plywood and a membrane for a distance of
10’ either side of the wall
• Rolling scaffold access to the work area
• Shoring of the structure as needed during the repair
• An allowance for removal and replacement of mechanical and electrical
items such as power outlets, switches, and data outlets
• The cost of clean-up and removal of temporary protection
8.4.1.4 Establish Repair Cost Ranges
Repair cost ranges are established for repair items based on a three point
estimating approach, using high, low, and best estimate values. For example,
furniture relocation is estimated at $1,000 for the best estimate value, with a

Second Edition
range from $500 to $3,000. Installation of reinforcing at the ends of a wall is
estimated at $2,000 for the median value, with a range from $0, representing
no boundary damage, to $4,000, representing extensive boundary damage.
The dispersion, β, value is set based on review of the resulting three point
estimate. For user-generated fragility specifications, it is possible to use an
overall β estimate, without first calculating the three point estimate,
depending on user comfort with range estimates.
In this example, baseline estimate pricing and efficiencies assumed four to

ten walls being repaired. Fewer than eight walls would result in lower
efficiency and increased waste due to short orders. More than twenty six
walls would lead to increases in efficiency of operations. The costs for wall
counts below four increase by 20%, and the costs for walls counts above
twenty six are decreased by 20%.
8.4.2 Estimation of Repair Time
For the above example, labor hours are estimated by line item, resulting in a
median estimate of 15.8 days for the repair.
8.4.3 Estimation of Environmental Impacts
In this example, global warming potential and embodied energy are

estimated by breaking down the repair cost estimate and multiplying the
costs with environmental impact factors from the United States
Environmentally Extended Input-Output (USEEIO) database (available at:
https://catalog.data.gov/dataset/useeio-v1-1-matrices, last accessed
September 28, 2018).
8.4.3.1 Obtain the Repair Cost Estimate
Refer to Table 8-1 for the median repair cost estimate for a B1044.001 wall
in damage state 2.
8.4.3.2 Allocate Costs to Industrial Sectors
The industrial sectors were selected from the USEEIO database. The
percentage cost allocations of the work activities are shown in Table 8-2.
The result total cost allocations are shown in Table 8-3.
8.4.3.3 Obtain Environmental Impact Factors
The environmental impact factors extracted from the USEEIO database for
the industrial sectors in this example are shown in Table 8-4. Note that in the
database, global warming potential is labeled “impact potential/gcc/kg co2
eq” and energy is labeled “resource use/enrg/mj.”

Second Edition
Table 8-2 Percentage Cost Allocations for Wall
Industrial Sectors
% Rental Tools/
% Plastic film
% Electrical
% Adhesive
% Concrete
% Cleaning
% Plywood
Fabricated
No impact
% Energy
% Steel –
% Labor
Work Activity
Remove, store and reinstall office
94 1 0 0 5 0 0 0 0 0
furniture and equipment
Temp- Floor protection 94 1 0 0 0 0 0 5 0 0

orary Scaffolding or work platform 74 1 25 0 0 0 0 0 0 0
Grout injection at wall
59 1 0 40 0 0 0 0 0 0
cracks…
Remove portion of concrete… 79 1 10 0 10 0 0 0 0 0
Wall
Remove portion of buckled
repair 59 1 0 0 0 0 0 0 0 40
#8…
Install eight #4 ties… 59 1 0 5 0 0 0 0 0 35
Cast 5000 PSI concrete… 49 1 10 0 0 35 0 0 5 0
Mechanical and electrical modifications
74 1 0 0 0 0 25 0 0 0
or as required for repair work
Table 8-3 Resulting Total Cost Allocations for Wall

Industrial Sectors
Rental Tools/ No
Plastic film ($)
Fabricated ($)
Total Cost ($)
Adhesive ($)
Concrete ($)
Electrical ($)
Cleaning ($)
Plywood ($)
Impact ($)
Energy ($)
Labor ($)
Steel –
31854 20876 319 1403 4062 470 1470 750 75 210 2220
Table 8-4 Environmental Impact Factors from USEEIO

Database for Select Industrial Sectors
Global Warming
Potential Embodied Energy
Industrial Sector (kg CO2e / $) (MJ / $)
Adhesive 0.89 21.09
Cleaning 0.17 4.94
Concrete 2.24 16.55
Electrical 0.25 4.73
Plastic film 0.74 18.59
Plywood 0.59 12.19
Steel - fabricated 0.85 26.02

Second Edition
The factors were then adjusted for inflation, converting from 2013 (USEEIO
reference year) to 2011 (PACT reference year), by dividing by the ratio of
gross domestic product in the construction sector in 2013 to 2011 from U.S.
Bureau of Economic Analysis (available at https://www.bea.gov/, last
accessed September 28, 2018). The inflation-adjusted gross outputs for the
construction sector were 1139.7 in year 2013 and 1008.60 in year 2011, and
thus the inflation or “deflation” factor going from year 2013 to 2011 was
calculated as 1008.60/1139.7 = 0.88. The inflation-adjusted environmental
impacts are shown in Table 8-5. Also, ‘global warming potential’ was
renamed to ‘embodied carbon.’
Table 8-5 Adjusted Environmental Impact Factors

Embodied Carbon Embodied Energy
Industrial Sector (kg CO2e/$) (MJ/$)
Adhesive 1.01 23.84
Cleaning 0.19 5.58
Concrete 2.54 18.70
Electrical 0.28 5.35
Plastic film 0.84 21.00
Plywood 0.66 13.77
Steel - fabricated 0.96 29.40
8.4.3.4 Calculate Total Environmental Impacts
The calculation for total environmental impact are shown below using matrix
multiplication. Essentially, Table 8-3 is multiplied with Table 8-5.
[Sector costs] × [environmental impact factors / $ in sector] = [overall
environmental impacts]
= [20,876 319 1,403 4,062 470 1,470 750 75 210 2,220]
1.01 23.84 
0.19 5.58 

 2.54 18.70 
 
× 0.28 5.35 
0.84 21.00 
 
0.66 13.77 
0.96 29.40 

= [9,785 178,184]
Thus, the total environmental impacts for this fragility, an 8” thick

standard concrete wall with dimensions of 25’ × 15’ in DS2 are:

Second Edition
Embodied carbon = 9,785 kg CO2e
Embodied energy = 178,184 MJ.
8.5 Consequence Functions in PACT
8.5.1 Provided Consequence Functions in PACT
Consequence functions have been developed for fragility groups that are
provided in the Fragility Database which is referenced by PACT. The Cost
Summary tab of the Fragility Database contains the data developed in both
of the Consequence Estimation Tools. In addition to the calculated low (10th
percentile, P10), best estimate (50th percentile, P50), and high (90th
percentile, P90) values of repair cost and repair time, the Cost Summary tab
also includes information regarding the expected distribution of the
consequence functions (normal or lognormal), relevant COV and dispersion
values, as well as the lower and upper quantity values. Note that the
Fragility Database provides the repair cost and repair time information per
fragility unit of measure. Thus, best estimate repair cost of a reinforced
concrete shear wall (B1044.001) is displayed as $36 × 144 square feet =
$5,180 in the Fragility Database and PACT. All of the above information is
also presented in the Fragility Specification.
8.5.2 User-Defined Consequence Functions
User-defined consequence functions are input, or existing ones are modified

using the PACT Fragility Specification Manager utility. This utility can be
accessed via the PACT Control Panel. The following sections describe the
steps for adding new or modifying existing fragility specifications.
Consequence functions should be provided for each fragility specification

included in the fragility database. For example, to add a new, low aspect
ratio concrete shear wall fragility specification, the + button on the Fragility
Specification Details tab of the Fragility Specification Manager is
selected, as shown in Figure 8-1:
This will bring up the Add New Fragility window as shown in Figure 8-2.
An appropriate fragility specification number per the modified NISTR
system is entered and clicking OK to bring up additional windows for entry
of data. The resulting, blank fragility specification is ready to be completed
with the associated fragility and consequence function information.
Repair cost consequence information is input at the Consequence Functions

tab as indicated in Figure 8-3. The unit of measure is dollars.

Second Edition
Repair cost consequence information is requested for both the lower quantity
and upper quantity values to incorporate economies of scale. The resulting
repair cost curve is automatically generated as shown in Figure 8-3. Figure
8-3 illustrates an example, where the repair cost for less than 5 units is
$5,000 and $3,000 for more than 15 units.
Users must indicate the Curve Type (either normal or lognormal) in the
dropdown menu in Figure 8-3. In this example, the repair cost curve
assigned is “normal” and the associated coefficient of variation (COV) is
input into the cell. If the repair cost curve is lognormal, the associated
dispersion (standard deviation of the logs of the values or β) value will be
requested.
Figure 8-1 Initiating a new fragility specification in Fragility Specification

Manager.
Figure 8-2 Add New Fragility

window.

Second Edition
Figure 8-3 Consequence Function input window.
Figure 8-4 PACT graph showing upper and lower bound

repair cost data.
Repair time consequence information is input in a similar manner to repair

costs. The unit of measure is days. The economy of scale treatment is
similar for both lower and upper quantities.
If a component damage state is likely to result in an unsafe placard posting,

the box Possible Unsafe Placard Consequences, shown in Figure 8-5 in the
Other Consequences tab, is checked.

Second Edition
The associated median fraction and dispersion values are input in the two
boxes below the flag.
When it is judged likely that the damage state will result in either fatality or
serious injury, the Potential for Non-Collapse Casualties box should be
checked, as shown in Figure 8-6.
Information regarding Occupied Area Affected Per Performance Group

Unit, along with the median and dispersion values can be input here.
If the repair time of a component is expected to be affected by long lead

times for procurement, such as HVAC equipment and elevators, then the Use
Long Lead Time Flag box should be checked. Note that the actual lead
time is not included in the repair time estimate.
The Environmental Losses produced by PACT are also shown in this tab.
If needed, the values for environmental impacts can be changed here.
Figure 8-5 Other Consequences tab, Unsafe Placards input.

Second Edition
Figure 8-6 Other Consequences tab, Casualty input.
8.5.3 Other Considerations
User-defined consequence functions are not adjusted by time or location

factors built into PACT. Thus, if the user-defined function is to be used at a
later time or a different location, it must be adjusted by the user to reflect
these changes.

Second Edition
Appendix A
Structural Component Fragility
Specifications
Table A-1 Structural Steel Elements (B103)
Construction Damage Mode, Other/

Design Design Configuration; Comments/
Material/System Characteristics Other Features Notes Fragility Classification Number
Gravity beam
Bolted shear tab - B1031.001
connection
Miscellaneous
Structural Steel Column base B1031.011a, B1031.011b,
- -
Components/ plate B1031.011c
Connections
B1031.021a, B1031.021b
Column splice Welded -
B1031.021c
B1033.001a, B1033.001b,
Chevron brace
B1033.001c
Wide flange braces B1033.002a, B1033.002b,
Single diagonal brace
(balanced design) B1033.002c
B1033.003a, B1033.003b,
X brace
B1033.003c
B1033.011a, B1033.011b,
Chevron brace
B1033.011c
Hollow structural
B1033.012a, B1033.012b,
section (HSS) braces Single diagonal brace
B1033.012c
(balanced design)
B1033.013a, B1033.013b,
X brace
B1033.013c
Special
B1033.021a, B1033.021b,
Concentrically Chevron brace
B1033.021c
Braced Frame
Structural Steel Hollow structural
(CBF): balanced B1033.022a, B1033.022b,
Braced Frame section (HSS) braces Single diagonal brace
design criteria, B1033.022c
(tapered gussets)
Tapered gussets,
B1033.023a, B1033.022b,
AISC minimum X brace
B1033.023c
standard
B1033.031a,
Chevron brace
B1033.031bB1033.031c
Hollow structural
B1033.032a, B1033.032b,
section (HSS) (AISC Single diagonal brace
B1033.032c
minimum standard)
B1033.033a, B1033.032b,
X brace
B1033.033c
B1033.041a, B1033.041b,
Chevron brace
B1033.041c
Double angle braces
B1033.042a, B1033.042b,
(AISC minimum Single diagonal brace
B1033.042c
standard)
B1033.043a, B1033.043b,
X brace
B1033.043c
FEMA P-58-2 A: Structural Component Fragility Specifications A-1

Second Edition
Table A-1 Structural Steel Elements (B103) (continued)
Construction Damage Mode, Other/

Design Design Configuration; Comments/
Material/System Characteristics Other Features Notes Fragility Classification Number
B1033.051a, B1033.051b,
Chevron brace
B1033.051c
Compact brace shape
B1033.052a, B1033.052b,
(wide flange, HSS, Single diagonal brace
B1033.052c
double angles)
B1033.052a, B1033.053b,
X brace
B1033.053c
B1033.061a, B1033.061b,
Chevron brace
B1033.061c
Ordinary
Concentrically Brace at ductile
B1033.062a, B1033.062b,
Braced Frame slenderness limit (wide Single diagonal brace
B1033.062c
flange, HSS, angle)
Structural Steel B1033.063a, B1033.063b,
X brace
Braced Frame B1033.063c
(continued)
B1033.071a, B1033.071b,
Chevron brace
B1033.071c
Basic design for
factored loads only B1033.072a, B1033.072b,
Single diagonal brace
(wide flange, HSS, B1033.072c
angle)
B1033.073a, B1033.073b,
X brace
B1033.073c
B1033.101a, B1033.101b,
Buckling- - Chevron brace
B1033.101c
Restrained Brace
(BRB) Frame B1033.111a, B1033.111b,
- Single diagonal brace
B1033.111c
Eccentrically No floor beams B1035.061a, B1035.061b
Braced Frame Shear link
(EBF) With floor beams B1035.062a, B1035.062b
Post-Northridge
B1035.001, B1035.002, B1035.011,
Reduced Beam Welded Web -
B1035.012
Section (RBS)
Structural Steel
Moment Frame Post-Northridge B1035.021, B1035.022, B1035.031,
- -
not RBS B1035.032
Welded Unreinforced B1035.041, B1035.042, B1035.051,

Pre-Northridge -
Flange-Bolted WUF-B B1035.052
A-2 A: Structural Component Fragility Specifications FEMA P-58-2

Second Edition
Table A-2 Reinforced Concrete Elements (B104)
Construction Damage Mode,

Design Design Configuration; Other/Comments/ Fragility Classification
Material/System Characteristics Other Features Notes Number
B1041.001a, B1041.001b,
Beam yield - B1041.002a, B1041.002b,
B1041.003a, B1041.003b
ACI 318 Special
Moment Frame ΣMc/ΣMb > 1.2
(SMF)
Pu(col) < 0.6 fc’Ag B1041.011a, B1041.011b,
Weak joints Joint V/Vn < 1.2 B1041.012a, B1041.012b,
B1041.013a, B1041.013b
Compliant transverse
reinforcement
ACI 318
B1041.021a, B1041.021b,
Intermediate Beam or column shear
Joint shear damage B1041.022a, B1041.022b,
Moment Frame response
B1041.023a, B1041.023b
(IMF)
Joint shear damage B1041.031a, B1041.031b,
Beam yield, weak
ΣMc/ΣMb > 1.2 B1041.032a, B1041.032b,
joints
Pu(col) < 0.6 fc’Ag B1041.033a, B1041.033b
Column yield ΣMc/ΣMb < 0.8 B1041.041a, B1041.041b,

B1041.042a, B1041.042b,
Joint shear yield Beam, Col: Vn>Ve B1041.043a, B1041.043b
Reinforced Concrete ACI 318 Ordinary Beam yield (flexure or B1041.051a, B1041.051b,
Moment Frame Moment Frame shear) Beam: Vn<Ve B1041.052a, B1041.052b,
(OMF) B1041.053a, B1041.053b
Weak joints
Pu(col) < 0.6 fc’Ag B1041.061a, B1041.061b,

Column yield (strong
B1041.062a, B1041.062b,
joints) Column: Vn<Ve B1041.063a, B1041.063b
Weak columns with Pu(col) > 0.6 fc’Ag B1041.071a, B1041.071b,
high axial load (strong B1041.072a, B1041.072b,
joints) Column: Vn<Ve B1041.073a, B1041.073b
ΣMc/ΣMb > 1.2
V/Vn>1.2
B1041.081a, B1041.081b,
Weak joints, beam Or B1041.082a, B1041.082b,
yield Non-compliant B1041.083a, B1041.083b
Transverse Beam
Reinforcement
ACI 318 Non-
Conforming ΣMc/ΣMb < 0.8
Moment Frames
(Vp/Vn) Col < 0.6
Joint V/Vn>1.2 B1041.091a, B1041.091b,
Weak joints, column
Or B1041.092a, B1041.092b,
yield
B1041.093a, B1041.093b
Transverse Column
Reinforcement with 90°
hooks

Second Edition
Table A-2 Reinforced Concrete Elements (B104) (continued)
Construction Damage Mode, Design
Design Configuration; Other Other/Comments/ Fragility Classification
Material/System Characteristics Features Notes Number
B1041.101a,
Beam with High V/bwd(fc’)1/2 B1041.101b,
Weak beams, strong
Or B1041.102a,
joints (beam flexure or
Non Compliant Transverse B1041.102b,
shear)
Beam Reinforcement B1041.103a,
B1041.103b
B1041.111a,
Weak columns (flexure ΣMc/ΣMb < 0.8 B1041.111b,
or shear yield) B1041.112a,
(Vp/Vn) Col > 1.0
Low to moderate axial B1041.112b,
Reinforced Concrete column loads Pu(col) > 0.6 fc’Ag B1041.113a,
ACI 318 Non-
Moment Frame Conforming B1041.113b
(continued) Moment Frames B1041.121a,
(continued) B1041.121b,
ΣMc/ΣMb < 0.8
Weak columns strong B1041.122a,
Col Vn<VE
joints, high axial B1041.122b,
Pu(col) > 0.6 fc’Ag B1041.123a,
B1041.123b
B1041.131a,
Inadequate B1041.131b,
reinforcement Inadequate Anchorage,
B1041.132a,
development Splices, Discontinuous
B1041.132b,
Reinforcement
(beam or column) B1041.133a,
B1041.133b
Aspect ratio between 1.0 B1042.001a, B1042.011a,
and 2.0 B1042.021a
Diagonally reinforced
Aspect ratio between 2.0 B1042.001b, B1042.011b,
and 4.0 B1042.021b
Concrete link beam
Aspect ratio between 1.0 B1042.002a, B1042.012a,
and 2.0 B1042.022a
Conventionally reinforced
Aspect ratio between 2.0 B1042.002b, B1042.012b,
and 4.0 B1042.022b
B1044.001,
Planar, single curtain
8 inches or less thick B1044.002,
Rectangular low reinforcement
Reinforced Concrete B1044.003
aspect ratio wall
Shear Wall
h/l < 1.0 B1044.011, B1044.021,
Planar, double curtain
8”-24” thick B1044.012, B1044.022,
reinforcement
B1044.013, B1044.023
Return flanges, single curtain B1044.031, B1044.032,
8 inches or less thick
reinforcement B1044.033
B1044.041, B1044.051,
Return flanges, double
8”-24” thick B1044.042, B1044.052,
Low-rise wall curtain reinforcement
B1044.043, B1044.053
B1044.061,
Boundary columns, single
8 inches or less thick B1044.062,
curtain reinforcement
B1044.063

Second Edition
Table A-2 Reinforced Concrete Elements (B104) (continued)
B1044.071, B1044.081,
Low-rise wall Boundary columns, double
8”-24” thick B1044.072, B1044.082,
(continued) curtain reinforcement
B1044.073, B1044.083
15 feet long B1044.091
12 inches thick 20 feet long B1044.092
Reinforced Concrete
Shear Wall (continued) 15 feet long B1044.101
Slender wall
h/l > 2.0
No continuity
B1049.001a
reinforcement
0 < Vg/V0 < 0.2
With continuity
B1049.001b
reinforcement
Flat slab without No continuity

B1049.002a
shear reinforcement
0.2 ≤ Vg/V0 < 0.4
reinforcement at With continuity
columns B1049.002b
reinforcement
No continuity
B1049.003a
reinforcement
0.4 ≤ Vg/V0 < 0.6
With continuity
B1049.003b
reinforcement
Flat slab with shear 0.2 ≤ Vg/V0 < 0.4 - B1049.011
reinforcement at
columns 0.4 ≤ Vg/V0 < 0.6 - B1049.012
Reinforced Concrete
No continuity reinforcement B1049.021a
Moment Frame
Post-tensioned slab 0.2 ≤ Vg/V0 < 0.4 With continuity
without shear B1049.021b
reinforcement
reinforcement at
columns No continuity reinforcement B1049.022a
0.4 ≤ Vg/V0 < 0.6 With continuity
reinforcement B1049.022b
Post-tensioned slab 0 < Vg/V0 < 0.4 - B1049.031

with shear
reinforcement at
columns 0.4 ≤ Vg/V0 < 0.6 - B1049.032
No continuity reinforcement B1049.041a

0 ≤ Vg/V0 < 0.4
Flat slab with drop With continuity
panel or capitol B1049.041b
reinforcement
without shear
reinforcement at No continuity reinforcement B1049.042a
columns 0.4 ≤ Vg/V0 < 0.6 With continuity
B1049.042b
reinforcement

Second Edition
Table A-3 Masonry Vertical Elements (B105)

B1051.001, B1051.012,
Shear dominated - B1051.002, B1051.021,
B1051.011, B1051.022
Partially grouted - B1051.003, B1051.023
B1051.004, B1051.024
Flexure dominated
Reinforced Masonry B1051.013
Bearing Wall
B1051.014
- B1052.001, B1052.011
Shear dominated
B1052.002, B1052.012
Fully grouted
- B1052.003, B1052.013
Flexure dominated
B1052.004, B1052.014
Table A-4 Cold-Formed Steel Structural Elements (B106)

Structural sheathing interior - B1061.001a

Wood panel
sheathing
Stucco finish exterior - B1061.001b
Cold-Formed Steel
Stud Wall - - B1061.011a
Flat strap ×
bracing Stucco finish exterior - B1061.011b
22 mil/31 mil - - B1061.021a

steel sheathing Stucco finish exterior - B1061.021b
Table A-5 Wood Light Frame Structural Elements (B107)

No holdowns - B1071.001
OSB-ply,
Light Wood Frame Holdowns - B1071.021
Shear Wall No holdowns - B1071.011
OSB-ply,
stucco Holdowns - B1071.002
Light Wood Frame, -
Diagonal let-in
Diagonal Strut No holdowns B1071.031
bracing
Bracing
Gypsum with
Exterior Wall Fixed below, fixed above B1071.041
wood studs

Second Edition
Appendix B
Nonstructural Component
Fragility Specifications
Table B-1 Exterior Nonstructural Walls (B201)
Gypsum with
Exterior Wall Fixed below, fixed above B1071.041
wood studs
Structural Sheathing Interior - B2011.001a

Wood panel
sheathing
Stucco Finish Exterior - B2011.001b
Exterior Wall
Cold Formed Steel - - B2011.011a
Flat strap
Stud Walls
X Bracing Stucco Finish Exterior - B2011.011b
22 Mil/31 Mil - - B2011.021a

Steel Sheathing Stucco Finish Exterior - B2011.021b
No holdowns - B2011.101
OSB-ply
Exterior Wall Holdowns - B2011.121
Light Wood Frame
Shear Wall OSB-ply, No holdowns - B2011.111
stucco Holdowns - B2011.102
Exterior Wall -
Light Wood Frame, Diagonal let-in
No holdowns B2011.131
Diagonal Strut bracing
Bracing
Exterior Wall In-plane deformation B2011.201a
Precast Concrete 4.5 inches thick
Panels Out-of-plane deformation B2011.201b
FEMA P-58-2 B: Nonstructural Component Fragility Specifications B-1

Second Edition
Table B-2 Exterior Window Systems (B202)
Fragility
Glass Classification
Basic Type Configuration Lamination Type Details Number
Monolithic Not Not Aspect ratio = 6:5 B2022.001
Generic (single pane) Specified Specified Other details Unknown
Midrise
Stick-Built Insulating Not Not Aspect ratio = 6:5 B2022.002
Curtain wall Glass Units Specified Specified Other details Unknown
(dual pane)
Asymmetric Laminated Annealed 1/4 in. (6 mm) inner AN / 1/2 in. B2022.011
insulating glass (LAM) (AN) (13 mm) outer AN LAM (0.030
units (dual- PVB) IGU; glass-frame clearance =
pane, 0.43 in. (11 mm); aspect ratio =
unequal- 6:5 sealant = dry
thickness IGU)
1/4 in. (6 mm) AN LAM (0.030 B2022.021
PVB); glass-frame clearance =
0.43 in. (11 mm); aspect ratio =
6:5 sealant = dry
1/4 in. (6 mm) AN laminated, 2- B2022.022
sided SSG, Center Panel; glass-
frame clearance = 0.43 in. (11
Laminated Annealed
mm); aspect ratio = 6:5 sealant =
wet
1/4 in. (6 mm) AN laminated, 2- B2022.023
sided SSG, Outside Panel; glass-
wet
Midrise 1/4 in. (6 mm) AN monolithic; B2022.031

Stick-Built glass-frame clearance = 0.43 in.
Curtain (11 mm); aspect ratio = 6:5
Wall sealant = dry
Monolithic 1/4 in. (6 mm) AN monolithic; B2022.032
(single pane) glass-frame clearance = 0 in. (0
dry
1/4 in. (6 mm) AN monolithic; B2022.033
glass-frame clearance = 0.13 in. (3
Not dry
Annealed
laminated 1/4 in. (6 mm) AN monolithic; B2022.034
glass-frame clearance = 0.25 in. (6
dry
glass-frame clearance = 0.43 in.
(11 mm); aspect ratio = 2:1
sealant = dry
sealant = dry
B-2 B: Nonstructural Component Fragility Specifications FEMA P-58-2

Second Edition
Table B-2 Exterior Window Systems (B202) (continued)
Fragility
1/4 in. (6 mm) AN monolithic, 2- B2022.037
sided SSG, Center Panel; glass-
wet
1/4 in. (6 mm) AN monolithic, 2- B2022.038
wet
Annealed 1/4 in. (6 mm) AN monolithic; B2022.039
(cont’d) glass-frame clearance = 0.43 in.
sealant = dry
sealant = dry
Monolithic Not sealant = dry
(single pane) laminated
Midrise (cont’d) (cont’d) 1/4 in. (6 mm) FT monolithic, 2- B2022.051
Stick-Built sided SSG, Center Panel; glass-
Curtain frame clearance = 0.43 in. (11
Wall mm); aspect ratio = 6:5 sealant =
(cont’d) wet
1/4 in. (6 mm) FT monolithic, 2- B2022.052
Fully mm); aspect ratio = 6:5 sealant =
tempered wet
(FT)
1/4 in. (6 mm) FT monolithic, B2022.053
Seamed Edge; glass-frame
clearance = 0.43 in. (11 mm);
aspect ratio = 6:5 sealant = dry
Polished Edge; glass-frame
1/4 in. (6 mm) HS monolithic; B2022.061
Heat
strengthe
ned (HS)
sealant = dry
Symmetric 1/4 in. (6 mm) inner AN / 1/4 in. B2022.071
insulating glass (6 mm) outer AN LAM (0.030
units (dual- Laminated Annealed PVB) IGU; glass-frame clearance =
pane, equal- 0.43 in. (11 mm); aspect ratio =
thickness IGU) 6:5 sealant = dry

Second Edition
Fragility
1/4 in. (6 mm) inner AN / 1/4 in. B2022.072
(6 mm) outer AN LAM (0.060
Laminated Annealed
PVB) IGU; glass-frame clearance =
(cont’d) (cont’d)
6:5 sealant = dry
1 in. (25 mm) AN insulating glass B2022.081
unit (IGU) [1/4 in. (6 mm) inner
and outer panes]; glass-frame
1 in. (25 mm) AN IGU [1/4 in. (6 B2022.082
mm) inner and outer panes]; glass-
dry
Annealed 1 in. (25 mm) AN IGU [1/4 in. (6 B2022.083

Symmetric
Midrise mm) inner and outer panes], 2-
insulating glass
Stick-Built sided SSG, Center Panel; glass-
units (dual-
Curtain frame clearance = 0.43 in. (11
pane, equal-
Wall mm); aspect ratio = 6:5 sealant =
thickness IGU)
(cont’d) Not wet
(cont’d)
laminated 1 in. (25 mm) AN IGU [1/4 in. (6 B2022.084
mm) inner and outer panes], 2-
wet
1 in. (25 mm) FT IGU [1/4 in. (6 B2022.091
mm) inner and outer panes]; glass-
Fully
tempered
dry
1 in. (25 mm) HS IGU [1/4 in. (6 B2022.101
Heat mm) inner and outer panes]; glass-
strengthe frame clearance = 0.43 in. (11
ned mm); aspect ratio = 6:5 sealant =
dry
Unitized 1-1/4 in. (32 mm) FT IGU [1/4 in.
Symmetric
Curtain (6 mm) inner and outer panes], 4- B2022.201
insulating glass
Wall (also Not Fully sided SSG, VHBTM SGTTM; glass-
units (dual-
generic laminated tempered frame clearance = 0.43 in. (11
pane, equal-
unitized mm); aspect ratio = Varies sealant
thickness IGU)
curtain wall) = wet
Monolithic Not Not Aspect ratio = 6:5 B2023.001
(single pane) Specified Specified Other details Unknown
Generic
Storefront IGU Aspect ratio = 6.5 B2023.002
Not Not
(dual pane) Specified Specified Other details Unknown

Second Edition
Fragility
Glass
Basic Type Configuration Lamination Classification
Type
Details Number
1/4 in. (6 mm) AN LAM (0.030 B2023.011
PVB); glass-frame clearance =
Laminated Annealed
6:5 sealant = dry
Annealed
sealant = dry
Planar Specimen; glass-frame
aspect ratio = 1.93:1 sealant =
wet
Monolithic 3/8 in. (10 mm) FT monolithic, B2023.032
(single pane) Corner Cond. Specimen, in-plane;
Not glass-frame clearance = 0.5 in. (13
laminated mm); aspect ratio = 1.93:1 sealant
Fully = wet
tempered 3/8 in. (10 mm) FT monolithic, B2023.033
Corner Cond. Specimen,
combined; glass-frame clearance
= 0.5 in. (13 mm); aspect ratio =
1.93:1 sealant = wet
Storefront Corner Cond. Specimen, Short
side; glass-frame clearance = 0.5
in. (13 mm); aspect ratio = 2.72:1
sealant = wet
1 in. (25 mm) AN IGU; glass-frame B2023.041
Annealed clearance = 0.59 in. (15 mm);
1 in. (25 mm) FT IGU [1/4 in. (6 B2023.051
mm) inner and outer panes],
Planar Specimen; glass-frame
aspect ratio = 1.89:1 sealant =
Symmetric dry
insulating glass 1 in. (25 mm) FT IGU [1/4 in. (6 B2023.052
Not
units (dual- mm) inner and outer panes],
laminated
pane, equal- Fully Corner Cond. Specimen, in-plane;
thickness IGU) tempered glass-frame clearance = 0.43 in.
(11 mm); aspect ratio = 1.89:1
sealant = dry
1 in. (25 mm) FT IGU [1/4 in. (6 B2023.053
mm) inner and outer panes],
Corner Cond. Specimen,
combined; glass-frame clearance
= 0.43 in. (11 mm); aspect ratio
= 1.89:1 sealant = dry

Second Edition
Fragility
1 in. (25 mm) FT IGU [1/4 in. (6 mm) B2023.054
inner and outer panes], Corner Cond.
Specimen, Short side; glass-frame
clearance = 0.43 in. (11 mm); aspect
ratio = 2.83:1 sealant = dry
1 in. (25 mm) FT IGU [1/4 in. (6 mm) B2023.055
inner and outer panes], Planar
Specimen; glass-frame clearance =
1.88:1 sealant = dry
Symmetric
1 in. (25 mm) FT IGU [1/4 in. (6 mm) B2023.056
insulating glass
Not Fully inner and outer panes], Corner Cond.
Storefront units (dual-
laminated tempered Specimen, in-plane; glass-frame
(cont’d) pane, equal-
(cont’d) (cont’d) clearance = 0.43 in. (11 mm); aspect
thickness IGU)
(cont’d)
1 in. (25 mm) FT IGU [1/4 in. (6 mm) B2023.057
Specimen, combined; glass-frame
1 in. (25 mm) FT IGU [1/4 in. (6 mm) B2023.058
Specimen, Short side; glass-frame
Table B-3 Roof Elements (B301, B303, B304)

Fragility
Seismic Installation Condition / Repair Classification
Item Description Size Approach Number
Concrete tiles, secured and compliant with UBC B3011.011
94
Roof Tile N/A Clay tiles, secured and compliant with UBC 94 B3011.012
Concrete tiles unsecured B3011.013
Clay tiles unsecured B3011.014
Replace with masonry B3031.001a

5’ above roof
Replace without masonry B3031.002a
Masonry Chimneys Replace with masonry B3031.001b

10’ above roof
(unreinforced) Replace without masonry B3031.002b
Replace with masonry B3031.001c

15’ above roof
Replace without masonry B3031.002c
Masonry Parapet N/A Unreinforced, unbraced B3041.001

Second Edition
Table B-4 Partitions (C101)
Full or Fragility
Item Partial Corner Classificatio
Description Finish Height Configuration Installation Variable n Number
Full height Fixed below, fixed C1011.001a
wallboard above
GWB partition Partial height Fixed below, laterally C1011.001b

Paint with wallboard braced above
(steel studs, N/A
gypsum board taped joints Fixed below, slip track C1011.001c
both sides) above with returns
Full height
wallboard Fixed below, slip track C1011.001d
above without returns
GWB partition Paint with Full height N/A Fixed below, fixed C1011.011a
(wood studs, taped joints wallboard above
gypsum board
both sides)
Table B-5 Stairs (C201)

Fragility
Item Seismic installation Classification
Description Type condition/repair action Number
With seismic joint C2011.001a
Prefab steel
No seismic joint C2011.001b
With seismic joint C2011.011a

Precast concrete
No seismic joint C2011.011b
No seismic joint replace with C2011.021a

cast in place concrete
Stair With seismic joint replace C2011.021b

with cast in place concrete
Cast in place concrete
No seismic joint replace with C2011.021c
steel stair
With seismic joint replace C2011.021d

with steel stair
Hybrid: Steel stringers with precast No seismic joint C2011.031a

treads rigidly linked to a stringer With seismic joint C2011.031b

Second Edition
Table B-6 Wall Finishes (C301)
Full or Fragility
Item Partial Classification
Description Finish Height Installation Variable Number
Full height Fixed below, fixed above C3011.001a
Partial height Fixed below, laterally braced above C3011.001b
Wall paper
Full height Fixed Track Below, Slip Track C3011.001c,
Above with and without Returns C3011.001d
Wall Finish Full height Fixed below, fixed above C3011.002a

Applied to
One Side of a Ceramic tile
GWB Full height Fixed Track Below, Slip Track C3011.002c,
Partition Above with and without Returns C3011.002d
Full height Fixed below, fixed above C3011.003a
High-end
Marble or
wood panel Full height Fixed Below, Slip Track Above with C3011.003c,
and without Returns C3011.003d
Table B-7 Floor Finishes, Raised Access Floors, and Floor Flooding (C302)
Fragility
Seismic Installation Classification
Item Description Condition Size/Other Number
Office (Dry) C3021.001a
Office (humid) C3021.001b
Lab (dry) C3021.001c
Lab (humid) C3021.001d
Hospital (dry) C3021.001e
Hospital (humid) C3021.001f
School (dry) C3021.001g

Generic Floor
Covering and Wall School (humid) C3021.001h
N/A
Damaged by Flooding Apartments (dry) C3021.001i
(Single Pipe Failure)
Apartments (humid) C3021.001j
Retail (dry) C3021.001k
Retail (humid) C3021.001l
Warehouse (dry) C3021.001m
Warehouse (humid) C3021.001n
Hospitality (dry) C3021.001o
Hospitality (humid) C3021.001p
Unanchored, not N/A C3027.001

Raised Access Floor seismically rated
Seismically rated N/A C3027.002

Second Edition
Table B-8 Ceilings and Ceiling Lighting (C303)
Fragility
Seismic Installation Classification
Item Description Condition Size/Area Number
SDC: A,B Vertical support < 250 SF C3032.001a
only
250 – 1000 SF C3032.001b
1000 – 2500 SF C3032.001c
> 2500 SF C3032.001d
SDC C: Vertical support and < 250 SF C3032.002a

lateral support for larger
ceiling areas 250 – 1000 SF C3032.002b
1000 – 2500 SF C3032.002c
> 2500 SF C3032.002d

Suspended Ceiling
SDC D/E: (Ip = 1.0) Vertical < 250 SF C3032.003a
and lateral support for larger
ceiling areas plus wider 250 – 1000 SF C3032.003b
perimeter angle 1000 – 2500 SF C3032.003c
> 2500 SF C3032.003d
SDC: D,E,F (Ip = 1.5) < 250 SF C3032.004a

Same as D/E plus stringent 250 – 1000 SF C3032.004b
design and installation Q/C
1000 – 2500 SF C3032.004c
> 2500 SF C3032.004d
Independent Pendant Not seismically rated C3034.001

Lighting
Independent Seismically rated C3034.002

Pendant Lighting
Table B-9 Elevators and Lifts (D101)

Fragility
Item Classification
Description Type Seismic Installation Condition Number
Installed in California after 1975, in most D1014.011
Traction western states after 1981 and in most other US
States after 1997
Installed in California before 1976, in most D1014.012

Traction western states before 1982 and in most other
Elevator US States before 1998
Installed in California after 1975, in most D1014.021

Hydraulic western states after 1981 and in most other US
states after 1997
Installed in California before 1976, in most D1014.022

Hydraulic western states before 1982 and in most other
US States before 1998

Second Edition
Table B-10 Domestic Water Distribution Including Hot Water Heaters(D202)
Fragility
Item Installation Classification
Description Size Material Type Variable Number
SDC A or B D2021.011a,
D2021.011b
Cold Water Pipe SDC C D2021.012a,

Service, Cold or diameter, Threaded Steel D2021.012b
Hot Potable Φ ≤ 2.5” SDC D, E or F D2021.013a,
D2021.013b
SDC D, E, or F D2021.014a,
(OSHPD or sim.) D2021.014b
SDC A or B D2021.021a
SDC C D2021.022a
Cold Water Pipe
SDC D, E or F D2021.023a,
Service, Cold or diameter, Welded steel
Hot Potable Φ > 2.5” D2021.023b
SDC D, E, or F D2021.024a,
D2022.011b
SDC C D2022.012a,
Hot Water Pipe D2022.012b
Service, Heating diameter, Threaded steel
Hot Water Φ ≤ 2.5” SDC D, E or F D2022.013a,
D2022.013b
SDC D, E, or F D2022.014a,
D2022.021b
SDC C D2022.022a
Hot Water Pipe D2022.022b
Service, Heating diameter, Welded steel
Hot Water Φ > 2.5” SDC D, E or F D2022.023a,
D2022.023b
SDC D, E, or F D2022.024a,

Second Edition
Table B-11 Sanitary Waste Piping System (D203)
Item Installation Fragility Classification
SDC A or B D2031.011b
SDC C D2031.012b
Cast iron with flexible
couplings SDC D, E or F D2031.013b
SDC D, E, or F D2031.014b
(OSHPD or
sim.)
Sanitary Waste SDC A or B D2031.021a,
N/A
Piping D2031.021b
SDC C D2031.022a,
Cast iron with bell and D2031.022b
spigot couplings
D2031.023b
SDC D, E, or F D2031.024a,
(OSHPD or D2031.024b
sim.)
Table B-12 Chilled Water Piping (D205)

Item Installation Fragility Classification
D2051.011b
SDC C D2051.012a,
D2051.012b
Chilled Water Φ≤
Threaded steel SDC D, E or F D2051.013a,
Piping 2.5”
D2051.013b
SDC D, E, or F D2051.014a,
sim.)
D2051.021b
SDC C D2051.022a,
D2051.022b
Chilled Water Φ>
Welded steel SDC D, E or F D2051.023a,
Piping 2.5”
D2051.023b
SDC D, E, or F D2051.024a,
sim.)

Second Edition
Table B-13 Steam Piping (D206)
Fragility
Item Classification
Description Size Material Type Installation Variable Number
D2061.011b
SDC C D2061.012a,
Φ≤ D2061.012b
2.5” Threaded steel
D2061.013b
SDC D, E, or F (OSHPD or sim.) D2061.014a,
Steam Piping D2061.014b
D2061.021b
Φ>
2.5” Welded steel SDC C D2061.022a,
D2061.022b
D2061.023b
SDC D, E, or F (OSHPD or sim.) D2061.024a,
D2061.024b
Table B-14 Chillers, Cooling Towers, and Compressors (D303)

Fragility
Equipment Dominant Damage Classification
Description Installation Variables Size or Rated Capacity Mode Number
< 100 Ton D3031.011a
Unanchored, not isolated 100 to < 350 Ton Internal damage (ID) D3031.011b
350 to < 750 Ton D3031.011c
750 to < 1000 Ton D3031.011d
Anchorage failure D3031.012a
(AF)
< 100 Ton
Internal damage D3031.012b
Simultaneously D3031.012c
occurring AF and ID
Anchorage failure D3031.012d
100 to < 350 Ton Internal damage D3031.012e
Chiller Simultaneously D3031.012f
Isolated, not restrained occurring AF and ID
Anchorage failure D3031.012g
350 to < 750 Ton Internal damage D3031.012h
Simultaneously D3031.012i
occurring AF and ID
Anchorage failure D3031.012j
750 to < 1000 Ton Internal damage D3031.012k
Simultaneously D3031.012l
occurring AF and ID
Hard anchored or isolated < 100 Ton
and restrained Internal damage D3031.013b
Simultaneously D3031.013c
occurring AF and ID

Second Edition
Table B-14 Chillers, Cooling Towers, and Compressors (D303) (continued)
Equipment Dominant Damage Fragility Classification
Description Installation Variables Size or Rated Capacity Mode Number
100 to < 350 Ton
Internal damage D3031.013e
100 to < 350 Ton
Simultaneously D3031.013f
occurring AF and ID
Chiller
(continued) Hard anchored or isolated 350 to < 750 Ton Internal damage D3031.013h
and restrained (continued) Simultaneously D3031.013i
occurring AF and ID
Simultaneously D3031.013l
occurring AF and ID
< 100 Ton D3031.021a
Unanchored, not isolated 100 to < 350 Ton D3031.021b
Internal damage
350 to < 750 Ton D3031.021c
750 to < 1000 Ton D3031.021d
< 100 Ton Anchorage failure D3031.022a
< 100 Ton Internal damage D3031.022b
< 100 Ton Simultaneously D3031.022c
occurring AF and ID
100 to < 350 Ton Anchorage failure D3031.022d
100 to < 350 Ton Simultaneously D3031.022f
Isolated, not restrained occurring AF and ID
350 to < 750 Ton Anchorage failure D3031.022g
350 to < 750 Ton Internal damage D3031.022h
350 to < 750 Ton Simultaneously D3031.022i
occurring AF and ID
750 to < 1000 Ton Anchorage failure D3031.022j
Cooling
Tower 750 to < 1000 Ton Internal damage D3031.022k
750 to < 1000 Ton Simultaneously D3031.022l
occurring AF and ID
< 100 Ton Anchorage failure D3031.023a
< 100 Ton Internal damage D3031.023b
< 100 Ton Simultaneously D3031.023c
occurring AF and ID
100 to < 350 Ton Anchorage failure D3031.023d
100 to < 350 Ton Simultaneously D3031.023f
Hard anchored or isolated occurring AF and ID
and restrained
350 to < 750 Ton Anchorage failure D3031.023g
350 to < 750 Ton Internal Damage D3031.023h
350 to < 750 Ton Simultaneously D3031.023i
occurring AF and ID
750 to < 1000 Ton Anchorage failure D3031.023j
750 to < 1000 Ton Simultaneously D3031.023l
occurring AF and ID
Compressor Unanchored, not isolated Small non medical air supply D3032.011a
Internal damage
Large non medical air supply D3032.011b

Second Edition
Table B-14 Chillers, Cooling Towers, and Compressors (D303) (continued)
Fragility
Equipment Classification
Description Installation Variables Size or Rated Capacity Dominant Damage Mode Number
Small medical quality air D3032.011c
Unanchored, not isolated supply
(continued) Large medical quality air D3032.011d
supply
Small non-medical air Anchorage failure D3032.012a
supply
Simultaneously occurring AF D3032.012c
and ID
Large non-medical air Anchorage failure D3032.012d
supply
Simultaneously occurring AF D3032.012f
and ID
Isolated, not restrained
Small medical quality air Anchorage failure D3032.012g
supply
Internal damage D3032.012h
Simultaneously occurring AF D3032.012i
and ID
Large medical quality air Anchorage failure D3032.012j
Compressor supply
Internal damage D3032.012k
(continued)
Simultaneously occurring AF D3032.012l
and ID
Small non-medical air Anchorage failure D3032.013a
supply
Simultaneously occurring AF D3032.013c
and ID
Large non-medical air Anchorage failure D3032.013d
supply
Simultaneously occurring AF D3032.013f
Hard anchored or isolated and ID
and restrained Small medical quality air Anchorage failure D3032.013g
supply
Simultaneously occurring AF D3032.013i
and ID
Large medical quality air Anchorage failure D3032.013j
supply
Simultaneously occurring AF D3032.013l
and ID

Second Edition
Table B-15 Distribution Systems (D304)
Fragility
Equipment Seismic Rating, Dominant Classification
Description Installation Variables Size, Material Damage Mode Number
SDC C D3041.001b
Fan independently supported and vibration SDC D, E or F D3041.001c
isolated N/A
SDC D,E,F (high seismic D3041.001d
design) Special
Regulation (OSHPD) for
HVAC installations
HVAC Fan
In Line Fan SDC A or B D3041.002a
SDC C D3041.002b
SDC D, E or F D3041.002c
Fan independently supported but not on N/A
vibration isolated SDC D,E,F (high seismic D3041.002d
design) Special
Regulation (OSHPD) for
HVAC installations
< 6 SF in SDC C D3041.011b
cross sectional SDC D, E or F D3041.011c
area
SDC D, E, or F (OSHPD D3041.011d
or sim.)
Galvanized Sheet Metal
SDC C D3041.012b
≥ 6 SF in cross
sectional area SDC D, E or F D3041.012c
HVAC or sim.)
Ducting SDC A or B D3041.021a
SDC C D3041.021b
< 6 SF in cross
sectional area SDC D, E or F D3041.021c
or sim.)
Stainless Steel
≥ 6 SF in cross
sectional area SDC C D3041.022b
SDC D, E or F D3041.022c
≥ 6 SF in cross
sectional area SDC D, E, or F (OSHPD D3041.022d
or sim.)
HVAC SDC A or B D3041.031a
drops/
diffusers in No independent safety wires N/A SDC C D3041.031b
suspended
ceilings
HVAC
drops/ SDC C D3041.032b
Support by ducting only.
diffusers N/A SDC D, E or F D3041.032c
without No independent safety wires
ceilings SDC D, E, or F (OSHPD D3041.032d
or sim.

Second Edition
Table B-15 Distribution Systems (D304) (continued)
Fragility
Size, Seismic Rating, Dominant Classification
Equipment Description Installation Variables Material Damage Mode Number
Variable Air Volume (VAV) SDC A or B D3041.041a
Box with in-line coil N/A N/A
SDC C D3041.041b
Unanchored, not isolated N/A Internal damage D3041.101a
Isolated, not restrained N/A
Simultaneously occurring AF and D3041.102c
HVAC Fan ID
Hard anchored or isolated Internal damage D3041.103b
and restrained N/A
ID
Table B-16 Packaged Air Handling Units (D305)

Fragility
Description Installation Variables Size or Rated Capacity Dominant Damage Mode Number
< 5000 CFM D3052.011a
Unanchored, not 5000 to < 10000 CFM D3052.011b
Internal damage
isolated 10000 to < 25000 CFM D3052.011c
25000 to < 40000 CFM D3052.011d
< 5000 CFM
ID
Packaged Air Internal damage D3052.013e
Handling Unit 5000 to < 10000 CFM
Simultaneously occurring AF and D3052.013f
Hard anchored or ID
isolated and
restrained Anchorage failure D3052.013g
10000 to < 25000 CFM
Simultaneously occurring AF and D3052.013i
ID
25000 to < 40000 CFM
Simultaneously occurring AF and D3052.013l
ID
Table B-17 Control Panels and Instrumentation (D306)

Fragility
Description Installation variables Dominant Damage Mode Number
Unanchored Internal Damage D3067.011a
Control Panel
Hard anchored Internal Damage D3067.013b
Simultaneously occurring AF & ID D3067.013c

Second Edition
Table B-18 Fire Protection (D401)
Fragility
Classification
Item Description Size Material Type Installation Variable Number
D4011.011b
SDC C D4011.012a,
New style D4011.012b
Victaulic,
threaded steel SDC D, E or F D4011.013a,
D4011.013b
Fire Sprinkler Water
Piping – Horizontal N/A SDC D, E, or F (OSHPD or sim.) D4011.014a,
Mains and Branches D4011.014b
Old style SDC C D4011.022a
Victaulic, thin-wall
steel SDC D, E or F D4011.023a
SDC D, E, or F (OSHPD or sim.) D4011.024a
Dropping into tile SOFT ceiling, SDC A D4011.031a
or B
Dropping into lay-in tile SOFT ceiling, D4011.032a
Soft unbraced lay- SDC C
in tile ceiling Dropping into lay-in tile SOFT ceiling, D4011.033a
SDC D, E or F
SDC D, E, or F (OSHPD or sim.)
Dropping into lay-in tile HARD ceiling, D4011.041a
Hard unbraced SDC A or B
lay-in tile ceiling Dropping into lay-in tile HARD ceiling, D4011.042a
6 ft. long
drop max. SDC C
standard Dropping into lay-in tile SOFT ceiling, D4011.053a
Fire Sprinkler Drop
steel SDC D, E, or F
threaded
Braced lay-in tile SDC D, E, or F (OSHPD or sim.)
ceiling Dropping into lay-in tile HARD ceiling, D4011.063a
SDC D, E, or F
Dropping into lay-in tile HARD ceiling, D4011.064a
SDC D, E, or F (OSHPD or sim.)
No Ceiling, SDC A or B D4011.071a
No Ceiling, SDC C D4011.072a
No ceiling No Ceiling, SDC D, E or F D4011.073a
No Ceiling, SDC D, E, or F (OSHPD or D4011.074a
sim.)

Second Edition
Table B-19 Electrical Service and Distribution (D501)
Fragility
Equipment Size or Rated Classification
Description Installation Variables Capacity Dominant Damage Mode Number
< 100 kVA Internal damage D5011.011a
100 to < 350 kVA D5011.011b
Unanchored
350 to < 750 kVA D5011.011c
750 to 1500 kVA D5011.011d
< 100 kVA Internal damage D5011.013b
Simultaneously occurring AF and ID D5011.013c
Transformer/ Anchorage failure D5011.013d
Primary Service 100 to < 350 kVA Internal damage D5011.013e
Simultaneously occurring AF and ID D5011.013f
Hard anchored
350 to < 750 kVA
350 to < 750 kVA
Simultaneously occurring AF and ID D5011.013i
750 to 1500 kVA Internal damage D5011.013k
Simultaneously occurring AF and ID D5011.013l
Unanchored N/A D5012.013a
Motor Control Anchorage failure D5012.013b
Center Hard anchored N/A D5012.013c
D5012.013d
100 to < 350 Amp Internal damage D5012.021a
350 to < 750 Amp D5012.021b
Unanchored
750 to < 1200 Amp D5012.021c
1200 to 2000 Amp D5012.021d
100 to < 350 Amp Internal damage D5012.023b
Low Voltage Anchorage failure D5012.023d
Switchgear 350 to < 750 Amp Internal damage D5012.023e
Hard anchored
750 to < 1200 Amp Internal damage D5012.023h
1200 to 2000 Amp Internal damage D5012.023k

Second Edition
Table B-19 Electrical Service and Distribution (D501) (continued)
Fragility
Equipment Installation Classification
Description variables Size or Rated Capacity Dominant Damage Mode Number
100 to < 350 Amp D5012.031a
Unanchored, not
350 to < 750 Amp D5012.031b
isolated Internal damage
750 to < 1200 Amp D5012.031c
1200 to 2000 Amp D5012.031d
100 to < 350 Amp Internal damage D5012.033b
Distribution
Panel
350 to < 750 Amp Internal damage D5012.033e
Hard anchored Simultaneously occurring AF and ID D5012.033f
750 to < 1200 Amp Internal damage D5012.033h
1200 to 2000 Amp Internal damage D5012.033k
Table B-20 Other Electrical Systems (D509)

Fragility
Unanchored Internal damage D5092.011a
Battery Rack N/A
Hard anchored Internal damage D5092.013b
Unanchored Internal damage D5092.021a
Battery Charger N/A
Hard anchored Internal damage D5092.023b
100 to < 350 kVA D5092.031a
Unanchored, not 350 to < 750 kVA D5092.031b
Internal damage
isolated 750 to 1200 kVA D5092.031c
1200 to 2000 kVa D5092.031d
100 to < 350 kVA Internal damage D5092.032b
Diesel
Generator
Isolated, not restrained 350 to < 750 kVA Internal damage D5092.032e
750 to 1200 kVA Internal damage D5092.032h

Second Edition
Table B-20 Other Electrical Systems (D509) (continued)
Fragility
Isolated, not restrained
1200 to 2000 kVa Internal damage D5092.032k
(cont’d)
100 to < 350 kVA Internal damage D5092.033b
Diesel Anchorage failure D5092.033d

Generator 350 to < 750 kVA Internal damage D5092.033e
(cont’d)
Hard anchored or Simultaneously occurring AF and ID D5092.033f
isolated and restrained Anchorage failure D5092.033g
750 to 1200 kVA Internal damage D5092.033h
1200 to 2000 kVa Internal damage D5092.033k
Table B-21 Movable Furnishing (E202)

Fragility
Classification
Item Description Size Seismic Installation Condition Number
Modular workstation N/A N/A E2022.001
Unsecured fragile Unknown E2022.010
N/A
objects on shelves
Secured by museum putty, or other weak but E2022.011
Fragile contents N/A
sticky stuff
Fragile contents on In storage cabinets with latches E2022.012
N/A
shelves
Unsecured fragile Low friction surface E2022.013
N/A
objects on shelves
Home entertainment Unknown or average E2022.020
N/A
equipment on shelves
Electronic Equipment On wall mount brackets E2022.021
on wall mount N/A
brackets
Desktop Electronics N/A On slip-resistant surface E2022.022
including computers,
N/A On smooth surface E2022.023
monitors, stereos, etc.
Unanchored E2022.102a
2 shelves
Anchored E2022.102b
Bookcase Unanchored E2022.103a
3 shelves
Anchored E2022.103b
4 shelves Unanchored E2022.104a

Second Edition
Table B-21 Movable Furnishing (E202)
Fragility
Classification
5 shelves
Anchored E2022.105b
Bookcase (cont’d)
6 shelves
Anchored E2022.106b
2 drawer
Anchored E2022.112b
Vertical Filing Cabinet
4 drawer
Anchored E2022.114b
4 drawer
Anchored E2022.124b
Lateral Filing Cabinet
5 drawer
Anchored E2022.125b
Table B-22 Special Structures (F101)

Fragility
Classification
Storage rack 3-5 levels, 14.5-20 ft Comply with Rack Manufacturing Institute F1012.001
high Seismic Provisions Prior to 2007 Edition

Second Edition
Appendix C
PACT User Manual
C.1 Introduction
The Performance Assessment Calculation Tool (PACT) provided in

Volume 3 is the computational companion to Volume 1. PACT provides a
user-friendly platform for scenario-based, intensity-based, and time-based
loss calculations. It can accommodate results obtained from nonlinear
response history analyses as well as simplified analyses.
PACT provides three basic but integrated functions:

1. Gathering and organizing building information, fragility functions, and
demand parameters;
2. Performing loss calculations including repair cost, downtime, and
casualty estimates; and
3. Providing overall and performance group specific loss information
obtained from the above calculations.
PACT and its associated data management utilities (i.e., the Fragility
Specification Manager and Building Population Modeler) are coded in
Microsoft C#.net software development platform.
In this Second Edition, PACT has been enhanced to include an algorithm for
environmental impact assessment. Further details about the methodology
and its development are provided in FEMA P-58-3 BD 3.7.20.
C.2 Hardware and Software Requirements
A computer with the following minimum software and hardware are

necessary to use PACT:
• Microsoft Windows 2000, NT, XP, Vista, or 7 operating system
• 200 MB of available hard disk
• A Pentium 500 MHz or faster central processing unit (CPU)
• 256 MB of RAM
• A monitor with a minimum of 800×600 pixels resolution
FEMA P-58-2 C: PACT User Manual C-1

Second Edition
Only one commercial third party control is used (for creating graphs) and
PACT users do not need to license any software or utilities in order to use
PACT. Those who want to modify the code or further develop PACT need
to license only one third party control.
C.3 Installing PACT
PACT is provided as a single file installation executable, version 3.1.2, in

Volume 3. The setup program will guide through the installation when
double clicked.
PACT requires the free Microsoft Net Framework 3.5 to be installed on the
computer. Most newer computers have this utility installed as a part of their
operating system. However, the PACT installer checks the computer for
presence of the Microsoft Net Framework 3.5 and if necessary it will
download it from the Microsoft website and install it. Therefore, internet
connectivity during the PACT installation process is prudent.
Once the installation is completed, a shortcut to PACT is placed the desktop

as shown in Figure C-1.
Figure C-1 PACT shortcut icon.
It is recommended that users run PACT as an Administrator. To do this,

right click on the PACT icon and select “Run as Administrator.”
C.4 PACT Control Panel
When PACT is started, the PACT Control Panel appears. This is the focal
point of activities when the program is used. The user will come back to this
panel before and after completion of each major task.
The basic PACT operations are grouped together. They are:

• Model the building and import analyses results
• Evaluate performance, and
• Examine results.
The two PACT utilities that can be operated either within PACT or as totally
independent functions are:
• Fragility Specification Manager, and
• Building Population Modeler.
C-2 C: PACT User Manual FEMA P-58-2

Second Edition
The Reporting option allows generation of Microsoft Word reports.
Finally, the Exit button terminates PACT operation.
Figure C-2 PACT Control Panel.
C.5 Model the Building and Import Analyses Results
Selection of the Model the Building and Import Analyses Results button
initiates the Building Modeler window where the building, its contents,
results of structural analyses are defined, as shown in Figure C-3.

Second Edition
Figure C-3 PACT Building Modeler window.
C.5.1 Building Modeler Menus
PACT initially displays a blank canvas assuming that a new performance

assessment model is started from scratch. Figure C-4 illustrates how to
retrieve existing PACT performance assessment models.
Figure C-4 File menu.
A default five story office building model is built into PACT and can be
retrieved as shown in Figure C-5. Two versions of the model are provided:
One using the nonlinear response history method and the other using the
simplified (linear) method.

Second Edition
Figure C-5 Open File menu.
Once an existing PACT file is opened, the blank fields are overwritten by the
relevant information contained in the opened file.
Figure C-6 Project Information tab with retrieved data from opened file.
Under the File Menu, New command can be used to start a new project from
scratch, Recent Projects to see a list of projects you have worked on recently,
Save to save the project under its current name, and Save As to save a copy of
the project under a new name. On occasion, this action may set all demands
to zero, so users are encouraged to review the newly created file for
completeness. In addition, Page Setup can be used to setup a printer and
print preferences, Print to print a copy of the form displayed on the screen,
and Print Review to see how the selection would look. Finally, Exit

Second Edition
command would terminate the Building Modeler and return to the Control
Panel.
Under the Edit menu, shown in Figure C-7 Cut, Copy, Paste, and Select All
commands can be used to cut and paste contents of one or several cells. This
can be used to cut and paste results of analyses from a spreadsheet program
like Excel or tables contained in a word processing program such as Word.
Figure C-7 Edit, Tools, and Help menus.
Under the Tools Menu, Options allows for switching between imperial and
metric units. PACT is unit independent and both units are accommodated.
C.5.2 Project Information Tab
This tab, shown in Figure C-6, contains input fields for project identification,
building description, and client and engineer identifications. In addition,
regional and date cost multipliers for the project are to be entered here. Note
that PACT cost estimates are based on the 2011 Northern California average
costs. The user may choose to use a multiplier based on the year the
evaluation is being run and the region where the building is located.
Since performance assessment with PACT involves generation of a multitude

of random numbers, in order to get exactly the same result each time the
program is executed for the same building, the same solver random seed
value should be used. Otherwise, results that are very close but are not
necessarily identical to those obtained from previous executions of the
program will be output. Putting in a value of zero will mean that the
program will generate a new random seed every time the results are
calculated.
C.5.3 Building Information Tab
This tab, shown in Figure C-8, contains input fields for adding, editing, and
reviewing basic building information such as number of stories, total
replacement cost and time, core and shell replacement cost, maximum
number of workers per unit floor area (needed for estimating repair time),

Second Edition
total loss threshold as a fraction of total replacement cost, and carbon
emissions and embodied energy replacement values.
Figure C-8 Building Information tab.
Filling the Most Typical Default Values for floor area and story height
populates these values in the table shown in the lower part of the form. At
any time, the tabulated values in the table can be changed by clicking on the
corresponding cell. For example, if a 10-story building has a first floor
height of 20 ft, second floor height of 16 ft, and all other floors are 13.5 ft
high each, then 13.5 ft will be entered as the most typical height and
consequently the column will be populated with 13.5 ft. The user can then
change the values for the first two floors manually.
Each floor can be assigned a custom name. This will have no effect on the
building logic, but will show up in the reports and output, making the
basement and roof floors easier to distinguish.
Height Factor, Hazmat Factor, and Occupancy Factor values for each
floor can also be entered here. The costs of all damaged items on each floor
will be multiplied by these three cost factors.
C.5.4 Population Tab
The Population tab, shown in Figure C-9 is divided into three separate
windows. In the first one, the most typical population models for the
building and the fraction of floor area that they apply to are selected. For

Second Edition
example, if the typical floor of the building consists of 80% commercial
office space and 20% retail space, on the first row Commercial Office should
be selected for Occupancy with 0.8 Fraction on the same row and then the
second row will be filled with Retail and 0.2.
Figure C-9 Population tab.
Default population models, presented on the right side of the form for this
particular project, can also be easily edited. Note, however, that if changes
are made to the population models, these changes will be saved for this
project only and will not change the values stored in the Building
Population Modeler utility. Eight population models are provided with
PACT in the Building Population Modeler utility, additional population
models can be added.
The values entered on the upper left of the window as typical values are
automatically applied to all floors except the roof. The dropdown list and
arrows on the bottom left window can be used to change these values for
each floor by moving from one floor to another using the floor arrows.
The information corresponding to the selected population model is shown on

the right window. Population distribution per month, day of the week, and
hour of the day may be viewed in tabular of graphical form, as shown in
Figure C-10.

Second Edition
Figure C-10 Population model data screens.
A variety of icons are displayed above PACT graphs and charts and aid in
copying, exporting, printing, changing, and customizing how graphs are
presented, printed and saved.
C.5.5 Component Fragilities Tab
The Component Fragilities tab is organized in two parts, as shown in Figure

C-11. The top half is used to identify the most typical fragility specifications
to be included in the model of the building and the horizontal directions they
apply to. The population model corresponding to the selected fragility
specification need also be input here. If the same fragility specification is
present at more than one place in the structure, the fragility function needs to
be repeated so that there is one for each of the population models that apply
to that fragility specification.
On the bottom half of the window, the list is copied to all floors. The
dropdown list and arrows can be used to add, remove, or change the fragility
specifications at each floor floors.

Second Edition
Figure C-11 Component Fragilities tab.
C.5.6 Performance Groups Tab
Once the fragility specifications are selected on the Component Fragilities

tab, the Performance Groups tab, shown in Figure C-12, is used to furnish
PACT with the relevant quantities and other needed information. First, the
table should be populated with the selected fragility specifications in the
previous tab using the button on the righthand side.
Different directions and floors can be navigated to by using the

and buttons. If there is
uncertainty about the quantities (for example, number of desktop computers
at a floor), an appropriate Quantity Dispersion can be specified. The
default Fragility Correlated flag assigned to each fragility specification can
be changed by checking or unchecking the corresponding box. Note,
however, that making fragilities uncorrelated may substantially increase

Second Edition
execution time. The population model applicable to each performance group
can be changed if more than one population model is utilized in the building.
Figure C-12 Performance Groups tab.
In addition to the commonly used engineering demand parameters (EDP),

such as story drift ratio and floor acceleration, the user can also specify
special or custom demand parameters. For example, if the fragility function
for door frames is defined by doors jamming if residual drift exceeds a
certain value, residual drift ratio can be defined as a new Special EDP, as
shown in Figure C-13. This will prompt for residual drift ratio values in
addition to other demand parameters.
Figure C-13 Demand parameter selection data block.
If one or more fragility specifications were not entered in the Component

Fragilities tab, they may be directly added to selected performance groups
using button at the lower left.
C.5.7 Collapse Fragility Tab
The Collapse Fragility tab is used to model structural collapse and its
consequences. If potential collapse is not necessary in the building
assessment, the corresponding box at the top of Figure C-14 can be

Second Edition
unchecked. Collapse fragility is defined in terms of the median spectral
acceleration, g, at the fundamental period of structure that can cause
structural collapse and the associated dispersion. At least one collapse mode
needs to be defined, but there is no upper limit on the number of collapse
modes that can be included. In addition, the mutually exclusive probability
of collapse in each mode conditioned on collapse occurring, is also entered.
The fraction of each floor affected by a collapse is specified in a table. For
example, Figure C-14 shows three collapse modes representing total
collapse, soft story collapse, and middle floor collapse are selected.
Once the collapse modes are defined, the arrows in the Collapse
Consequences data block are used to cycle through modes and define the
consequences of occurrence of each mode including the fatality and injury
rates per floor and the corresponding dispersions. For example, in Figure
C-14, the mean fatality rate of 0.90 means that if this type of collapse occurs,
90% of the floor occupants at the time of the event will be perished.
Figure C-14 Collapse Fragility tab.
C.5.8 Structural Analysis Results Tab
The Structural Analysis Results tab is used to input results of structural

analyses. As shown in Figure C-15, the first input is needed to specify the
assessment type. Intensity-based assessments should be conducted if there is
an intention of conducting a time-based evaluation down the road. Next, the
type of analysis results should be selected: Nonlinear or linear (simplified).
For nonlinear analysis, the typical number of demand vectors to import,
should be defined, although the number can change for each intensity or

Second Edition
scenario. For simplified analysis, the number of demand vectors is always
one (see Figure C-16). The number of realizations that PACT will perform
for each intensity/scenario is provided next. Note, that in order to obtain a
realistic realization of casualty figures, a rather large number of realizations
is needed because PACT assigns a random date and time to each realization
affecting the number of people in the building and enough realizations are
needed to cover the population distribution of the building during a whole
year in a statistically meaningful way.
PACT will automatically determine the non-directional demand parameters

by multiplying the maximum value for the two inputted directions by a
conversion factor. The conversion factor can be entered here.
Figure C-15 Structural Analysis Results tab for entry of nonlinear analysis results.

Second Edition
Figure C-16 Structural Analysis Results tab for entry of simplified analysis results.
Under the Identify Intensity data block, click to add as many

scenarios/intensities as needed.
For each scenario/intensity, a description for each scenario/intensity should

be provided in the space provided for this purpose. If a particular
scenario/intensity has a different number of demand vectors than the
“Typical” value set above, it can be changed. A modeling dispersion can
also be set for each scenario/intensity, as well as spectral acceleration that
will be used for calculating collapse probabilities.
The table needs to be filled with the results of analyses for each
scenario/intensity in each direction and for each demand parameter. The
arrows can be used to cycle through various scenarios/intensities. If an Excel
spreadsheet containing this information is created, the data could be cut and
paste into the PACT table.
C.5.9 Residual Drift Tab
In the Residual Drift tab, a lognormal fragility function for maximum

building residual drift ratio, as well as the residual drift ratio results from the
nonlinear analysis for each intensity or scenario can be entered. If there is no
need to use residual drift calculations, the box at the top of the tab should be
unchecked, as shown in Figure C-17.

Second Edition
Figure C-17 Residual Drift tab.
C.5.10 Hazard Curve Tab
PACT will transfer the spectral accelerations entered in the previous tab into
the hazard curve table contained in the Hazard Curve tab, shown in Figure
C-18. These values are grayed out so that they cannot be changed here, they
can only be changed in the previous tab. Once the mean annual frequency of
exceedance (MAFE) values corresponding to each spectral acceleration are
entered, the hazard curve will be drawn.
C.5.11 Saving the PACT Model
Once all of the information has been entered in the tabs, the model should be
saved under a desired name.

Second Edition
Figure C-18 Hazard Curve tab.
C.6 Evaluate Performance
Once the model has been developed, the Evaluate Performance button is
activated in the PACT Control Panel and the window shown in Figure C-19
will appear.
Figure C-19 PACT Engine window.
As illustrated in Figure C-20, from the File menu, selecting Open Building
Project will allow the selection of the desired model. If the computer has a
dual- or multi-core processor, selecting the Use Threading under the Options

Second Edition
menu and clicking Yes will utilize the parallel processing capabilities of the
computer to speed up the evaluation process.
Figure C-20 Analysis options menu.
When the Run Evaluation button is clicked, depending on the speed of the
computer, the number of realizations selected and whether fragilities are
correlated or uncorrelated, the PACT execution time may take anywhere
between a few seconds to a number of hours. Progress bars at the bottom of
the form indicate the status of operation, as shown in Figure C-21, and upon
completion of execution the word “Done” will be displayed.
Figure C-21 PACT Engine progress screen.

Second Edition
Upon completion of execution, the results can be evaluated immediately
because PACT keeps these results available for viewing as long as the user
does not exit from PACT or initiate a new model. The file can also be saved
as a .bin file or a .xml file; however, this action may cause out of memory
errors.
C.6.1 Command Line Evaluation
There is also a command line version of the PACT Engine. The filename is
PactEngine.exe, and is located in the main installation directory. This file
can be run from the command line, in a batch file, or from an external
program. It uses command line switches to change the default options. It
will accept multiple input files at one time. The format of the command line
is:
“PactEngine InputFileName(s) -t[+-] -e[cnd] -s -e -d –b”
The available command line switches are:

• -h: Display the help information
• -t+: Turn multi-threading on
• -t-: Turn multi-threading off
• -s #: Override the seed value
• -e: Export internal data
• -d: Add date/time stamp to output filename
• -b: Save results as a binary formatted file (default is XML format)
The output file will be saved in the same directory as the input file, with the
default name of “inputFileName_results.xml,” unless the -d option is
selected, where it will add the date/time stamp of the analysis to the run, in
which case the format will be “InputFileName_results_YYYY-MM-
DDTHH.MM.SS.xml” (e.g., InputFileName_results_2011-12-
31T14.30.00.xml is Dec 31, 2011, at 2:30 in the afternoon).
If a particular command line switch is not used it will use the default option
last selected when using the graphical interface, unless otherwise specified
above.
C.7 Examine Results
Once the engine has run, the Examine Results button is activated in the
PACT Control Panel and the window shown in Figure C-22 will appear.

Second Edition
Figure C-22 Results window.
C.7.1 Scenario/Intensity Results
By default, PACT initially displays the Scenario/Intensity Results window,

shown in Figure C-22.
In all of the tabs in this window, by default, the cursor moves along the data
bins and not the lognormal curve. This can be changed using the Options
menu, as shown in Figure C-23.
Figure C-23 Performance curve display options.
Clicking the legend of a performance group on the bar chart causes the bar
chart for that performance group to stand out by darkening the color of other
bars. A click on the top of the legend box restores the original colors. The
radio buttons provided at the bottom left of the form can be used to either
show the entire graph or use scale breaks for viewing details at either end of
the horizontal scale, as shown in Figure C-24.

Second Edition
Figure C-24 Results displays.
For any of the graphs, fragilities can be grouped by their Uniformat ID,
which will simplify the keys on the graphs. Under the Tools/Options menu
there is an entry titled PG Grouping Level, as shown in Figure C-25. The
categories range from All Fragilities where each fragility will be one entry in
the graphs with own key entry and color, all the way to Group by Uniformat
Major Group Elements, where all the fragilities will be grouped into top level
categories such as “shell,” “interior,” and “equipment and furnishings.” This
will be particularly helpful with buildings containing many fragilities. Figure
C-26 illustrates the effects of changing the performance grouping levels.
Figure C-25 Changing performance grouping levels.

Second Edition
Figure C-26 Effects of changing performance grouping levels.
C.7.1.1 Repair Cost Tab
The Repair Cost tab, shown in Figure C-27, contains two graphs. By
placing the cursor at the line separating the two graphs and holding the
mouse key down, and the line can be dragged to make either graph smaller or
larger. The desired Scenario/Intensity number can be selected using the drop
down list at the top left of the form.
The graph at the bottom of the form depicts the loss curve for cost associated
with the selected scenario/intensity for all performance groups in all
directions and throughout the height of the building. By default the cursor on
the graph is placed at the median value. However, clicking anywhere on the
curve will allow examining losses at various probabilities.

Second Edition
Note that the values corresponding to the selected point on the curve are
displayed on the text boxes at the bottom of the form. Instead of clicking the
cursor on the curve, a cost or probability can be entered in the text box and
the corresponding value will be shown in the other text box and the cursor
will move to the corresponding location on the curve.
The bar chart on the top depicts the distribution of losses among various
performance groups. Note that when the cost or probability on the loss
curves is changed the distribution of costs among performance groups shown
on the bar chart changes.
Figure C-27 Repair Cost tab.
C.7.1.2 Casualties Tab
The Casualties tab, shown in Figure C-28, contains four graphs. By placing
the cursor at the lines separating the graphs and holding the mouse key down,
and the lines can be dragged to resize the graphs. The desired
scenario/intensity number can be selected using the drop down list at the top
left of the form.
The graphs at the bottom of the form depict casualty probabilities associated
with the selected scenario/intensity. By default, the cursor on the graph is
placed at the median value. However, clicking anywhere on the curve will
allow examining losses at various probabilities.

Second Edition
Note that the values corresponding to the selected point on the curve are
displayed on the text boxes at the bottom of the form. Instead of clicking the
cursor on the curve, a cost or probability can be entered in the text box and
the corresponding value will be shown in the other text box and the cursor
will move to the corresponding location on the curve.
The bar chart on the top depicts the distribution of casualties caused by
various performance groups and collapse (if occurring). Note that when the
cursor position is changed, the distribution of casualties per performance
groups shown on the corresponding bar chart changes.
Figure C-28 Casualties tab.
C.7.1.3 Repair Time Tab
PACT estimates repair time by considering two repair strategies:

1. Serial time in which repair activities are limited to one floor at a time,
and
2. Parallel time in which all floors will be subjected to repair activities at
the same time.
Repair strategy can be selected by clicking the desired radio button, as shown
in Figure C-29. Also note that PACT identifies and reports items that may
require a long lead time just below these radio buttons.

Second Edition
Figure C-29 Repair strategy selection menu.
For realizations that result in structural collapse, the program equates the
repair time to the user supplied total replacement time for the building.
In Figure C-30 and Figure C-31, the bottom graph shows the repair time
probability and the top chart shows the total repair time and contribution of
each performance group to the total repair time.
Figure C-30 Repair time output assuming floors are repaired in parallel.

Second Edition
Figure C-31 Repair time output assuming floors are repaired sequentially.
C.7.1.4 Environmental Impacts
For each scenario/intensity, the probability of various environmental impacts

for the building is displayed, as shown in Figure C-32. The desired
scenario/intensity number can be selected using the drop down list at the top
left of the form.
Figure C-32 Environmental Impacts tab.
C.7.1.5 Unsafe Placards Tab
For each scenario/intensity, the probability of various performance groups

triggering an unsafe placard for the building is displayed, as shown in Figure

Second Edition
C-33. The desired scenario/intensity number can be selected using the drop
down list at the top left of the form.
Figure C-33 Unsafe Placards tab.
C.7.1.6 Realizations Tab
The Realizations tab provides a graphical snapshot of results of each

realization performed by the program. As shown in Figure C-34, there are
four categories to choose from: (1) repair cost; (2) repair time; (3) number of
fatalities; and (4) number of injuries.
Figure C-34 Realizations categories.
PACT displays the consequences reported by each realization and the

breakdown of each consequence into contributing performance groups. The
realizations are ordered in the ascending order of the consequence, not the
number assigned to the realization. For example, Figure C-35 shows the
total cost and repair time reported by realizations. Note that only about 15 of
the 200 realizations resulted in structural collapse, another 15 had irreparable
residual drift. There is a big difference in cost between those 30 realizations
and those that just had component damage. The total replacement time is
reported in Figure C-35 for realizations reporting structural collapse or
irreparable residual drift.

Second Edition
Cost
Time
Figure C-35 Individual Realization display of repair cost (top) and repair time
(bottom) contributions by performance group.
C.7.1.7 Data Drill Down and Exports Tab
This Data Drill Down And Export Tab allows for examination or export of
results from a single realization or export of results of all realizations in a

Second Edition
comma separated value (CSV) format compatible with Microsoft Excel and
most other spreadsheet programs.
To examine results obtained from individual realizations, clicking

will display the Results Drill Down window. On this window a particular
intensity or scenario, a realization, a particular floor and direction, a
performance group, and even a particular item within the performance group
can be examined for more information.
The table on the top of the window shown in Figure C-36 provides a
summary for Intensity/Scenario #1, Realization #1, Direction 1, and
Performance Group B1035.001. Clicking will allow even
deeper examination.
Figure C-36 Result Drill Down screen.
There are two choices for exporting the results of realizations. High-level
exports produce files containing more significant information while low-
level exports contain every bit and piece of the information generated by the
program. The available options are shown in Figure C-37.

Second Edition
Figure C-37 Screen showing options for exporting results.
C.7.2 Time-Based Results
Selecting the button on the Scenario/Intensity Results window

shown in Figure C-22, will initiate the time-based results window, shown in
Figure C-38. This window has two tabs that show time based results and loss
area charts. The time based results tab is split into 6 sub tabs that show
annualized total values for repair cost, repair time, fatalities, injuries, carbon
emissions, and embodied energy. The shaded colors in each graph represent
the contribution of various intensities to the total annual loss.

Second Edition
Figure C-38 Time-based results output screen.
Clicking on the legend of any of the charts causes the chart to highlight the
selected data by darkening the color of other data. A click on the top of the
legend box restores the original colors.
By default, the annualized total and graphs are the integration of the data bins
from each analysis. In order to view annualized totals integrated using the
lognormal fitted curves, the option can be selected as shown in Figure C-39.
Figure C-39 Options menu showing base curve mode selection.
When integrating using the data bins, the number of data bins can have an
effect on the totals and the graphs. The number of bins used can be adjusted
in the Options menu, as shown in Figure C-40. The number of bins actually
used will depend on the exact data being evaluated, but it will never be larger
than the number specified by the user.
Figure C-40 Options menu showing histogram bins option.
The graph type (either color coded by the contribution of each intensity, or as
a single curve showing the total amount) can be changed with the radio
buttons on the bottom left of the form.

Second Edition
Figure C-41 Average annualized loss output screen.
On the left side of the form the annualized probabilities of structural collapse
and unsafe placard are displayed.
Loss Area Charts tab provides a three-dimensional representation of the

charts in their unadjusted form or as adjusted by application of mean annual
frequency of exceedance values. Figure C-42 shows the different options
available for display.
Figure C-42 Loss area charts buttons.
Figure C-43 Loss Area Charts tab.

Second Edition
C.8 Fragility Specification Manager
Attention is needed when experimenting with fragility functions definitions,

because any changes made here can radically change the outcome of the loss
analysis. It is recommended that a backup of the “\ATCCurves” folder
originally provided with PACT be made before making any changes to the
fragility functions.
The Fragility Specification Manager utility allows editing of fragilities and

creation of new fragility specifications.
The File menu provides the option to create a new fragility specification
from scratch, save a fragility specification. This menu is shown in Figure
C-44. In the Edit menu, an option to duplicate an existing fragility
specification is given. The duplicate fragility would be used as the basis for
a new fragility. The Tools menu provides the ability to switch default
measurement system between imperial and metric units.
Figure C-44 File, Edit, and Tools menus.
C.8.1 Overview Tab
The Overview tab, shown in Figure C-45 provides a list of all the fragility
specification categories in a tree structure, along with all the existing
fragilities. Any category that has one or more fragilities underneath it will be
marked as bold. Any subcategory can be expanded by clicking the small
symbol, or hidden by clicking the symbol.

Second Edition
Figure C-45 Overview tab.
The search box on the top can be used for searching all the existing fragilities
for an ID or description containing that text.
The fragilities are organized with the following:

• Top Level Category:
o A – Substructure
o B – Shell
o C – Interiors
o D – Services
o E – Equipment & Furnishings
o F – Special Construction
• 2nd Level Category:
o Top level followed by 2 numbers from 10 to 99, e.g., A10
• 3rd Level Category:
o 2nd Level followed by one number 1-9, e.g., A101
• 4th Level Category:
o 3rd Level followed by one number 1-9, e.g., A1011
In addition, a specific fragility classification number would include all four

levels and an optional letter, e.g., A1011.000 or A1011.555v.

Second Edition
C.8.2 Fragility Specification Details Tab
At the top of the Fragility Specification Details tab, shown in Figure C-46,
is a navigation bar, which lists the fragility specification being examined. It
also allows for navigation through all of the existing fragilities using the
arrows. It also provides buttons to add a new fragility and delete an
existing fragility , and save an existing fragility .
Below the navigation bar, the information on the current fragility, divided
between a tree view on the left showing the basic structure of the fragility
and its damage states and consequences, and detail information on the right,
is displayed. The border between the tree view and the detail information
can be moved left or right by clicking and dragging on the border between
them.
Figure C-46 Fragility Specification Details tab.
When a new fragility is added or an existing one is duplicated, the Uniformat

ID number will be requested. The format of the ID field that the fragility
manager expects is a single upper case letter between A-F followed by two
digits from 10 to 99, two more digits from 11 to 99, followed by a period,
then 3 more digits from 000 to 999, followed by an optional lower case letter
from a to z.

Second Edition
C.8.2.1 Fragility Structure Tree View
The fragility structure consists of general information for the fragility and
various damage state groups and sub-damage state groups, individual
damage states, and consequences.
The easiest structure is that of a sequential fragility. It consists of general

information, a single damage state group (designated with the title of
“Damage State Type”), one or more individual damage states, and
consequences for each of those damage states.
The structure of a mutually exclusive or simultaneous fragility is more

complicated. In either case, a “master” sequential damage state group
(designated with the title of “Damage State Type”) with only a single
damage state is provided. This damage state is used to determine if the item
in question is damaged at all based on the value of the demand parameter of a
particular realization. If it is determined that there is some damage, the sub-
damage state group is marked as either simultaneous or mutually exclusive,
with two or more damage states to determine the exact kind of damage, with
consequence functions for each damage state. Both types of fragilities are
illustrated in Figure C-47.
(a) (b)
Figure C-47 Tree view showing: (a) sequential; and (b) simultaneous
fragilities.
PACT does not limit the user to these options. It is possible to have multiple
damage state groups of all different types nested underneath each other, but
the utility of these options is limited. Two examples are shown in Figure
C-48.

Second Edition
Figure C-48 Example complicated fragility structures consisting of multiple sub-damage
state groups and sub-sub-damage state groups of several different types.
A new fragility can be created by clicking or using the File menu. The
tree structure will look as shown in Figure C-49.
Figure C-49 Basic damage state level.
The program provides general info and a master damage state by default.
When Add Damage State is clicked it will add Damage State 1.
If a simple sequential fragility function is being created, more damage states

can be added by clicking on Add Damage State. Clicking on Consequence
Functions for each damage state will allow adding of consequence
information, as shown in Figure C-50.
Figure C-50 Consequence functions.
Any damage state, consequence, or damage state group (except the very top
level one), can be deleted by right clicking on the item and choosing Delete
or selecting it and hitting .
In order to create a mutually exclusive or simultaneous fragility, first the top-
level damage state needs to be created, then a damage state group can be
added. In Figure C-51, note that there are two places that say Add Damage
State. The first is connected by a line to the master damage state group, and
is the same level as Damage State 1. That one will add a Damage State 2 to

Second Edition
the master group. The second one is under the level of the “Sequential
Damage State Group,” and will add a damage state to this subgroup. When
this damage state is selected, the selection in the combobox shown in Figure
C-52 can be changed to Mutually Exclusive or Simultaneous. Selecting
“Add Damage State” that is right under the Sequential Damage State Group
text changes the text to say “Mutually Exclusive Damage State Group,” with
an option to add the first damage state to it. This is shown in Figure C-53.
Figure C-51 Adding damage state to different groups.
Figure C-52 Combobox for selecting fragility type.
Figure C-53 Creating mutually exclusive damage states.
C.8.2.2 General Info Tab
When General Info is selected on the tree view, the screen shown in Figure
C-54, separated into two tabs on the right, is displayed. The General Info
tab will show the most important general information about the fragility.
The Fragility ID is shown, but is not editable, along with the short name and
a longer description of the fragility. The demand parameter can be selected
from a dropdown menu, or created new, as shown in Figure C-55. In
addition, the fragility can be specified as directional or non-directional, and
correlated or not correlated.
Fragilities can be marked as using the demand values of the floor above, by
using the checkbox marked Use Demand Value from Floor Above. This is
used for components such ceiling tiles or sprinkler pipes, which affect the
casualties of the floor they are on, but use the accelerations of the floor above
them to determine their damage state.
Some fragilities will require user-entered values, and are therefore

incomplete. These fragilities are marked with the checkbox labeled “User
Supplied Data Needed.” The user is expected to edit these fragilities for their

Second Edition
specific building, depending on the exact nature of the component. For
instance, there are fragilities for computers that have the demand values to
determine if the computer is damaged, but not the cost values because that
differs based on the specific circumstance. The program will provide a
warning if these fragilities are being used without entering in the required
values.
Figure C-54 General Info tab on Fragility Specification Details screen.
Figure C-55 Create New Demand Parameter window.
C.8.2.3 Notes Tab
The Notes tab is shown in Figure C-56 and contains miscellaneous

information about the fragility. Included here is the date the fragility was
created, whether it is an official and approved fragility, the rating of its
quality, the author information, and any other notes.

Second Edition
Figure C-56 Notes tab on Fragility Specification Details screen.
C.8.2.4 Damage State Group
When any Damage State Group is selected on the tree view, including the top
level group named “Damage State Type,” a screen on the right will be
displayed. The information consists of a dropdown list showing the damage
state group type (Figure C-57), and a graph of all the damage states in that
group. If it is the top level group, the fragility type cannot be changed. If the
damage group type is sequential, a line graph showing the probability of each
damage state against the possible demand parameter values will be shown
(see Figure C-58).
Figure C-57 Damage State Group Type dropdown menu.
If the damage state group type is simultaneous, a bar graph with the
probabilities of its damage states will be displayed, as shown in Figure
C-59a. If the damage state group type is mutually exclusive, a pie chart with
the probabilities of its damage states will be displayed, as shown in Figure
C-59b.

Second Edition
Figure C-58 Damage State Group screen.
(a)
(b)
Figure C-59 Fragility data display showing:
(a) simultaneous damage state
group; and (b) mutually
exclusive damage state group.

Second Edition
C.8.2.5 Damage State
When any of the damage states on the tree view are clicked, a screen with
information about that damage state will be displayed, as shown in Figure
C-60. This screen provides a description, an image if one has been specified,
and buttons to allow for deletion or addition of an image. If the selection is
sequential (or a master damage state group), the median and dispersion
values are displayed. If the selection is simultaneous or mutually exclusive
the fraction of the time it is in this damage state is displayed.
Figure C-60 Damage state information screen.
C.8.2.6 Consequence Functions
When any consequence function on the tree view is clicked, a screen with
different consequence information, separated into four tabs, is displayed, as
shown in Figure C-61.
In the General Consequence Info tab, a description of the repair measures

for this damage state is provided.
Figure C-61 Consequence Functions screen.

Second Edition
In the Repair Cost Consequence tab, shown in Figure C-62, information for
costs and cost discounts for this damage state are listed. The lowest quantity
at which cost discounts occur and the average cost per unit to repair at that
quantity or lower, and the quantity above which there will no longer be cost
discounts, and the average cost to repair per unit at that cost or higher should
be entered. The cost uncertainty type can also be entered, and depending on
if normal or lognormal distribution is chosen, and the dispersion or COV
value associated with the uncertainty.
Figure C-62 Repair Cost Consequence tab.
In the Repair Time Consequence tab, shown in Figure C-63, the

information for time and time discounts for this damage state are entered.
In the Other Consequences tab, shown in Figure C-64, information on the

unsafe placard consequences, non-collapse casualty consequences, long lead
time flags, and environmental losses are provided.
In this tab, use of unsafe placard consequences can also be specified. If not,
the fraction of the total quantities required to trigger a unsafe placard need to
be entered. PACT will count all the items of this type in the building, and
count how many of those items are in this damage state. If the ratio is higher
than the median value entered here (adjusted by the dispersion) for a
particular realization, it will be marked as getting an unsafe placard. It will
not add different damage states together, even different damage states for the
same fragility.

Second Edition
Figure C-63 Repair Time Consequence tab.
Figure C-64 Other Consequences tab on Fragility Specification Details

screen.

Second Edition
Non-collapse casualties can also be defined here by first entering information
on how much each floor is affected by an item in this damage state, then
what the fatality and serious injury rate are for anyone in that affected area.
Finally, items with a long lead time flags in this damage state can also be
indicated here.
C.9 Population Manager
Attention is needed when experimenting with the population library, because

any changes made here can change the outcome of the casualty analysis. It is
recommended that a backup of the population library originally provided
with PACT be made before making any changes to the occupancy functions.
The Population Manager is divided into two separate areas, as shown in

Figure C-65. The layout is very similar to the population manager in the
building manager, except that changing the values here will change them for
all future projects.
Figure C-65 Building Population Manager window.
To create a new population model, New can be selected from the File menu.
Fields are provided for entering the name of the population model, the peak
number of occupants per 1000 square feet, and the dispersion, population

Second Edition
distribution per month, day of the week and hour of the day. These may be
viewed graphically by clicking on the graph tab (see Figure C-66).
Figure C-66 Population model data screens.
C.10 Reporting
The reporting screen, shown in Figure C-67, will take project or result files,
and a Microsoft word template, and create a Microsoft Word report based on
them. Two templates are included, one to be used for a Project file (showing
the echo of the inputted information) and one to be used for a results file
(showing the inputted information and the results).
Figure C-67 Create Report window.

Second Edition
The Tools menu provides two sets of options for image size: (1) “Size of
created images (pixels)”; and (2) “Size of images in Word Document
(Inches).” The first option is the size that created images (i.e., graphs) will
be in the Word document. The default values generally allow for 2 graphs to
fit on one page. The second option alters the size of images that are created
internally before they are outputted to the Word document, and the size has
several effects on the output. As the resolution of the images increases, the
graphs get clearer and less blocky, but the fonts do not scale the same way,
and so begin to look smaller. It is recommended to use smaller values (200-
800) for a report that will most likely be viewed on the computer, and larger
values (1000+) when creating reports to print.
Regardless of what values are used for these two options, the aspect ratio (the
ratio of width/height) should be the same or very close, or the image will
look distorted.
C.11 A Look under the Hood
All data files reside in the Common Application Data folder. For Windows
XP, the default location of that folder is “C:\Documents and Settings\All
Users\Application Data\Pact,” and for Windows 7 it is “C:\Documents and
Settings\All Users\Pact.”
C.11.1 Project Files
All building and project information inputted for a particular project are
saved in a single XML file with the project name in the Projects subfolder
under the folder where PACT is installed. If the default path was used for
installing PACT, the project files may be found in the data folder in
“Projects.” Although XML files are text files that can be viewed and
changed using any text or word processor, extreme care is necessary if these
files are manipulated, as even minor manual changes may render a project
file unusable. A copy of the project file can be easily created with no
adverse effects.
C.11.2 Results Files
All result information if saved, is saved in a single file with the result name
in the Projects subfolder under the folder where PACT is installed. If the
default path was used for installing PACT, the project files may be found in
data folder at “Projects.” Result files can be saved as binary files, with the
file extension .bin or as XML files with the file extension XML. Binary files
are smaller, typically half the size of XML files, and much faster to load and
save. Results files can be quite large, so this is not a trivial concern.

Second Edition
However, binary files can only be read by the PACT Result Viewer. In
addition, binary files have been reported to cause out of memory errors when
being saved or loaded. XML files, while much larger, can be read by any
text editor, and can easily be parsed and manipulated by other programs.
Either file format contains the original project file, all the fragilities used by
that project, and all the result information, including every single analysis,
realization, performance group, damage state, and even every random
number used in any calculation.
C.11.3 Fragility Specification Files
Fragility specifications are directories and XML files that bear the same
identifier as the fragility specification they represent. For each fragility
specification, there is a folder with the same name as the fragility ID,
containing an XML file and any representative photos of the damage states
associated with the fragility function. These folders are stored in a subfolder
under the folder where PACT is installed. If the default path was used for
installing PACT, these files may be found in the data directory in
“ATCCurves.” PACT will show any proper fragility specification stored in
this subfolder.
C.11.4 Population Specification Files
Population specification files are saved in XML files bearing the name of the
population specification. They are stored in a subfolder under the folder
where PACT is installed. If the default path was used for installing PACT,
these files may be found at “Templates\Occupancy.”
C.11.5 Reporting Template Files
A Word 2007/2010 file is used as a template. The file must be in the DOCX
format; older Word DOC file types will not work. The default location for
the template files is in the data folder at “Reports.” It uses mail merge fields
in the template as placeholders for putting information from the actual
project. Each merge field in the body and header/footer will be reviewed,
and if it is in the proper format, will be replaced by values from the Project or
Results file. The fields are too many too list out, but the sample templates
have all of the available fields, with one exception: If all the fragilities used
want to be shown, the field “Special:AllFragilities” should be selected. If
only fragilities that are marked as unofficial want to be shown, the alternate
merge field of “Special:UnofficialFragilities” can be used.

Second Edition
C.11.6 Programming Notes
PACT is coded using Microsoft Visual Studio 2010 in C#. Besides the
functions and utilities contained in the above platform the following third-
party controls have been used:
Table C-1 PACT Third Party Controls

Control Name Vendor Link
Chart for .NET v. 9.11.3.12 Nevron Software http://www.nevron.com/
used under commercial license
Math.NET Iridium Numeric http://www.mathdotnet.com/
LPGL open source license Foundation
Log4Net Apache Logging http://logging.apache.org/

Apache open source license Services
C.11.6.1 Logfiles
The default location for logfiles is the Project directory, located under the
PACT main directory. If for any reason they cannot be written to that
location, Local Application Data path, which defaults to C:\Documents and
Settings\%username%\Local Settings\Application Data in Windows XP or
C:\Users\%username%\AppData\Local in Windows 7 will be used, where
%username% is the current username. When reporting bugs, the logfile
should be sent to aid with tracking down the bugs.

Second Edition
Appendix D
Normative Quantity Estimation
Tool User Manual
D.1 Introduction
The Normative Quantity Estimation Tool is provided in Volume 3 to assist

the user when establishing the quantities of various building components for
entry into the building performance model. The tool is limited in its
functionality and a detailed user review of the output is necessary.
Normative quantities are an estimate of the quantity of components and

contents likely to be present in a building of a specific occupancy on a gross
square foot (gsf) basis. The tool requires as input a listing of the floors of the
building and the gross area per floor. The tool produces a floor-by-floor list
of the suggested component types and the quantities of those components for
input into PACT. To facilitate the building definition, the tool includes eight
occupancy types (same as those provided in PACT) with default quantity
data for the components commonly appearing in that occupancy.
D.2 Usage Notes
The Normative Quantity Estimation Tool contains several worksheets

displayed as tabs. These are the following:
• Notes. This tab summarizes important development information about
the tool.
• Normative Quantity Estimate. This tab is used for inputting relevant
information and obtaining the output.
• Normative Quantity Database. This tab contains the tables which can
be used to modify the output of the tool in accordance with the specific
building being modeled.
• Alternate Component Matrix. This tab presents normative quantities
for 10th or 90th percentile values for use in the tool.
• Name Manager. This tab presents the file hierarchy used in the PACT
Fragility Specification Manager utility.
• Fragility Database. This tab contains a copy of the Fragility Database.
FEMA P-58-2 D: Normative Quantity Estimation Tool User Manual D-1

Second Edition
In each of the tabs, the user should only change cells with color shading and
select options with the radio buttons. Unshaded (white) cells are calculated
or created as the tool executes and thus should not be changed or moved.
Also, columns or rows of data should not be added or deleted except where
specifically noted.
Pop-up notes with comments have been included to further explain options
within the workbook. A red triangle at the upper right of a cell indicates the
existence of a pop up note. Hovering over the cell will display the note.
The tool uses Excel macros. The macros must be enabled in order for the
tool to run. The spreadsheet was created in Excel Office 2010 version
14.0.6106.5005 as a macro enabled Excel file (.xlsm).
D.3 Normative Quantity Estimate Tab
D.3.1 Building Definition Table
Within the Building Definition table each floor is defined by its total floor
area and up to three occupancies. To define a building, the user first inputs
the name of each floor (starting at the roof) as well as the total floor area on
each floor. The user then selects the occupancy types present on each floor
and assigns a percentage of the total floor area to each occupancy type. If the
percentage values do not add up to 100% over three occupancies, the column
to the right of the Building Definition table will issue a warning. The user
must also provide a floor number for later reference in the output summary.
The floor name can be any alphanumeric entry; the floor number must be an
integer value starting at 1 at the first floor.
D.3.2 Component Summary Matrix
The “Compile” button executes the tool and populates the Component
Summary Matrix with a floor-by-floor list of component types for input
into PACT. The output matrix is ordered by floor, then occupancy type. The
data can be organized and filtered using the dropdown options provided in
the column headers.
The Fragility Database provides a unit of measure for each fragility group,
listed here as “Assumed Quantity per component within PACT.” The
“Actual Quantity” value reported recognizes this assumed PACT component
quantity and multiplies it by the user input building occupancy area and per
occupancy unit quantity defined in the Building Occupancy Component
Matrix provided in the Normative Quantity Database tab. The “Actual
Quantity” value should be used to confirm that the components selected are
appropriate given the building size and quantity of discreet units required.
D-2 D: Normative Quantity Estimation Tool User Manual FEMA P-58-2

Second Edition
The Component Summary Matrix lists output that is based upon default
fragilities and quantities. The user can make more specific selections using
the two tables provided in the Normative Quantity Database tab, discussed
in Section D.4.
The output listed in the Component Summary Matrix can be revised by the
user to return quantities specific to the building studied by using the
additional features listed in Section D.3.3 or the information provided in the
Normative Quantity Database tab.
D.3.3 Additional Features
The Normative Quantity Estimate tab contains four additional data

processing features. These features may be enabled using the ratio buttons
next to the “Compile” button. Each feature is described below.
D.3.3.1 Processing of Incomplete Data
This option may be selected to allow incomplete data to be eliminated from

being listed in the Component Summary Matrix. Selecting “Ignore
Incomplete Data” will consider any component types with a unit mismatch
and eliminate them from the output matrix. This most often occurs when a
component has not been assigned to the Building Occupancy Component
Matrix, or if a component requires cost and consequence data entered by the
user. These components require case-by-case consideration to determine
quantity and consequence values.
D.3.3.2 Processing of Zero Data
The tool can be set to check if the occupancy category selected produces a
component with zero quantity. If the user has chosen to “Exclude Zero
Data,” any component types producing zero quantity will not be listed in the
Component Summary Matrix when the tool is run.
D.3.3.3 Summation of Fragility by Floor
When considering a building with multiple occupancies, certain components

such as generators, cooling towers, or elevators may be more appropriately
counted if the component is summed across a floor or across the building and
input to PACT as single component. For this reason, the tool has the option
of summing individual components on a floor-by-floor or total building
basis.
On the Fragility Database tab, column AB identifies each component to be

classified for summation according to the following:
• NO – The component will not be summed.

Second Edition
• FLR – The component will be summed on a floor-by-floor basis.
• BLDG – The component will be summed across the entire building.
The Normative Quantity Estimate tab contains an option to sum the

components according to their designation in column AB of the Fragility
Database tab:
• Sum Components by Building and Floor – Components designated as
FLR or BLDG are summed on a floor-by-floor basis or across the entire
building, respectively.
• Sum Components by Floor – Components designated as FLR or BLDG
are summed on a floor-by-floor basis. This option overrides the
summation across the entire building.
• List All Components – All components are listed by the floor and
occupancy by which they were defined. The summary output is not
summed.
Depending upon the summation method chosen by the user, the Occupancy
Type and Floor Level of the component will be listed within the Component
Summary Matrix as according Table D-1.
Table D-1 Summation Presentations in the Component Summary Matrix
Component Summary Matrix

Occupancy Type and
Summation Type Number Floor Number
Sum by Building and Floor ALL ALL
Lists the level on which the
Sum by Floor ALL
component exists.
Lists the occupancy the Lists the level on which the
List All Components
component belongs to. component exists.
“ALL” is reported when a component has been summed over multiple

occupancies or floors. When entering a summed component into PACT, the
user should place the summed component in the most appropriate occupancy
or floor given the component type. For example, mechanical units or
distribution panels are likely to be placed in a mechanical room on each floor
and the sum of all components should be assigned to the floor in question,
given an occupancy type appropriate to the room. Components summed on a
total building basis, such as cooling towers or generators should be assigned
to the floor where this equipment will be installed, and given an occupancy
appropriate to this space.

Second Edition
D.3.3.4 Rounding to Integer Quantity
Certain components such as stairs, elevators, and certain mechanical

equipment will be installed only in discreet quantities. On the Fragility
Database tab, column AA identifies fragilities that can only be installed in
discreet quantities, such as stairs, elevators, and certain mechanical
equipment. When “Round to Integer Unit” radio button is selected, the tool
will round the calculated component quantity up to the next integer value.
For comparative or study purposes, the user may unselect this option. This
setting supersedes the rounding setting noted in the Fragility Database tab.
If rounding is disabled, the resulting estimates will include fractional
quantities of the components and the user will have to evaluate if some level
of rounding is appropriate.
D.4 Normative Quantity Database Tab
The Component Summary Matrix on the Normative Quantity Estimate

tab lists output that is based upon default fragilities and quantities. The user
can make more specific selections using the two tables discussed below.
D.4.1 Component Definition Table
This table presents the selection of fragility groups referenced by the tool.
The Component Definition Table is shaded in blue and thus allows for
changes by the user.
For example, as a default, four fragility group selections are provided for
D2021.011a (piping fragility for cold water piping with a diameter less than
2.5 inches, designed for Seismic Design Category A or B). The user who has
information that the building contains large diameter welded steel piping
designed for Seismic Design Category D, E, or F, can use the dropdown list
available at the end of column E and select the appropriate fragility group
(D2051.023a). If this change is made before pressing the “Compile” button
to execute the tool, the output in the Component Summary Matrix will list
the quantity relevant to D2051.023a.
While making selections, it is important to retain consistency with the

description of the component category and relevant component unit quantity
and value listed in columns K through M in the Building Occupancy
Component Matrix. Note that some component categories listed in the
Component Definition Table do not have any fragilities associated with
them in the database. These are identified with “none available” in column
E.

Second Edition
In some cases, multiple similar components exist within the PACT database
providing various per component quantities. For example, chillers are
defined with varying levels of cooling between 350 tons and 1000 tons.
Once the tool has run, the user should review the resulting discreet quantity
components; an alternate component may exist with better agreement with
what may actually be used in the building.
D.4.2 Building Occupancy Component Matrix
Buildings of similar type and occupancy were reviewed during the

development of this tool and the quantity of each component per occupancy
type was bracketed into 10th, 50th, and 90th percentiles. The tool assumes 50th
percentile values by default in the Building Occupancy Component
Matrix. If the user wishes to use values other than the default values, the
component quantity values can be the edited in the Building Occupancy
Component Matrix. 10th, 50th, and 90th percentile values of normative
quantities are provided in the Alternate Quantity Data tab for reference.
This matrix also contains entries for some components, such as fixed
casework and fume hoods, that do not have provided fragilities.

Second Edition
Glossary
Definitions
Annual frequence of exceedance. The average number of times per year
that an event having magnitude larger than a reference value is likely to
occur (typically less than 1).
Annualized loss. The average value of loss, per year, over a period of many
years.
Backbone curve. A component force-deformation relationship that
envelopes either monotonic or cyclic loading response.
Building performance model. An organized collection of data necessary to
define building assets that are at risk and vulnerable to the effects of
earthquake shaking.
Capacity boundary. A component force-deformation relationship that
represents the maximum strength that a component can develop at a given
level of deformation; generally equal to the response of the component to
monotonic loading.
Casualties. Loss of life, or serious injury requiring hospitalization.
Collapse floor area ratio. The portion of floor area at each level that is
subject to collapse in each collapse mode.
Collapse fragility. A mathematical relationship that defines the probability
of incurring structural collapse as a function of ground motion intensity.
Collapse mode(s). One or more ways in which a building would be
expected to collapse, ranging from partial to complete collapse. Possible
collapse modes include single-story collapse, multi-story collapse, or total
collapse.
Component. One of many parts, both structural and nonstructural, that
together comprise a building.
Conditional probability. The probability that an outcome will occur, given
that a particular event occurs or condition exists.
Consequences. The losses resulting from earthquake damage in terms of
potential casualties, repair and replacement costs, repair time, and unsafe
placarding.
FEMA P-58-2 E: Glossary E-1

Second Edition
Consequence function. A relationship that indicates the potential
distribution of losses as a function of damage state.
Core and shell replacement cost. The cost to replace the basic building
structure, exterior enclosure, and mechanical, electrical, and plumbing
infrastructure that is typically present in a building before tenant
improvements are made, including costs associated with demolition of the
damaged building and clearing the site of debris.
Correlation. A mathematical relationship that defines the extent to which
the value of one parameter is dependent on the value of one or more other
parameters.
Cyclic envelope. A component force-deformation relationship that
envelopes the hysteretic behavior of a component in response to cyclic
loading.
Damage state. For a particular component, or the building as a whole, a
condition of damage associated with a unique set of consequences.
Decision-maker(s). An individual or group responsible for selecting the
performance objectives for a performance-based design.
Demand. A parameter that is predictive of component or building damage
states, including peak floor (or ground) acceleration, peak component
deformation, peak (or residual) story drift, peak floor (or ground) velocity, or
peak component force (or stress).
Discount rate. A factor used to indicate the time-dependent value of money
in economic analyses.
Dispersion. A measure of uncertainty associated with prediction of the true
value of a random, or otherwise uncertain, behavior.
Distribution. A mathematical function describing the statistical probability
that elements in a set will have specific values.
Earthquake scenario. A specific earthquake event, defined by a magnitude
and geographic location (distance) relative to a building site.
Floor acceleration. At a floor level, the acceleration of the center of mass
relative to a fixed point in space.
Floor velocity. At a floor level, the velocity of the center of mass relative to
a fixed point in space.
Fragility function. A mathematical relationship that defines the probability
of incurring a damage state conditioned on the value of a single demand
parameter.
E-2 E: Glossary FEMA P-58-2

Second Edition
Fragility group. A set of similar building components having the same
potential damage characteristics in terms of fragility and consequence
functions.
Fragility specification. A detailed description of potential component
damage states, fragility functions, and consequence functions.
Fragility unit of measure. The basic unit used as a basis for defining
quantities in a fragility group for the purposes of estimating damage and
consequences. Components in a fragility group can be defined in individual
units of “each,” units of lineal feet (e.g., 1000 LF), units of square feet (e.g.,
250 SF), or other units.
Geometric mean (geomean). In characterizing ground motion intensity, the
square root of the product of the value of a ground motion parameter in each
of two orthogonal directions (i.e., xy ).
Intensity. The severity of ground shaking as represented by a 5%-damped,
elastic acceleration response spectrum, or the 5% damped spectral response
acceleration at a specific natural period of vibration.
Intensity-based assessment. An assessment of the probable performance of
a building subjected to a ground motion of a specified intensity.
Lognormal distribution. A distribution of values, characterized by a
median value and dispersion, with the property that the natural logarithm of
the values is normally distributed.
Mean. The average value of a parameter (or population of values). If the
population of values is normally distributed, the mean will be equal to the
median. If the distribution is skewed, the mean may be less than or greater
than the median.
Mean recurrence interval. The average amount of time between expected
repeat occurrences of an event (e.g., an earthquake scenario or loss), typically
expressed in years.
Median. The value of an uncertain parameter (or population of values) that
will be exceeded 50% of the time.
Mutually exclusive damage states. A series of related damage states in
which the occurrence of any one type of damage will preclude the occurrence
of another type of damage.
Net present value. For a particular discount rate, the present value of
benefits that will be received in the future, less the associated costs of these
benefits, considering the time-dependent value of money.

Second Edition
Normal distribution. A symmetric distribution described by a mean value
and standard deviation. When plotted as a probability density function, a
normal distribution has the shape of a bell curve.
Normative quantity. An estimate of the quantity of nonstructural
components and contents likely to be present in a building of a specific
occupancy on a gross square foot basis.
Nonstructural component. A building component that is not part of the
structural system.
Performance. The probable damage and resulting consequences as a result
of earthquake shaking or other hazards.
Performance function. A statistical distribution that indicates the
probability that losses of a specified or smaller magnitude will be incurred as
a result of future earthquakes.
Performance group. A subset of components described by a single fragility
group that will experience the same demand in response to earthquake
shaking.
Performance group quantity. The number of fragility units of measure
present in a building, in the performance group under consideration.
Performance measure. A means of quantifying the consequences
associated with earthquake shaking in terms that are meaningful to decision-
makers. Performance measures include casualties, repair and replacement
costs, repair time, and unsafe placarding.
Performance objective. An expectation as to the probable damage state and
resulting consequences that a building may experience in future earthquakes.
Period. The time, in seconds, necessary for a structure to complete one
cycle of motion in free vibration.
Probable Maximum Loss. The probable cost of repairing earthquake
damage expressed as percentage of building replacement cost with a
particular confidence of non-exceedance, conditioned on a particular
earthquake return period.
Realization. One possible performance outcome for a particular earthquake
scenario or intensity comprising a unique set of demands, damage states, and
consequences.
Repair cost. The cost, in present dollars, necessary to restore a building to
its pre-earthquake condition, or in the case of total loss, to replace the
building with a new structure of similar construction.

Second Edition
Repair fragility. A mathematical relationship that defines the probability
that a building will be considered irreparable as a function of residual drift
ratio.
Repair time. The time, in weeks, necessary to repair a damaged building to
its pre-earthquake condition.
Replacement cost. The cost, in present dollars, necessary to replace a
building that has been damaged beyond the point of practicable repair,
including costs associated with demolition and removal of debris.
Replacement time. The time, in weeks, necessary to replace a building that
has been damaged beyond the point of practicable repair, including time
associated with demolition and removal of debris.
Residual drift. The difference in permanent displacement between the
center of mass of two adjacent floors as a result of earthquake shaking.
Residual drift ratio. Residual drift divided by story height.
Return on investment. The annual income that can be derived from an
investment divided by the present value of the investment.
Return period. See mean recurrence interval.
Rugged. Not vulnerable to damage, or having a very high threshold for
damage; alternatively, not considered significant for assessing seismic
performance impacts.
Scenario-based assessment. An assessment of the probable performance of
a building subjected to a specified earthquake scenario.
Scenario Expected Loss. A probable maximum loss with mean chance of
occurrence.
Scenario Upper Loss. A probable maximum loss with a 10% chance of
exceedance.
Seismic hazard curve. A plot of the annual frequency of exceedance of the
possible values of a ground shaking intensity parameter.
Sequential damage states. A series of related component damage states that
must occur in sequence following the occurrence of an initial damage state.
Simulated demands. A vector set of demands representing the building
response in a particular realization, reflecting the correlation between the
various response quantities predicted in the structural analyses and
uncertainties inherent in structural response prediction.
Simultaneous damage states. A series of independent and unrelated
component damage states that can occur, but need not necessarily occur, at
the same time.

Second Edition
Stakeholder(s). An individual or group affected by the performance of a
building, including building owners, lenders, insurers, tenants, and the
general public.
Story drift. The instantaneous difference in lateral displacement between
the center of mass of two adjacent floors.
Story drift ratio. Story drift divided by story height.
Structural component. A building component that is part of the intended
gravity or seismic-force-resisting system, or that provides measurable
resistance to earthquake shaking.
Structural system. A collection of structural components acting together in
resisting gravity forces, seismic forces, or both.
Time-based assessment. An assessment of the probable performance of a
building, considering all earthquake scenarios and the associated probability
of occurrence of each, over a specified period of time.
Total loss threshold. A user-defined cap on the level of repair effort at
which a building is likely to be replaced rather than repaired.
Total replacement cost. The cost to replace an entire building (core, shell,
and tenant improvements) as it exists in its pre-earthquake condition,
including costs associated with demolition of the damaged building and
clearing the site of debris.
Unsafe placarding. A post-earthquake inspection rating that deems a
building, or portion of a building, damaged to the point that entry, use, or
occupancy poses immediate risk to safety. Unsafe placards are typically
colored red.
Yield drift. The value of drift at which a structure reaches its effective yield
strength.
Yield story drift. The value of story drift at which a story in a structure
reaches its effective yield strength.
Yield story drift ratio. Yield story drift divided by story height.

Second Edition
Symbols
Notation
ai* in the simplified analysis method, estimated median peak floor
acceleration at level “i”
b slope of a straight line fit to a series of data having x and y values
c y intercept of a straight line fit to a series of data having x and y
values
di value of the demand at which a particular damage state was first
observed to occur in specimen “i” in a series of m tests, or the
maximum value of demand to which a specimen was subjected
without occurrence of the damage state
dj maximum demand experienced by a series of test specimens in bin
“j”
dmax maximum demand, d, experienced by a specimen in a series of tests
dmin for a series of tests of specimens, the minimum value of demand, d,
at which any specimen exhibited damage or distress
ei in time-based assessment the mid-range intensity within an
incremental range of intensities, ∆ei
fr rocking frequency of an unanchored object, (hz)
fs frequency associated with sliding of an unanchored object, (hz)
h building height (also designated as H)
hi in the simplified procedure, the height of level “i” above the building
base
i interest rate, or expected rate of return on money
k in the simplified procedure, a coefficient used to account for
nonlinear mode shape
mj number of specimens in bin “j” for which damage conforming to a
particular damage state is observed
t period of years over which an investment decision is amortized
FEMA P-58-2 F: Symbols F-1

Second Edition
vi* in the simplified analysis method, estimated median peak floor
velocity at level “i”
vsi reference floor velocity parameter used to determine estimated peak
floor velocity in the simplified analysis procedure
wj in simplified analysis, weight lumped at a structure’s level “j”
xj natural logarithm of d j
x mean value of a series of values, xj
yj inverse lognormal distribution value for the number of specimen
tests that experienced a damage state in a bin “j” of specimen tests
y mean of the values of yj
A amount of money that is earned or expended each year over a period
of several years
B parameter used to calculate the critical tipping demand on an
unanchored object
C1 in the simplified analysis procedure, a coefficient used to account for
the difference in displacement response between elastic and inelastic
structures, assuming non-degrading inelastic behavior
C2 in the simplified analysis procedure, a coefficient used to account for
the effects of stiffness degradation on displacement response
C3 coefficient used to determine the rocking capacity of an unanchored
object
Cq coefficient that adjusts computed median strength capacity for
construction quality
CR coefficient used to determine the rocking capacity of an unanchored
object
Csm coefficient used to derive an approximate estimate of the fraction of
total mass effective in first mode response
Cvx in the simplified analysis procedure, the fraction of the total pseudo
lateral force, applied to level “x”
Damp spectral velocity amplification factor
Fx in the simplified analysis procedure, the pseudo lateral force applied
to level “x”
H total building height above the base; also, the height of a wall panel
M total number of test specimens for which test data is available and for
which the damage state of interest occurred
F-2 F: Symbols FEMA P-58-2

Second Edition
MA in a series of M tests, the number of specimens that did not
experience any damage or distress but which experienced demands
greater than or equal to a reference value, da
MB in a series of M tests, the number of specimens that experienced
some damage or distress but which did not experience the damage
state of interest
Mj number of test specimens in a bin “j” of a series of tests
N total number of vulnerable stories in a building performance model
NPV net present value of a future stream of incomes or expenditures
PGA peak ground acceleration, (g)
PGV peak ground velocity, (in/sec)
PM10 value of a performance measure (e.g., repair cost) with a 10% chance
of exceedance
PM50 median value of a performance measure (e.g., repair time) with a
50% of exceedance
PM90 value of a performance measure (e.g., casualties) with a 90% chance
of exceedance
PM mean value of a performance measure, also known as the expected
value or average value
P(C) conditional probability of building collapse, given an intensity of
shaking
P(Ui) probability, for a given intensity or scenario, that damage sustained
by components in performance group “i” will result in an unsafe
placard
P(UT) total conditional probability, given a specific intensity of motion or
earthquake scenario, that a building will receive an unsafe placard
PEY probability of exceedance of an event in Y years
PR mean return period for an event
R number of realizations in a performance assessment
Rn nominal resistance in accordance with load and resistance factor
design (LRFD) specifications
S in the simplified analytical procedure, the ratio of elastic base shear
force to the structure’s yield shear strength
Sa(T) spectral response acceleration at period T, (g)

Second Edition
Sâ (T ) median collapse capacity of a structure, measured in terms of
spectral response acceleration at period T, (g)
Samax for time-based assessment, the maximum spectral response
acceleration at which performance is assessed, (g)
Samin for time-based assessment, the minimum spectral response
acceleration at which performance is assessed, (g)
SaD effective value of design spectral acceleration at which a structure
satisfies all applicable seismic design criteria of ASCE/SEI 7-10, (g)
SaO spectral acceleration that will tip an unanchored object to a critical
angle at which overturning instability will occur, (g)
S gm (T ) geometric mean (geomean) spectral response acceleration at period
T, (g)
Sv(1) spectral velocity at a period of 1 second
T period of structural response, i.e., the amount of time that a structure
in free vibration will take to complete one cycle of motion, (sec)
T1 period of fundamental mode of vibration
T1X period of fundamental mode of translation along the horizontal X
axis
T1Y period of fundamental mode of translation along the horizontal Y
axis
T average of the periods of the fundamental mode of translation along
each of two orthogonal building axes
Tmax maximum period in a range of periods used to scale ground motions
for nonlinear response history analysis
Tmin minimum period in a range of periods used to scale ground motions
for nonlinear response history analysis
V in the simplified analysis procedure, a pseudo lateral force used to
estimate story drifts
Vs average shear wave velocity of soil soil in the upper 100 feet (30
meters)
VPT peak total floor velocity at a floor level, in/sec
Vyt effective yield strength of a structure responding in the first mode
W1 in simplified analysis, the first mode weight effective of a structure
Y number of years considered in a time-based assessment

Second Edition
α for an unanchored object, the critical tip angle at which the object
will become unstable and overturn, radians
β measure of dispersion in the true value of a random parameter, also
called uncertainty
βa uncertainty associated with analysis record-to-record response
variability
βaa in the simplified analysis procedure, default record-to-record
variability (dispersion) associated with floor acceleration response
βav in the simplified analysis procedure, default record-to-record
variability (dispersion) associated with floor velocity response
βa∆ in the simplified analysis procedure, default record-to-record
variability (dispersion) associated with story drift response
βc modeling uncertainty associated with ambiguity in details of
construction; also, the uncertainty associated with estimates of repair
cost
βC in calculation-based fragilities, uncertainty associated with the
quality of installation (or construction) of a component
βD uncertainty associated with the ability of a design equation to predict
actual behavior when material strength is defined
βFA in the simplified analysis procedure, the total dispersion associated
with floor acceleration
βFV in the simplified analysis procedure, the total dispersion associated
with floor velocity
βgm dispersion accounting for uncertainty in ground motion intensity as
reflected in attenuation relationships
βm modeling uncertainty associated with dispersion in the predicted
response from structural analysis
βM uncertainty in material strength
βq modeling uncertainty associated with the quality and completeness
of the nonlinear analysis model, considering component hysteretic
modeling, estimates of damping, and the number of elements
included in the model
βr in a series of data used to establish fragility functions, the random
variability evident in the data; also, the damping ratio for a rocking
object (fraction of critical damping)

Second Edition
βSD in the simplified analysis procedure, the total dispersion associated
with story drift
βu in a series of data used to establish fragility functions, a measure of
uncertainty representing differences between the conditions inherent
in the data set and actual conditions present in a real building
δ sliding displacement that will cause an unanchored component to hit
another object, resulting in damage
δi peak displacement of floor “i” relative to the base
δr peak displacement of the roof relative to the base
δy component deformation at yield
γ factor used to determine effective damping for a rocking component
∆i in the simplified procedure, the uncorrected story drift ratio at level
“i”
∆i* estimated median peak story drift ratio at story “i”
∆r estimated median residual story drift ratio
∆ei in time-based assessments, an incremental range of earthquake
intensities
∆λi in time-based assessments, the probability of occurrence in a period
of time, for earthquake intensity in range ∆ei
∆y estimated story drift at development of yield strength
θ in lognormal distributions, the median value of demand
θDS in lognormal distributions, the median value of demand
λ mean annual frequency of exceedance of a ground motion parameter
µ ratio of component deformation at onset of damage, to component
deformation at yield
µD kinetic coefficient of friction between an unanchored object and its
resting surface
µS static coefficient of friction between an unanchored object and its
resting surface
φ resistance factor, as specified in load and resistance factor design
(LRFD) specifications
φj1 in simplified analysis, the ordinate of the first mode deflection at
floor level “j”

Second Edition
References
ACI, 2002, Building Code Requirements for Structural Concrete and

Commentary, ACI 318-02, American Concrete Institute, Farmington
Hills, Michigan.

Hills, Michigan.
ACI, 2008, Building Code Requirements for Masonry Structures and

Specifications for Masonry Structures, ACI-530-08/ASCE 5-08/TMS
402-08, Masonry Standards Joint Committee of American Concrete
Institute, American Society of Civil Engineers, and The Masonry
Society, Farmington Hills, Michigan.

Hills, Michigan.
AF&PA, 2005a, National Design Specification for Wood Construction,

ANSI/AF&PA NDS-2005, American Forest & Paper Products
Association, Washington, D.C.
AF&PA, 2005b, Special Design Provisions for Wind and Seismic,

ANSI/AF&PA SDPWS-2005, American Forest & Paper Products
Association, Washington, D.C.
AISC, 2005, Seismic Provisions for Structural Steel Buildings, ANSI/AISC

341-05, American Institute of Steel Construction, Chicago, Illinois.
ASCE, 1996, Standard Load and Resistance Factor Design (LRFD) for
Engineered Wood Construction, AF&PA/ASCE 16-95, American
Society of Civil Engineers, Reston, Virginia.
ASCE, 2002, Minimum Design Loads for Buildings and Other Structures,
ASCE/SEI 7-02, American Society of Civil Engineers, Reston, Virginia.
ASCE, 2005a, Minimum Design Loads for Buildings and Other Structures,
FEMA P-58-2 References G-1

Second Edition
ASCE, 2005b, Seismic Design Criteria for Structures, Systems, and
Components in Nuclear Facilities, ASCE/SEI 43-05, American Society
of Civil Engineers, Reston, Virginia.
ASCE, 2007, Seismic Rehabilitation of Existing Buildings, ASCE/SEI 41-06,

American Society of Civil Engineers, Reston, Virginia.
ASCE/SEI 7-10 including Supplement No. 1, American Society of Civil
Engineers, Reston, Virginia.
ASCE, 2014, Seismic Evaluation and Retrofit of Existing Buildings,

ASCE/SEI 41-13, American Society of Civil Engineers, Reston,
Virginia.
ASCE, 2017, Minimum Design Loads and Associated Criteria for Buildings
and Other Structures, ASCE/SEI 7-16, American Society of Civil
Engineers, Reston, Virginia.
ASME, 1981, Safety Code for Elevators and Escalators, ASME A17.1,
American Society of Mechanical Engineers, New York, New York.
FEMA, 2018a, Seismic Performance Assessment of Buildings, Volume 1 –

Methodology, Second Edition, FEMA P-58-1, prepared by the Applied
Technology Council for the Federal Emergency Management Agency,
Washington, D.C.
FEMA, 2018b, Seismic Performance Assessment of Buildings, Volume 3 –

Supporting Electronic Materials and Background Documentation,
Second Edition, FEMA P-58-3, prepared by the Applied Technology
Council for the Federal Emergency Management Agency, Washington,
D.C.
FEMA, 2018c, Seismic Performance Assessment of Buildings, Volume 5 –

Expected Seismic Performance of Code Conforming Buildings, FEMA
P-58-5, prepared by the Applied Technology Council for the Federal
Emergency Management Agency, Washington, D.C.
FEMA, 2018d, Guidelines for Performance-Based Seismic Design of

Buildings, FEMA P-58-6, prepared by the Applied Technology Council
for the Federal Emergency Management Agency, Washington, D.C.
G-2 References FEMA P-58-2

Second Edition
FEMA, 2018e, Updates to Fragilities, FEMA P-58/BD-3.7.19, Technical
Background Documentation, prepared by the Applied Technology
Council for the Federal Emergency Management Agency, Washington,
D.C.
Huang, Y.-N., Whittaker, A.S., and Luco, N., 2008, Performance Assessment
of Conventional and Base-Isolated Nuclear Power Plants for Earthquake
and Blast Loadings, Technical Report MCEER-08-0019,
Multidisciplinary Center for Earthquake Engineering Research,
University at Buffalo, Buffalo, New York.
ICC, 1994, 1997, Uniform Building Code, International Code Council,

Washington, D.C.
ICC, 2000, 2009, 2012, International Building Code, International Code

Council, Washington, D.C.
ICC, 2007, California Building Code, International Code Council,

Washington, D.C.
Lehman, D., Roeder, C.W., Herman, D., Johnson, S., and Kotulka, B., 2008,
“Improved seismic performance of gusset plate connections,” Journal of
Structural Engineering, Vol. 134, Issue 6, pp. 890-901.
Newmark, N.M., and Hall, W.J., 1982, Earthquake Design and Spectra,
Earthquake Engineering Research Institute, Oakland, California.
NFPA, 2007, Standard for Installation of Sprinkler Systems, NFPA 13,

National Fire Protection Association, Quincy, Massachusetts.
NIST, 1999, UNIFORMAT II Elemental Classification for Building

Specifications, Cost Estimating and Cost Analysis, NISTIR 6389 Report,
National Institute of Standards and Technology, Gaithersburg, Maryland.
Selna, L.G., 1981, “Slender wall tests instrumentation and results,”

Proceedings of the 50th SEAOC Convention, Coronado, California.
Tissel, J.R., 1967, 1966 Horizontal Plywood Diaphragm Tests, APA

Laboratory Report 106, American Plywood Association, Tacoma,
Washington.
Vamvatsikos, D., and Cornell, C.A., 2006, “Direct estimation of the seismic
demand and capacity of oscillators with multi-linear static pushovers
through IDA,” Earthquake Engineering and Structural Dynamics, Vol.
35, Issue 9, pp. 1097-1117.
FEMA P-58-2 References G-3

Second Edition
Project Participants
SECOND EDITION PROJECT TEAM
Federal Emergency Management Agency

Mike Mahoney (Project Officer) Robert D. Hanson (Technical Monitor)
Federal Emergency Management Agency Federal Emergency Management Agency
400 C Street, SW 5885 Dunabbey Loop
Washington, D.C. 20472 Dublin, Ohio 43017
Applied Technology Council

Christopher Rojahn Ayse Hortacsu (Project Manager)
Applied Technology Council Applied Technology Council
201 Redwood Shores Parkway, Suite 240 201 Redwood Shores Parkway, Suite 240
Redwood City, California 94065 Redwood City, California 94065
Jon A. Heintz (Project Executive)

Project Management Committee

Ronald O. Hamburger (Project Technical Director) Jack P. Moehle
Simpson Gumpertz & Heger, Inc. University of California, Berkeley
100 Pine Street, Suite 1600 3444 Echo Springs Road
San Francisco, California 94111 Lafayette, California 94549
John Gillengerten Khalid Mosalam

1055 Rivermeade Dr. University of California, Berkeley
Hebron, Kentucky 41048 Structural Engineering, Mechanics, and Materials
Civil and Environmental Engineering
William T. Holmes 723 Davis Hall
Rutherford + Chekene Berkeley, California 94720
375 Beale Street, Suite 310
San Francisco, California 94105 Laura Samant
2547 Diamond Street
John D. Hooper San Francisco, California 94131
Magnusson Klemencic Associates
1301 Fifth Avenue, Suite 3200 Steven R. Winkel
Seattle, Washington 98101 The Preview Group, Inc.
2765 Prince Street
Stephen A. Mahin (deceased) Berkeley, California 94705
University of California, Berkeley
777 Davis Hall
Berkeley, California 94720
FEMA P-58-2 Project Participants H-1

Second Edition
Project Steering Committee
William T. Holmes Jonathan C. Siu
Rutherford + Chekene City of Seattle, Dept. of Planning and
375 Beale Street, Suite 310 Development
San Francisco, California 94105 P.O. Box 34019
Seattle, Washington 98124
Lucy Arendt
St. Norbert College Jeffrey R. Soulages
100 Grant Street, Cofrin Hall, 310 Intel Corporation
De Pere, Wisconsin 54115 2501 NW 229th Street MS: RA1-220
Hillsboro, Oregon 97124
Deborah Beck
Beck Creative Strategies LLC Eric Von Berg
531 Main Street, Suite 313 Newmark Realty Capital, Inc.
New York, New York 10044 595 Market Street, Suite 2550
San Francisco, California 94105
Christopher Deneff
FM Global Engineering Standards Williston Warren (ATC Board Contact)
270 Central Avenue SESOL Inc.
Johnston, Rhode Island 02919 1918 Santiago Drive, Suite A
Newport Beach, California 92660
H. John Price
9839 Caminito Rogelio
San Diego, California 92131
Stakeholder Products Team

Laura Samant (Team Leader) Maryann T. Phipps
2547 Diamond Street Estructure
San Francisco, California 94131 1144 65th Street
Oakland, California 94608
David Mar
Mar Structural Design Sharyl Rabinovici
2629 Seventh Street, Suite C 1720 Le Roy Ave.
Berkeley, California 94710 Berkeley, California 94709
Lori Peek L. Thomas Tobin

University of Colorado Tobin & Associates
Natural Hazards Center 444 Miller Ave.
Institute of Behavioral Sciences Building Mill Valley, California 94941
1440 15th Street
Boulder, Colorado 80302
Performance Products Team

John Gillengerten (Team Leader) David R. Bonneville
1055 Rivermeade Dr. Degenkolb Engineers
Hebron, Kentucky 41048 375 Beale St., Suite 500
San Francisco, California 94105
H-2 Project Participants FEMA P-58-2

Second Edition
Dominic Campi Vesna Terzic
Rutherford + Chekene California State University, Long Beach
375 Beale Street, Suite 310 Department of Civil Engineering & Construction
San Francisco, California 94105 Engineering Management
Long Beach, California 90815
Products Update Team

John D. Hooper (Team Leader) Peter Morris
Magnusson Klemencic Associates AECOM
1301 Fifth Avenue, Suite 3200 2020 L Street
Seattle, Washington 98101 Sacramento, California 95811
Russell Larsen
1301 Fifth Avenue, Suite 3200
Stakeholder Products Consultants

Sandra L. Grabowski Stacia Sydoriak
Adler Enterprises LLC 2049 Bronson St
4600 Miller Street Fort Collins, Colorado 80526
Wheat Ridge, Colorado 80033
Jennifer Tobin-Gurley
Taline Mitten University of Colorado
Mar Structural Design Natural Hazards Center
2629 Seventh Street Suite C Institute of Behavioral Sciences Building
Berkeley, California 94710 1440 15th Street
Performance Products Consultants

Shreyash Chokshi Abe Lynn
California State University, Long Beach Degenkolb Engineers
1250 Bellflower Blvd, VEC-129 375 Beale St., Suite 500
Long Beach, California 90840 San Francisco, California 94105
Travis Chrupalo Daniel Saldana

Degenkolb Engineers California State University, Long Beach
375 Beale St., Suite 500 1250 Bellflower Blvd, VEC-129
San Francisco, California 94105 Long Beach, California 90840
Erica Hays Vinit M. Shah

Degenkolb Engineers California State University, Long Beach
375 Beale St., Suite 500 1250 Bellflower Blvd, VEC-129
San Francisco, California 94105 Long Beach, California 90840
Nirmal Kumawat Udit S. Tambe

California State University, Long Beach California State University, Long Beach
1250 Bellflower Blvd, VEC-129 1250 Bellflower Blvd, VEC-129
Long Beach, California 90840 Long Beach, California 90840

Second Edition
Duy Vu To Peny Villanueva
15662 Jefferson Street California State University, Long Beach
Midway City, California 92655 1250 Bellflower Blvd, VEC-129
Long Beach, California 90840
Products Update Consultants

Robert Bachman Gilberto Mosqueda
R.E. Bachman Consulting Structural Engineers University of California, San Diego
25152 La Estrada Dr. Department of Structural Engineering
Laguna Niguel, California 92677 9500 Gilman Drive, Suite 8085
La Jolla, California 92093
Jack Baker
Haselton Baker Risk Group, LLC Farzad Naeim
728 Cherry Street, Suite C Farzad Naeim, Inc.
Chico, California 95928 100 Spectrum Center Drive, Suite 900
Irvine, California 92618
Dustin Cook
Haselton Baker Risk Group, LLC Barbara Rodriguez
728 Cherry Street, Suite C Carbon Leadership Forum
Chico, California 95928 Center for Integrated Design
1501 East Madison St. Suite 200
Scott Hagie
John A. Martin and Associates, Inc. Kathrina Simonen
950 S. Grand Ave. 4th Floor Carbon Leadership Forum
Los Angeles, California 90015 Center for Integrated Design
1501 East Madison St. Suite 200
Angie Harris Seattle, Washington 98122
19413 Broadacres Ave.
Carson, California 90746 Siavash Soroushian
K.N. Toosi University of Technology
Curt Haselton Department of Civil & Environmental Engineering
Haselton Baker Risk Group, LLC Tehran, Iran
728 Cherry Street, Suite C
Chico, California 95928 Katherine Wade
Haselton Baker Risk Group, LLC
Wyatt Henderson 728 Cherry Street, Suite C
Magnusson Klemencic Associates Chico, California 95928
Seattle, Washington 98101 Farzin Zareian
University of California, Irvine
Monica Huang Department of Civil and Environmental
Carbon Leadership Forum
Engineering
Center for Integrated Design
1501 East Madison St. Suite 200 Irvine, California 92697

Second Edition
FIRST EDITION PROJECT TEAM
Federal Emergency Management Agency

Mike Mahoney (Project Officer) Robert D. Hanson (Technical Monitor)
Federal Emergency Management Agency Federal Emergency Management Agency
400 C Street, SW 5885 Dunabbey Loop
Washington, D.C. 20472 Dublin, Ohio 43017

Christopher Rojahn (Project Executive Director) Ayse Hortacsu
Applied Technology Council Applied Technology Council
201 Redwood Shores Parkway, Suite 240 201 Redwood Shores Parkway, Suite 240
Redwood City, California 94065 Redwood City, California 94065
Jon A. Heintz (Project Manager)

Project Management Committee

Ronald O. Hamburger (Project Technical Director) Peter J. May
Simpson Gumpertz & Heger University of Washington
100 Pine Street, Suite 1600 3630 Evergreen Point Road
San Francisco, California 94111 Medina, Washington 98039
John Gillengerten Jack P. Moehle

1055 Rivermeade Dr. University of California, Berkeley
Hebron, Kentucky 41048 3444 Echo Springs Road
Lafayette, California 94549
William T. Holmes (ex-officio)
Rutherford + Chekene Maryann T. Phipps (ATC Board Contact)
375 Beale Street, Suite 310 Estructure
San Francisco, California 94105 1144 65th Street
Oakland, California 94608
Steering Committee
William T. Holmes (Chair) Anne Bostrom
Rutherford + Chekene University of Washington
375 Beale Street, Suite 310 Parrington Hall, Room 327
San Francisco, California 94105 Seattle, Washington 98195
Roger D. Borcherdt Bruce Burr

U.S. Geological Survey Burr & Cole Consulting Engineers
345 Middlefield Road, MS977 3485 Poplar Avenue, Suite 200
Menlo Park, California 94025 Memphis, Tennessee 38111

Second Edition
Kelly Cobeen R. Jay Love
Wiss, Janney, Elstner Associates, Inc. Degenkolb Engineers
2200 Powell Street, Suite 925 1300 Clay Street
Emeryville, California 94608 Oakland, California 94612
Anthony B. Court David Mar

A.B. Court & Associates Mar Structural Design
4340 Hawk Street 2629 Seventh Street, Suite C
San Diego, California 92103 Berkeley, California 94710
Terry Dooley Steven McCabe

Morley Builders National Institute of Standards and Technology
2901 28th Street, Suite 100 100 Bureau Drive, MS 8630
Santa Monica, California 90405 Gaithersburg, Maryland 20899
Dan Gramer Brian J. Meacham

Turner Construction Company Worcester Polytechnic Institute
830 4th Avenue South, Suite 400 Dept. of Fire Protection Engineering
Seattle, Washington 98134 100 Institute Road
Worcester, Massachusetts 01609
Michael Griffin
CCS Group, Inc. William J. Petak
1415 Elbridge Payne Road, Suite 265 University of Southern California
Chesterfield, Missouri 63017 School of Policy Planning and Development
Los Angeles, California 90089
Risk Management Products Team

John D. Hooper (Co-Team Leader) Mahmoud Hachem
Magnusson Klemencic Associates Degenkolb Engineers
1301 Fifth Avenue, Suite 3200 1300 Clay Street
Seattle, Washington 98101 Oakland, California 94612
Craig D. Comartin (Co-Team Leader) Gee Heckscher

Comartin Engineers Architectural Resources Group
7683 Andrea Avenue 1301 53rd Street
Stockton, California 95207 Port Townsend, Washington 98368
Mary Comerio Judith Mitrani-Reiser

University of California, Berkeley National Institute of Standards and Technology
Dept. of Architecture 100 Bureau Drive
232 Wurster Hall Gaithersburg, Maryland 20899
Peter Morris
Allin Cornell (deceased) AECOM
Stanford University 2020 L Street
Dept. of Civil & Environmental Engineering Sacramento, California 95811
110 Coquito Way
Portola Valley, California 94028

Second Edition
Farzad Naeim Hope Seligson
Farzad Naeim, Inc. MMI Engineering
100 Spectrum Center Drive, Suite 900 2100 Main Street, Suite 150
Irvine, California 92618 Huntington Beach, California 92648
Keith Porter
University of Colorado
Dept. of Civil, Environmental, and Architectural
Engineering
1111 Engineering Drive
Risk Management Products Consultants

Travis Chrupalo Juan Murcia-Delso
Degenkolb Engineers University of California, San Diego
375 Beale St., Suite 500 9500 Gilman Drive, MC0085
San Francisco, California 94105 La Jolla, California 92093
D. Jared DeBock Scott Shell

California State University, Chico Esherick Homsey Dodge & Davis Architects
400 West 1st Street 500 Treat Avenue, Suite 201
Chico, California 95929 San Francisco, California 94110
Armen Der Kiureghian P. Benson Shing

ADK & Associates University of California, San Diego
1974 18th Avenue Dept. of Structural Engineering
San Francisco, California 94116 409 University Center
9500 Gilman Drive, MC0085
Scott Hagie La Jolla, California 92093
John A. Martin & Associates, Inc.
950 South Grand Avenue Mohamed M. Talaat
Los Angeles, California 90015 Simpson Gumpertz & Heger
100 Pine Street, Suite 1600
Curt B. Haselton San Francisco, California 94111
Haselton Baker Risk Group, LLC
728 Cherry Street, Suite C Farzin Zareian
Chico, California 95928 University of California, Irvine
Department of Civil and Environmental Engineering
Russell Larsen Irvine, California 92697
Structural Performance Products Team

Andrew S. Whittaker (Team Leader) Gregory Deierlein
University at Buffalo Stanford University
Dept. of Civil, Structural and Environmental Dept. of Civil & Environmental Engineering
Engineering Blume Earthquake Engineering Center, MC 3037
230 Ketter Hall Stanford, California 94305
Buffalo, New York 14260

Second Edition
John D. Hooper Nicolas Luco
Magnusson Klemencic Associates U.S. Geological Survey
1301 Fifth Avenue, Suite 3200 Box 25046, DFC, MS 966
Seattle, Washington 98101 Denver, Colorado 80225
Yin-Nan Huang Andrew T. Merovich

University at Buffalo A.T. Merovich & Associates, Inc.
127 Paramount Parkway 1950 Addison Street, Suite 205
Buffalo, New York 14223 Berkeley, California 94704
Laura Lowes
University of Washington
Dept. of Civil and Environmental Engineering
233C More Hall, Box 352700
Structural Performance Products Consultants

Jack Baker Kerem Gulec
Stanford University University at Buffalo
Dept. of Civil & Environmental Engineering 230 Ketter Hall
240 Terman Engineering Center Buffalo, New York 14260
Stanford, California 94305
Dawn Lehman
Dhiman Basu University of Washington
University at Buffalo Dept. of Civil and Environmental Engineering
230 Ketter Hall 214B More Hall
Buffalo, New York 14260 Seattle, Washington 98195
J. Daniel Dolan Jingjuan Li

Washington State University University of Washington
Dept. of Civil and Environmental Engineering 4747 30th Avenue NE B118
405 Spokane Street, Sloan 101 Seattle, Washington 98105
Pullman, Washington 99164
Eric Lumpkin
Charles Ekiert University of Washington
University at Buffalo 6353 NE Radford Drive, Apt. #3917
12 Hillcrest Drive Seattle, Washington 98115
Hamlin, New York 14464
Juan Murcia-Delso
Andre Filiatrault University of California, San Diego
University at Buffalo 9500 Gilman Drive, MC0085
Dept. of Civil, Structural and Environmental La Jolla, California 92093
Engineering
134 Ketter Hall Hussein Okail
Buffalo, New York 14260 University of California, San Diego
Aysegul Gogus La Jolla, California 92093
University of California, Los Angeles
5731 Boelter Hall
Los Angeles, California 90095

Second Edition
Charles Roeder Victor Victorsson
University of Washington Stanford University
Dept. of Civil and Environmental Engineering 1742 Sand Hill Road, Apt. 308
233B More Hall Palo Alto, California 94304
John Wallace
P. Benson Shing University of California, Los Angeles
University of California, San Diego Dept. of Civil Engineering
Dept. of Structural Engineering 5731 Boelter Hall
409 University Center Los Angeles, California 90095
La Jolla, California 92093
Christopher Smith
Stanford University
MC:4020, Building 540
Nonstructural Performance Products Team

Robert E. Bachman (Team Leader)
RE Bachman Consulting Manos Maragakis
25152 La Estrada Drive University of Nevada, Reno
Laguna Niguel, California 92677 Dept. of Civil Engineering,
Mail Stop 258
Philip J. Caldwell Reno, Nevada 89557
Schneider Electric
7 Sleepy Hollow Lane Eduardo Miranda
Six Mile, South Carolina 29682 Stanford University
Dept. of Civil & Environmental Engineering
Andre Filiatrault 240 Terman Engineering Center
University at Buffalo Stanford, California 94305
Dept. of Civil, Structural and Environmental
Engineering Gilberto Mosqueda
134 Ketter Hall University at Buffalo
Buffalo, New York 14260 Dept. of Civil, Structural, and Environmental
Engineering
Robert P. Kennedy (deceased) Buffalo, New York 14260
RPK Structural Mechanics Consulting, Inc.
28625 Mountain Meadow Road Keith Porter
Escondido, California 92026 University of Colorado
Dept. of Civil, Environmental, and Architectural
Helmut Krawinkler (deceased) Engineering
Stanford University 1111 Engineering Drive
Dept. of Civil & Environmental Engineering Boulder, Colorado 80309
240 Terman Engineering Center

Second Edition
Nonstructural Performance Products Consultants
Richard Behr David L. McCormick
Pennsylvania State University Simpson Gumpertz & Heger
Dept. of Architectural Engineering 100 Pine Street, Suite 1600
104 Engineering Unit A San Francisco, California 94111
University Park, Pennsylvania 16802
Ali M. Memari
Gregory S. Hardy Pennsylvania State University
Simpson Gumpertz & Heger Dept. of Architectural Engineering
4000 MacArthur Boulevard, Suite 710 104 Engineering Unit A
Newport Beach, California 92660 University Park, Pennsylvania 16802
Christopher Higgins William O'Brien

Oregon State University Pennsylvania State University
School of Civil and Construction Engineering The Pointe
220 Owen Hall 501 Vairo Boulevard
Corvallis, Oregon 97333 State College, Pennsylvania 16803
Gayle S. Johnson John Osteraas

Halcrow Exponent
500 12th Street, Suite 310 149 Commonwealth Drive
Oakland, California 94607 Menlo Park, California 94025
Paul Kremer Xin Xu

Pennsylvania State University 7 Xibei Xincun, Room 403
430 Canterbury Drive Wuxi, Jiangsu 214062, P.R. China
State College, Pennsylvania 16802
Fragility Review Panel

Bruce R. Ellingwood Stephen Mahin (deceased)
Colorado State University University of California, Berkeley
Dept. of Civil and Environmental Engineering Dept. of Civil Engineering
Fort Collins, Colorado 80523 777 Davis Hall
Robert P. Kennedy (deceased)
RPK Structural Mechanics Consulting, Inc.
28625 Mountain Meadow Road
Escondido, California 92026
Validation/Verification Team
Charles Scawthorn (Chair) Jack Baker
C. Scawthorn & Associates Stanford University
744 Creston Road Dept. of Civil & Environmental Engineering
Berkeley, California 94708 240 Terman Engineering Center

Second Edition
David Bonneville Hope Seligson
Degenkolb Engineers MMI Engineering
375 Beale St., Suite 500 2100 Main Street, Suite 150
San Francisco, California 94105 Huntington Beach, California 92648
Special Reviewers
Thalia Anagnos Fouad M. Bendimerad
San Jose State University 125B 24th Avenue
General Engineering Office 169 San Mateo, California 94403
San Jose, California 95192

Second Edition
FEMA P-58-2
Catalog No. 1894-6

FEMA P-58-2-SE Volume2 Implementation

Uploaded by

Copyright:

Available Formats

FEMA P-58-2-SE Volume2 Implementation

Uploaded by

Document Information

Original Description:

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

FEMA P-58-2-SE Volume2 Implementation

Uploaded by

Copyright:

Available Formats

What is the document about?

What is the document about?

Who prepared the document?

Who prepared the document?

Seismic Performance

FEMA P-58-2 / December 2018

Seismic Performance Assessment of Buildings

SECOND EDITION PROJECT TEAM

APPLIED TECHNOLOGY COUNCIL STAKEHOLDER PRODUCTS TEAM

FIRST EDITION PROJECT TEAM

APPLIED TECHNOLOGY COUNCIL RISK MANAGEMENT PRODUCTS

The three volumes together describe the seismic performance assessment

The purpose of this Volume 2 –Implementation Guide, Second Edition is to

FEMA P-58-2 Preface v

Updates include technical improvements to the methodology, and revisions

vi Preface FEMA P-58-2

ATC acknowledges the Pacific Earthquake Engineering Research Center

Special acknowledgment is extended to C. Allin Cornell and Helmut

ATC also gratefully acknowledges Michael Mahoney (FEMA Project

Ayse Hortacsu Jon A. Heintz

FEMA P-58-2 Preface vii

List of Tables ............................................................................................ xxiii

1. Introduction ...................................................................................... 1-1

2. Building Performance Model .......................................................... 2-1

FEMA-P-58-2 Table of Contents ix

3. Building Analytical Model and Performance Assessments .......... 3-1

4. Example Application: Intensity-Based Performance Assessment

x Table of Contents FEMA-P-58-2

5. Example Application: Time-Based Performance Assessment

6. Structural Fragility Calculation ..................................................... 6-1

7. Nonstructural Fragility Calculation ............................................... 7-1

8. Consequence Function Development.............................................. 8-1

FEMA-P-58-2 Table of Contents xi

Appendix A: Structural Component Fragility Specifications ............... A-1

Appendix B: Nonstructural Component Fragility Specifications ......... B-1

Appendix C: PACT User Manual ............................................................ C-1

xii Table of Contents FEMA-P-58-2

Symbols ....................................................................................................... F-1

References .................................................................................................. G-1

Project Participants .................................................................................. H-1

FEMA-P-58-2 Table of Contents xiii

Figure 1-1 Performance assessment process ....................................... 1-4

Figure 2-1 Format of PACT screenshots used, indicating the tab

Figure 2-2 PACT Project Information tab ........................................... 2-2

Figure 2-3 PACT Building Information tab ........................................ 2-3

Figure 2-5 PACT Population tab showing commercial office

Figure 2-6 PACT Population Manager utility ................................... 2-10

Figure 2-7 PACT Component Fragilities tab showing fragility

Figure 2-8 PACT Performance Groups tab. ...................................... 2-28

Figure 2-9 Normative Quantity Estimation Tool, Building

Figure 2-10 Normative Quantity Estimation Tool, Component

Figure 2-11 SPO2IDA idealized pushover curve for hypothetical

Figure 2-12 Plan of example structure................................................. 2-52

Figure 2-13 Section of example structure............................................ 2-52

Figure 2-14 Pushover curve for 2-story reinforced masonry

Figure 2-15 SPO2IDA Input............................................................... 2-54

Figure 2-16 SPO2IDA Output ............................................................. 2-54

Figure 2-17 Comparison of collapse fragility functions obtained