Applying Probabilistic Risk Management in

Design and Construction Projects

Construction
Industry Implementation Resource 280-2
Institute® Version 1.2
CII Member Companies
Abbott AMEC
Air Products and Chemicals AZCO
Ameren Corporation Alstom Power
American Transmission Company Audubon Engineering Company
Anheuser-Busch InBev BIS Industrial Services
Aramco Services Company Baker Concrete Construction
ArcelorMittal Barton Malow Company
Architect of the Capitol Bechtel Group
BP America Bentley Systems
Barrick Gold Corporation Black & Veatch
Bristol-Myers Squibb Company Burns & McDonnell
CITGO Petroleum Corporation CB&I
Cameco Corporation CCC Group
Cargill CDI Engineering Solutions
Chevron CH2M HILL
ConocoPhillips CSA Group
DTE Energy Coreworx
The Dow Chemical Company Day & Zimmermann
DuPont Dresser-Rand Company
Eastman Chemical Company eProject Management, LLC
Ecopetrol Emerson Process Management
Eskom Holdings Faithful+Gould
ExxonMobil Corporation Flad & Associates
General Electric Company Fluor Corporation
General Motors Corporation Foster Wheeler USA Corporation
GlaxoSmithKline GS Engineering & Construction Corporation
Global Infrastructure Partners Gross Mechanical Contractors
Huntsman Corporation Hargrove Engineers + Constructors
International Paper Hatch
Irving Oil Limited Hilti Corporation
Kaiser Permanente IHI E&C International Corporation
Koch Industries IHS
Eli Lilly and Company Industrial Contractors Skanska
Linde North America JMJ Associates
LyondellBasell JV Driver Projects
Marathon Petroleum Corporation Jacobs
National Aeronautics & Space Administration KBR
NOVA Chemicals Corporation Kiewit Corporation
Occidental Petroleum Corporation Kvaerner North American Construction
Ontario Power Generation Lauren Engineers & Constructors
Petroleo Brasileiro S/A - Petrobras Matrix Service Company
Petroleos Mexicanos McCarthy Building Companies
Petroliam Nasional Berhad McDermott International
Phillips 66 Midwest Steel
Praxair Parsons
The Procter & Gamble Company Pathfinder
Reliance Industries Limited (RIL) Quality Execution
SABIC - Saudi Basic Industries Corporation The Robins & Morton Group
Sasol Technology S&B Engineers and Constructors
Shell Global Solutions US SAIC Constructors
Smithsonian Institution SKEC USA
Southern Company SNC-Lavalin
Statoil ASA The Shaw Group
SunCoke Energy Siemens Energy
TNK-BP Technip
Teck Resources Limited Tenova
Tennessee Valley Authority TOYO-SETAL Engenharia
TransCanada Corporation URS Corporation
U.S. Army Corps of Engineers Victaulic Company
U.S. Department of Commerce/NIST/ Walbridge
Engineering Laboratory Wanzek Construction
U.S. Department of Defense/ Wilhelm Construction
Tricare Management Activity Wood Group Mustang
U.S. Department of Energy WorleyParsons
U.S. Department of Health & Human Services Yates Construction
U.S. Department of State Zachry Holdings
U.S. Department of Veterans Affairs Zurich
Vale
Applying Probabilistic Risk Management in
Design and Construction Projects

Research Team 280, Methods for Dealing with Uncertainty:

Applying Probabilistic Controls in Design and Construction

Construction Industry Institute

Implementation Resource 280-2

Version 1.2

March 2013
© 2013 Construction Industry Institute™

The University of Texas at Austin

CII members may reproduce and distribute this work internally in any medium at no cost to internal recipients. CII members
are permitted to revise and adapt this work for their internal use, provided an informational copy is furnished to CII.

Available to non-members by purchase; however, no copies may be made or distributed and no modifications may be
made without prior written permission from CII. Contact CII at http://construction-institute.org/catalog.htm to purchase
copies. Volume discounts may be available.

All CII members, current students, and faculty at a college or university are eligible to purchase CII products at member
prices. Faculty and students at a college or university may reproduce and distribute this work without modification for
educational use.

Printed in the United States of America.

Version 1.0 (June 2012)

Version 1.1 (October 2012) – Minor text edits throughout
Version 1.2 (March 2013) – Minor software edits
Contents

1. Introduction 1
2. Example Project Using Risk Management Levels 5
3. Risk Register Tool: Selecting the Appropriate File 13
4. Interpreting the Results 21
5. Conclusion 31
Glossary 33
References 37
 1. Introduction

CII Research Team (RT) 280 developed this comprehensive guide to probabilistic
risk analysis, first to help organizations determine whether the probabilistic approach
is appropriate for their projects, and, if appropriate, to help them implement it. The
guide takes the user through the typical steps of adopting probabilistic risk analysis,
from identifying risks to interpreting a probability distribution. In addition to this process
description, the guide also provides an overview of the interactive risk register tool
developed by the research team. Along with this overview and instructions, it offers
an example of a construction project that uses the tool.

Benefits of Probabilistic Risk Management

In the course of conducting the case studies that were a part of the RT 280
research process, the team interviewed project managers who used probabilistic
methods for risk management. These interviewees consistently reported that, among
the biggest advantages of using this type of risk analysis, is its ability to give them
more certainty in their capacity to successfully execute a project by a set completion
date. They reported that probabilistic controls enable them to anticipate, plan for, and
mitigate extra costs and delays. After analyzing the data from the case studies, the
team determined that the key advantages of using probabilistic over deterministic risk
analysis include the following:

• Comprehensive risk assessment – Results incorporate the cumulative

effect of known risks and the quantification of project uncertainties.

• Clear visual representation – Graphic representation of results allows

for robust means of communicating risk, including probability density
graphs, cumulative distributions, and tornado diagrams.

• Communication – The method allows for clear communication of project

cost or schedule uncertainty to stakeholders.

• Confidence in project outcomes – Results provide additional data

to support decision making and illuminate the possible outcomes of
decisions.

• Ability to manage cost and schedule – The method assesses the

impact of risk events on project cost or schedule performance and the
likelihood of achieving project outcomes.

1
1. Introduction

Additional information and details on the case study analysis can be found in Research
Summary 280-1, Probabilistic Risk Management in Design and Construction Projects,
and Research Report 280-11, Methods for Dealing with Uncertainty: Applying
Probabilistic Controls in Design and Construction.

Risk Management Levels

In order to address the benefits and implications of using a probabilistic approach,

the research team had to determine the process and framework for implementing
probabilistic approaches. Based on the case study analysis, the team developed a
three-level approach. Implementation Resource (IR) 280-3, What’s the Risk?, presents
the three levels of risk management, along with information on interpreting probabilistic
results. The levels, as identified by the research team, are presented in Figure 1, with
brief definitions.

Risk Category Level 1: Risk Identification is the formal act of

Poor soil conditions may identifying risks and opportunities for projects.
require deep foundations, Cost Typically, this includes the use of a risk checklist
resulting in cost increase.
and/or the start of a risk register.
Utility relocation delay could
Schedule
result in longer schedule.

HEAT MAP Level 2: Deterministic Risk Analysis is

E M M H H H the act of analyzing risks through a single-
Probability

D L M M H H point estimate of potential impacts. Typically,

C L L M M H
this involves a probability × impact matrix, a
B L L L M M
prioritized list of risks, and/or an expected value
A L L L L M
for contingency allocation of schedule or cost.
a b c d e
Impact

0.14
Level 3: Probabilistic Risk Analysis is the act
0.12
of analyzing risk through probability distribution
0.10

estimates of potential impact, known as Monte

Probability

0.08

0.06 Carlo simulations. Typically, this approach

0.04
involves cumulative distributions of potential
0.02
outcomes and determination of the probability
0.00
640 650 660 670 680 690 700 710 720
of meeting project targets.
Distribution for Cost to Completion ($ million)

Figure 1. The three levels of risk management

2
 1. Introduction

Because these three levels provided a clear framework for evaluating and
categorizing the case study data, formulating them helped the research team assess
the similarities and differences among the organizations studied; using the framework,
the team was better able to understand how each organization implements project risk
management and probabilistic approaches. The three-tiered framework also made it
easier to define, compare, and understand the benefits and barriers the organizations
experienced at each level.

3
 2. Example Project Using Risk Management Levels

To facilitate the implementation of the research, the team developed the three
separate risk registers included with this guide. These registers correlate to the three
levels of risk analysis: Level 1 – Risk Identification; Level 2 – Deterministic Risk Analysis;
and Level 3 – Probabilistic Risk Analysis. This section explains how to use the risk
register for each level and how to interpret the sample outputs of an actual bridge
construction project.

Project Context

To improve safety, a city is planning to change an at-grade intersection to a grade-

separated interchange. The construction will occur at highways US 555 and SH 111,
in a rapidly-developing suburban area. The highways have inadequate capacity and
are expensive to maintain. They are also the only viable east-west (US 555) and north-
south (SH 111) routes for commercial traffic for several miles in either direction. The
existing and proposed site plans are provided in Figures 2 and 3.

The project will upgrade the existing unlimited-access, two-lane US 555 into a
limited-access, four-lane highway. This improvement will include reconstruction of
the existing roadway section. The city anticipates that, where possible, US 555 will
be widened to the north of the existing facility, since right-of-way is more readily
available to the north. If, as assumed, the roadway embankment is supported by
retaining walls, widening the road to the north will affect a 10- to 15-foot-wide strip
of existing Class III wetlands along the eastern half of the upgrade. This widening
will also require construction near a historic building to the south and the crossing of
Wandering Creek, which is home to protected salmon and other animal species. The
project will also involve upgrading the existing unlimited-access, two-lane SH 111 into
a limited-access, four-lane highway. This change will include the reconstruction of the
existing roadway section. Finally, the project will convert the at-grade intersection of
US 555 and SH 111 into a grade-separated interchange.

5
2. Example Project Using Risk Management Levels

Figure 2. Existing site plan

Figure 3. Proposed site plan

6
 2. Example Project Using Risk Management Levels

Application of Level 1: Risk Identification

At an early stage in the design, the city conducted a facilitated risk identification
workshop to identify, categorize, and document the risks for the project. The project
team gathered preliminary data for the workshop, including the project scope
description, cost estimate, and construction schedule. The team members were
certain to include all the assumptions about scope, schedule, and cost estimate, as
well as any allowances and contingencies that were made as they were developed.
While the city had hired outside consultants to lead their risk workshops in the past,
this workshop was facilitated by an internal team member who was knowledgeable in
risk management and had risk elicitation training. The facilitator brought a series of risk
checklists maintained by the city to ensure that the team did not miss any commonly
encountered risks. However, these were only used after the team had completed its
brainstorming about project risks.

Key project team members attended the workshop, including the project manager,
design manager, construction manager, and safety officer. People from outside
the project team also attended, including internal discipline leads, a few mangers
with experience from other projects, and a few external consultants, including an
environmental consultant. The workshop facilitator began by asking the team members
to express their key issues and concerns with the project (i.e., what was keeping them
up at night). After a few hours of structured brainstorming and discussion, which
included an examination of the design/estimating and scheduling assumptions, the
team ended up with a categorized list of approximately 50 risks. Figure 4 presents
some of the risks identified in this session. In addition to identifying and describing
the risk, the project team also determined the project phase at which the risk could
occur and the category of its impact (i.e., cost or schedule).

At this point in the project life cycle (design), if the city only wanted to identify
the risks and did not want or need to conduct a more robust analysis, the city would
most likely use its historic contingency average of 10 percent and manage the risks
identified in this Level 1 risk identification workshop. Accordingly, the team would have
budgeted approximately $24 million.

7
 2. Example Project Using Risk Management Levels
8

LEVEL 1 – RISK IDENTIFICATION

Date Category of
Item Status Brief Risk Description Detailed Risk Description CII Project Phases
Identified Impact

Drop Manual
Manual Manual Manual Drop Down Drop Down
Down (MM/DD/YY)

On-site structures may be eligible for historic register

1 Active 09/14/12 Historic site approval Front End Planning Schedule
causing project delay.

Change in pavement Traffic studies may result in a change in pavement

2 Active 09/15/12 Detailed Engineering Cost
section and/or type section and/or type resulting in a project cost increase.

If required by the hydraulic study, the culvert at

Replace culvert over
3 Active 09/15/12 Wandering Creek may need to be replaced resulting in Front End Planning Cost
Wandering Creek
a project cost increase.

Widening to the north will impact a 10- to 15-foot-wide

Additional wetland
4 Active 09/15/12 strip of existing Class III wetlands along the east half of Detailed Engineering Cost
mitigation
the upgrade, causing an increase in project cost.

Delay in request for The city may not meet its RFP development schedule,
5 Active 09/15/12 Procurement Schedule
proposal resulting in a project delay.

Figure 4. Identified risks

 2. Example Project Using Risk Management Levels

Application of Level 2: Deterministic Risk Analysis

The city felt that the risk identification workshop did not provide enough information
for this project, due to the uncertainty and intuitively high costs associated with the
identified risks. In addition, city staff felt that the identified risks should be prioritized.
Thus, the city decided to progress to Level 2, deterministic risk analysis. Using the
same facilitator and team from the risk identification workshop, the city first defined
the meaning of “impact” and “probability” for this specific project. The facilitator and
the workshop team completed the Level 2 risk register input sheet. (See Figure 5.)

Category of Impact
Cost
Value Mean Value
Description
L H (Dollars)
VL Insignificant Cost lncrease 0% 0% $0
L Minimal Cost lncrease 0% 5% $450,000
M Significant Cost lncrease 5% 10% $1,350,000
H Very Significant Cost Increase 10% 20% $2,700,000
VH Critical Cost Increase 20% 40% $5,400,000
Schedule
Value Mean Value
Description
L H (Days) (Dollars)
VL Impact recoverable without affecting critical path 0% 0% 0.0 $0
L Impact recoverable affecting critical path 0% 5% 27.4 $547,500
M Critical path affected 5% 10% 82.1 $1,642,500
H Restructuring of project required 10% 20% 164.3 $3,285,000
VH Major restructuring of project required 20% 40% 328.5 $6,570,000

Risk Probability Ranking

Value
Description
L H
vl Very low likelihood of occurring 0% 10%
l Low likelihood of occurring 10% 30%
m Medium likelihood of occurring 30% 50%
h High likelihood of occurring 50% 70%
vh Very high likelihood of occurring 70% 100%

Figure 5. Deterministic impact and probability assessment

9
2. Example Project Using Risk Management Levels

The facilitator elicited qualitative assessments for the likelihood and consequences
of each major risk on the project. This was done using historic data when available,
but, most frequently, the facilitator built team consensus to assign the ratings. The
frequency and probability ratings for each risk were used to calculate the overall risk
ratings and mean values for the project. Figure 6 illustrates the completed deterministic
sheet of the risk register tool.

Since the deterministic risk ranking sheet highlights in red the critical risks that
deserve the most attention, it allowed the city to determine where to focus its mitigation
efforts. And, because the ranking sheet also provides risk mean values, the city was able
to determine that it would need approximately $3 million of contingency by summing
these values. The facilitator explained that this sum should only be used as a point of
reference, since overlapping risks can create a contingency value that is unreasonably
high. Conversely, eliminating risks can create a contingency value that is too low. To
increase their confidence and help assess a reasonable contingency value, the city
team members referenced their historic values for contingency on similar projects.
Overall, by conducting the Level 1 and the Level 2 methods, the city increased the
project’s contingency and budgeted approximately $25 million for the project.

10
 LEVEL 2 – DETERMINISTIC
Pre-Mitigated Risk Ranking

Most Likely Impact Most Likely Probability

Brief Risk Description Risk Mean Value Overall Risk
Quantitative Mean Quantitative (Dollars) Ranking
Qualitative Value Qualitative Value
Value (Dollars or Day) Mean Value

Drop Down Auto Drop Down Auto

Auto Auto Auto
(Selection Required) (Manual Override) (Selection Required) (Manual Override)

Historic site approval Medium 82 Medium 40% $657,000 Med

Change in pavement
High $2,700,000 High 60% $1,620,000 High
section and/or type

Replace culvert over

High $2,700,000 Low 20% $540,000 Med

2. Example Project Using Risk Management Levels

Wandering Creek

Additional wetland
Medium $1,350,000 Very Low 5% $67,500 Low
mitigation

Delay in request for

Medium 82 Low 20% $328,500 Med
proposal

Figure 6. Deterministic risk ranking

11
2. Example Project Using Risk Management Levels

Application of Level 3: Probabilistic Risk Analysis

After careful review of the Level 2 outputs, the city felt that a probabilistic approach
to risk management was warranted. Although Level 2 would be sufficient for many
of the city’s smaller projects, this project was larger and more complex than usual.
Thus, the feeling was that a probabilistic risk analysis (Level 3) would provide more
confidence that adequate funds were being allotted for contingency and that the
project’s completion date could be achieved. To conduct the analysis, the facilitator
from the first two workshops once again worked with key project leaders—many of
whom had participated in the previous workshops. In the workshop, the analysis used
three-point estimates for each risk (i.e., worst case, best case, and most likely); the
mean value from the Level 2 analysis served as a point of reference for the “most likely”
value. The facilitator used historic data whenever possible, but in most cases, the best
case, worst case, and most likely values were estimated by team consensus. After the
workshop, the facilitator used the values for best case, most likely, and worst case
columns on the probabilistic risk register sheet and ran a Monte Carlo simulation to
calculate the range and confidence for the project’s cost and schedule. (See Figure 7.)

Based on the Monte Carlo analysis, the project team determined that they had
80-percent confidence that the project would not exceed $26 million and would not
take more than 3.8 years.

12
 LEVEL 3 – PROBABILISTIC
Category of Cost (Dollars) Schedule (Days)
Brief Risk Description
Impact Best Case Most Likely Worst Case Best Case Most Likely Worst Case

Auto (Manual Auto (Manual

Auto Auto Manual Manual Manual Manual
Override) Override)

Historic site approval Schedule 40 82 120

Change in pavement section

Cost $950,000 $1,620,000 $1,800,000
and/or type

Replace culvert over

Cost $250,000 $540,000 $750,000
Wandering Creek

2. Example Project Using Risk Management Levels

Additional wetland mitigation Cost $40,000 $67,500 $90,000

Delay in request for proposal Schedule 50 82 90

Figure 7. Probabilistic risk analysis inputs

13
2. Example Project Using Risk Management Levels

Risk Management and Complete Risk Register

After conducting the three levels of risk analysis, the team needed to ensure that
the risks were properly managed. Using the allocation and risk mitigation tab in the
risk register tool, the team identified appropriate response strategies for each risk,
assigned a responsible individual to each one, estimated the cost of mitigation, and
determined potential trigger events for all of them. If the team wanted to re-rank the
risks and calculate a new contingency based on the mitigation strategies employed, they
could re-calculate post-mitigated risk values for Levels 1 and 2 using the risk register
for post-mitigation scenarios. The summary tab of the risk register tool populates a
complete risk register on the basis of the previous sheets; this can be used by the
project team to manage and control risks throughout the project’s lifecycle. To monitor
risk exposure and mitigation efforts, retire risks, and identify and quantify new risks,
the team has planned regular status reviews of the project risks. The team also plans
to run an updated Monte Carlo analysis at critical decision points throughout the
project’s life cycle.

Project Example Conclusion

By going through all three levels of risk management, the city was able to estimate
the cost and duration of the project more accurately and confidently. Due to the project’s
attributes, including the large project cost and tight schedule, the team decided to
employ all three levels, rather than only implementing the first two. By doing so, the
project’s deterministic and probabilistic cost contingencies were each adjusted by $1
million. Further, the probabilistic analysis also provided the city with a range of potential
outcomes and determined the project cost and schedule at an 80-percent confidence
level. Arriving at such a high level of confidence would not have been possible had
the team stopped at the deterministic level. Again, for many of the city’s smaller
projects with less risk and uncertainty, simply completing Levels 1 or 2 would likely be
sufficient. In contrast, Level 3 has particular value for the more complex projects that
inherently involve greater risk and uncertainty. Because more information is gained as
the analysis progresses through the levels, each provides additional benefits; these
benefits need to be weighed against the additional resources required to successfully
complete each level.

14
 3. Risk Register Tool: Selecting the Appropriate File

The purpose of the risk register tool is to help project personnel create project
risk registers. The tool consists of three separate Microsoft Excel®-based risk register
files, one for each corresponding level of risk management: Level 1, Identification;
Level 2, Deterministic Analysis; and Level 3, Probabilistic Analysis. Each risk register
file is independent; the user should select the file that best satisfies the project’s risk
management needs. Use the information in this section to help select the appropriate
risk register file.

Important Tool Note

While the risk register tool is complete and ready to use, it is not locked and is editable. Users
can modify the risk register according to their respective companies’ risk management
and project controls procedures. For example, the default risk register does not contain an
option for modeling opportunities along with risks. Some organizations may wish to model
opportunities and can easily modify the spreadsheet.

Each file contains additional instructions on its particular use. Figure 8 outlines
the risk management process. For a comprehensive overview of each level, including
key considerations of the benefits, results, requirements, and limitations of each level,
please refer to IR 280-3, What’s the Risk? This handy two-page resource will help you
select the level of risk analysis that best suits the project in question. After determining
the appropriate risk management level, open the corresponding risk register file.

15
 3. Risk Register Tool: Selecting the Appropriate File
16

Level 3 – Probabilistic Analysis

Analyze risk estimates of potential impacts Select
Yes
for input into Monte Carlo simulation. probabilistic
risk register.

No
Level 2 – Deterministic Analysis
Analyze risks through single-point estimates
Is this Select
of potential impacts.
enough? Yes
deterministic
risk register.

No
Level 1 – Identification
Start Identify risks and opportunities, and
progress to other levels if warranted. Is this Select
enough? Yes
identification
risk register.

Figure 8. Risk management flow chart

 3. Risk Register Tool: Selecting the Appropriate File

Risk Register Instructions

After selecting the appropriate risk register, open the corresponding risk register
file. Specific instructions for each level are provided below and in the file.

Level 1: Risk Identification

This file provides a template for a Level 1 – Risk Identification risk register. Use this
template if the highest level of risk management for a project is risk identification. While
the risk register for this level will provide a complete Level 1 risk register, the template
can be modified for user-added preferences or additional information. Please review
the tool instructions before proceeding to the subsequent sheets.

The Level 1 risk register tool consists of individual input sheets titled “Level 1 –
Identification” and “Allocation & Risk Mitigation.” In order to populate the file’s last
sheet—the risk register summary sheet—the user must complete these input sheets
as fully as possible. (See Figure 9 for an overview of the Level 1 sheets.) Data should
be entered from left to right.

Level 1 – Identification
Risk Register Sheet Overview

Below is a list of each sheet and the purpose it serves in creating the final risk register.

Provides necessary instructions and definitions for the risk

Instructions register tool. The sheet serves as a reference and requires
no input.

This is the first sheet to complete to create the Level 1

risk register. Enter initial risks on this sheet, including risk
Level 1 – Identification description, project phase, and category of impact (time
or cost).

This is the next sheet to complete to create the Level 1 risk

Allocation & Risk register. The input focuses on mitigating and monitoring the
Mitigation risks. The user can identify response strategies, assign key
individuals, and note potential trigger events.

This is the final sheet of the Level 1 risk register. The entire
Summary sheet is auto-populated on the basis of the data entered
on the previous sheets.

Figure 9. Level 1 risk overview

17
3. Risk Register Tool: Selecting the Appropriate File

Data can be entered in each cell in one of four ways: 1) manually (user manually
enters information); 2) using the drop down list (user selects from a pre-populated list
of choices); 3) automatically (user does not enter data in these cells; information is
populated on the basis of data previously entered); and 4) automatically, with manual
override (data are auto-populated, however, users can override that information
with their own values after unprotecting the sheet). Columns are marked according
to the input method used, and each cell displays a message that further explains the
specific information that should be entered.

Level 2: Deterministic Risk Analysis

The Level 2 file provides a template for a deterministic risk analysis risk register.
Use this file if the highest level of risk management for a project is deterministic risk
analysis. This risk register will provide a complete Level 2 risk analysis, but it can be
modified for user-added preferences or additional information. Please review this file’s
instruction sheet before proceeding to the subsequent sheets.

The risk register tool consists of four individual input sheets: 1) Input Sheet;
2) Level 1 – Identification; 3) Level 2 – Deterministic; and 4) Allocation & Risk Mitigation.
In order to populate the file’s last sheet—the risk register summary sheet—the user
must complete these input sheets as fully as possible. (See Figure 10 for an overview
of the Level 2 sheets.) Data should be entered from left to right.

Data can be entered in each cell in one of four ways: 1) manually (user manually
enters information); 2) using the drop down list (user selects from a pre-populated list
of choices); 3) automatically (user does not enter data in these cells; information is
populated on the basis of data previously entered); and 4) automatically, with manual
override (data are auto-populated, however, users can override that information
with their own values after unprotecting the sheet). Columns are marked according
to the input method used, and each cell displays a message that further explains the
specific information that should be entered.

18
 3. Risk Register Tool: Selecting the Appropriate File

Level 2 – Deterministic
Risk Register Sheet Overview

Below is a list of each sheet and the purpose it serves in creating the final risk register.

Provides necessary instructions and definitions for the risk

Instructions register tool. The sheet serves as a reference and requires
no input.

Defines values for impact and probability. Values are used

Input Sheet in deterministic calculations. Default values can be used,
if desired.

Risk identification is the first step in the creation of the Level

2 risk register. Enter initial risks on this sheet, including risk
Level 1 – Identification description, project phase, and category of impact (time
or cost).

Deterministic risk analysis is the second step in the creation

of the Level 2 risk register. Each risk is rated on a qualitative
impact × probability scale. In addition, the user can manually
Level 2 – Deterministic enter the expected risk impact × probability value. This sheet
will populate the risk mean value (i.e., the most likely case)
and overall risk ranking for each identified risk.

This is the last sheet to complete to create the Level 2 risk

register. The input focuses on mitigating and monitoring the
risks. The user can identify response strategies, assign key
Allocation & Risk individuals, and note potential trigger events. The user can
Mitigation also estimate the cost of mitigation and then re-rank the
risks’ impacts and probability. This sheet will populate the
post-mitigated risk mean value (i.e., the most likely case)
and give the overall risk ranking for each identified risk.

This is the final sheet of the Level 2 risk register. The entire
Summary sheet is auto-populated on the basis of the data entered
on the previous sheets.

Figure 10. Level 2 risk overview

19
3. Risk Register Tool: Selecting the Appropriate File

Level 3: Probabilistic Risk Analysis

The Level 3 file provides a template for a probabilistic risk analysis risk register.
Use this file if the highest level of risk management for a project is probabilistic risk
analysis. The risk register will provide a complete Level 3 risk analysis and provide
input values for probabilistic modeling software (e.g. Crystal Ball, @Risk, and Primavera
Risk Management, among others). It can be modified for user-added preferences or
additional information. Please review the file’s instruction sheet before proceeding to
the subsequent sheets.

The risk register tool consists of five individual input sheets: 1) Input Sheet;
2) Level 1 – Identification; 3) Level 2 – Deterministic; 4) Level 3 – Probabilistic; and
4) Allocation & Risk Mitigation sheets. In order to populate the file’s last sheet—the
risk register summary sheet—the user must complete these input sheets as fully as
possible. (Level 3 values are not populated in the complete risk register). Data should
be entered from left to right. (See Figure 11 for an overview of the Level 3 sheets.)

Data can be entered in each cell in one of four ways: 1) manually (user manually
enters information); 2) using the drop down list (user selects from a pre-populated list
of choices); 3) automatically (user does not enter data in these cells; information is
populated on the basis of data previously entered); and 4) automatically, with manual
override (data are auto-populated, however, users can override that information
with their own values after unprotecting the sheet). Columns are marked according
to the input method used, and each cell displays a message that further explains the
specific information that should be entered.

20
 3. Risk Register Tool: Selecting the Appropriate File

Level 3 – Probabilistic
Risk Register Sheet Overview

Below is a list of each sheet and the purpose it serves in creating the final risk register.

Provides necessary instructions and definitions for the risk

Instructions register tool. The sheet serves as a reference and requires
minimal input.

Defines values for impact and probability. Values are used

Input Sheet in deterministic calculations. Default values can be used,
if desired.

Risk identification is the first step in the creation of the Level 3

risk register. Enter initial risks on this sheet, including risk
Level 1 – Identification description, project phase, and category of impact (time
or cost).

Deterministic risk analysis is the second step in the creation

of the Level 3 risk register. Each risk is rated on a qualitative
impact × probability scale. In addition, the user can manually
Level 2 – Deterministic enter the expected risk impact × probability value. This sheet
will populate the risk mean value (i.e., the most likely case)
and overall risk ranking for each identified risk.

Probabilistic risk analysis is the third step in the creation of the

Level 3 risk register. The risk mean value (i.e., the most likely
case calculated in the deterministic analysis) is auto-populated
Level 3 – Probabilistic
on this sheet. The best case and worst case scenarios must
Input be entered for each identified risk. This will result in a three-
point estimate that can be exported to a Monte Carlo-based
software program for probabilistic analysis.

This is the last sheet to complete to create the Level 3 risk

register. The input focuses on mitigating and monitoring the
risks. The user can identify response strategies, assign key
Allocation & Risk individuals, and note potential trigger events. The user can
Mitigation also estimate the cost of mitigation and then re-rank each
risk’s potential impact and probability. This sheet will populate
the post-mitigated risk mean value (i.e., the most likely case)
and give the overall risk ranking for each identified risk.

This is the final sheet in the Level 3 risk register. The entire
Summary sheet is auto-populated on the basis of the data entered
on the previous sheets.

Figure 11. Level 3 risk overview

21
 4. Interpreting the Results

The Monte Carlo simulation is the most commonly used probabilistic approach in
construction and, notably, it was the only probabilistic method used by the case study
participants. A Monte Carlo simulation begins with a deterministic project cost and/or
schedule estimate (i.e., the base estimate) and adds multiple iterations of cost/time
estimates for uncertainty and risk (i.e., the risk estimate). The result is a total estimate
for cost and schedule, presented as a range. The uncertainty can be modeled as
inputs for probability distributions for “known” items (e.g., ranges of units, costs, and
durations) and for “unknown” items (i.e., risks events). Using a computational algorithm,
the Monte Carlo simulation runs hundreds or thousands of simulations, each time using
a different set of values from the cost and duration risk model inputs.

Several software packages are available to perform probabilistic Monte Carlo

simulations. The software can be add-on packages for spreadsheets, such as Microsoft
Excel®, or stand-alone products that build on the functionality of scheduling software
such as Primavera. For most projects using probabilistic analysis, the simulation
objective is to find the effects of multiple uncertainties and risk events on the total
project cost or project duration. Monte Carlo simulations provide a model of the cost
or schedule for all identified risks based on the probability that they will occur and the
interactions among the risks.

Probabilistic risk-based information can provide decision makers with increased

confidence in the accuracy of their estimates, and it enables them to more clearly
communicate uncertainty to project stakeholders. The tradeoff for achieving these
benefits is the level of effort, time, and resources required for risk identification,
assessment, and modeling. These can be significant, especially for large and/or
complex projects.

Key Inputs to Probabilistic Risk Management

Historic data can generate or inform probabilistic input; however, the uniqueness
of projects often requires expert opinion to adjust and determine the input values for
probabilistic analysis. With proper training and facilitation, experts can provide credible
and defensible probabilistic input values. Group participation with consensus building
helps to increase the accuracy of these input values.

23
4. Interpreting the Results

A risk workshop is a commonly used approach for gathering input for a Monte
Carlo model: a team of subject matter experts (SMEs) reviews the risk profile of the
project and uses its insight and expertise to 1) identify potential risk events (Level 1)
and 2) quantify the most likely impact if a risk were to occur (Level 2). For Level 3
analyses, these workshops include the level of uncertainty of each risk. For instance,
in the triangular distribution, the SMEs would determine inputs for the minimum,
most likely, and maximum values. Although a formal risk workshop is not required to
gather the input values, a workshop provides additional benefits—including knowledge
transfer— that enhance team understanding and alignment regarding potential project
risks. Workshops also help reduce the potential bias of individual participants, though
facilitators need to avoid the potential for group think. If a project is not large or complex
enough to warrant the level of effort required by a full workshop, experienced project
management personnel can develop a reasonable risk-based estimate with individual
input from project team members.

The following factors have a significant effect on the quality and success of a
probabilistic risk analysis approach:

• valid cost and schedule estimates, including the following:

–– a valid cost and schedule estimate that includes distinctly separate base
and contingency estimates
–– a valid critical path method (CPM) schedule with consistent and
complete logic
–– an appropriate work breakdown structure (WBS)

• valid risk register for probabilistic analysis, including the following:

–– input from multiple stakeholders through formal project team interaction
–– ranges of possible outcomes/impacts (cost and/or duration) for each
item
–– a risk breakdown structure (RBS) or comprehensive risk categories
–– mitigation measures for the impacts.

The most advantageous probabilistic analyses are those that require a probability
distribution for schedule and costs risks as an input. When defining an input probability
value, it is important to consider both the central tendency and its range or dispersion.
The central tendency can be described as the typical expected value: in deterministic

24
 4. Interpreting the Results

approaches, this is the single-point, expected value. The range is the difference between
the least and greatest values of all the values in a data set. It measures the distance
from the lowest expected value to the highest expected value. Generally accepted
practices include the use of high (maximum), low (minimum), and expected value
(most likely) to make the three-point estimates that are then applied to a triangular
probability distribution. Figure 12 illustrates probability distributions that are common
to probabilistic analysis using Monte Carlo simulation techniques.

Continuous Distribution: Continuous Distribution:

Normal Uniform
Probability

Probability

Cost or Duration Cost or Duration

Continuous Distribution:
Triangular Discrete Distributions
Probability

Probability

Cost or Duration Cost or Duration

Figure 12. Monte Carlo-type probability distributions of risk analysis input

All four distributions have a single high point (the mode), and a mean value that may
or may not equal the mode. It is notable that some of the distributions are symmetrical
around the average or expected value, while others are not. The most appropriate
probability distributions to use as the bases of Monte Carlo simulations are the ones

25
4. Interpreting the Results

that most closely reflect the distribution of actual or expected data. Historical data on
unit prices, activity durations, and quantity variations should be used when available, but
are often difficult to obtain. In cases in which there are not enough data to completely
define probability distributions, risk managers must rely on a subjective assessment
of the needed input variables. Elicitation of these subjective assessments in a risk
workshop can increase accuracy and add credibility to the analysis.

Interpreting and Communicating a Probabilistic Risk Analysis

Simulation

Project teams should consider the types of outputs that a probabilistic analysis
produces when selecting their risk analysis methods or tools. Generally speaking,
techniques that require increased rigor, demand stricter assumptions, or need more
input data produce more informative and helpful results. Figures 13–16 illustrate several
standard outputs related to cost and time that are unique to probabilistic simulations.
Because these outputs provide richer information regarding cost and schedule, they
support more effective decision-making than do deterministic outputs. These figures
illustrate the output of the Level 3 analysis for the highway overpass described in
Section 4.

What is the range of cost or schedule?

The probability mass function graphics in Figures 13 and 14 (i.e., histograms) depict
the results of a Monte Carlo simulation defining the probability distribution of cost and
time based on input assumptions. This type of information is useful for understanding
the expected cost/duration (i.e., mean cost/mean duration) and the range/dispersion
(i.e., standard deviation) of the projected costs and durations. Using this information,
project teams can formulate specific predictions of possible outcomes and use them
for decision-making. In this case, the range of project cost is $14 to $31 million, and
schedule duration range is between 2.5 to 4.25 years.

26
 4. Interpreting the Results

0.025 Mean = $22.5

0.020
80% Confidence
Probability

0.015 = $26

0.010

0.005

0.000
15 20 25 30
80% 20%

Total Cost ($ million)

Figure 13. Probability distribution for cost

Mean = 3.6 Years

0.025

0.020 80% Confidence

= 3.8 Years
Probability

0.015

0.010

0.005

0.000
3.0 3.5 4.0
80% 20%

Figure 14. Probability distribution for schedule

27
4. Interpreting the Results

What is my confidence in cost or schedule?

Monte Carlo simulations also provide key insights into the probability of specific
outcomes. The cumulative probability functions shown in Figures 15 and 16 provide
the same information shown in Figures 13 and 14, but in a cumulative fashion. The
cumulative functions provide a quick reference to the mean (50-percent probability)
and a confidence level in the estimate or schedule. In this example, an 80-percent
confidence level is shown. This result indicates that a decision maker should be
80-percent confident that the project will not exceed $26 million or 3.8 years. In other
words, the project would not exceed these values eight out 10 times. The accuracy of
these models is, of course, predicated on the accuracy of the inputs and the accuracy
of the model itself.

28
 4. Interpreting the Results

80% Confidence = $26

1.00

Cumulative Probability
0.80
Mean = $22.5
0.60

0.40

0.20

0.00
15 20 25 30
80% 20%

Total Cost ($ million)

Figure 15. Cumulative probability distribution for cost

80% Confidence
= 3.8 Years
1.00
Cumulative Probability

0.80
Mean = 3.6 Years
0.60

0.40

0.20

0.00
3.0 3.5 4.0
80% 20%

Year Completion

Figure 16. Cumulative probability distribution for schedule

29
4. Interpreting the Results

What risks have the greatest impact and where should I focus mitigation
efforts?

Sensitivity analysis is a primary modeling output that can be used in the valuation
of the impacts of individual risks. These are calculations that are extremely useful to
risk management and risk mitigation support. Figure 17 provides a sensitivity analysis
in the form of a “tornado diagram.” Tornado diagrams are valuable in that they depict
the sensitivity or influence of individual risks on the overall variability of the risk model.
By graphically showing the correlation between variations in model inputs and the
distribution of the outcomes, tornado diagrams highlight the greatest contributors
to the overall risk. The length of the bars on a tornado diagram corresponds to the
influence of the items on the overall risk (in this case, risk to schedule duration).

Unmitigated Mean Severity

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Risk Event

Historic Site

Replace culvert over

Wandering Creek

Delay in Request for

Proposal

Change in pavement section

and/or type

Ground improvement
required in interchange

Embankment delays

Figure 17. Tornado diagram sensitivity analysis

30
 4. Interpreting the Results

How much contingency do I need and when can I release it?

Probabilistic risk-based methods are powerful tools for contingency estimation. The
amount of contingency on a project can be defined in a variety of terms, depending
upon the modeler’s preference. As shown in Figures 13 and 15, a contingency can
be defined by the $3.5 million difference between the mean value and the 80-percent
confidence level value. Simulation-based methods allow for the development of a
contingency drawdown curve, such as the one shown in Figure 18. The contingency
can be reduced as risks are realized. If a risk is realized, the contingency will be spent.
If the risk is not realized, the contingency can be returned to the funding source.
Whether spent or returned, the contingency will no longer be needed after the point
in time at which the project risk event is expected to occur. As seen in Figure 17, the
historic site approval/rejection is the risk with the greatest impact on the project cost.
The correlation to this is shown in Figure 18, as a reduction in contingency at the point
in time when the historic site designation is approved or rejected.

Historic site
approval

3
RFP delay
resolved
2

1 Contingency

0
0 1 2 3 4

Time (Years)

Figure 18. Risk resolution graph

31
4. Interpreting the Results

As this series of illustrations shows, a probabilistic approach to risk analysis can

provide improved decision making capability. Monte Carlo simulations are the most
common method for project risk analysis because they provide detailed, illustrative
information about potential risk impacts on the project cost and schedule. Their
popularity is no doubt greatly due to the fact that they utilize standard project control
tools that most project teams produce to successfully manage their projects, e.g., cost
estimate, CPM schedule, and risk register. Probabilistic methods do require knowledge
and training for successful implementation. To produce quality input to Monte Carlo
simulations, the user must be familiar with probability distribution information, e.g.,
expected outcome values (mean), possible range/dispersion of outcome (standard
deviation), and distribution shape. Probabilistic modeling also requires a knowledge
of probability and certain software capabilities.

32
 5. Conclusion

RT 280 has shown that the use of probabilistic approaches for risk management
is valuable and effective. Employing probabilistic risk management enables project
teams to determine the likelihood of finishing a project per cost and schedule targets.
It also illuminates the risks that have the greatest potential impact to project cost and
schedule durations. Accordingly, the results of probabilistic analysis help project teams
make better decisions and optimize resources by focusing on the most important
risks to mitigate. Using probabilistic methods, team members are empowered to
work proactively rather than react to risk events. These benefits are driving many
organizations in the construction industry to implement probabilistic approaches on
their capital projects.

Clear benefits exist at each of the three risk management levels defined by the
research: identification, deterministic analysis, and probabilistic analysis. While risk
managers gain more information and confidence as the analysis progresses through
the levels, these benefits must be weighed against the additional resources required
to successfully complete each level. Probabilistic risk management is not appropriate
for every design and construction project, due to this cost/benefit tradeoff. This
research describes the benefits, results, requirements, and limitations of each level
to help organizations and projects teams effectively select the appropriate level of
risk management.

Organizations and project teams can implement probabilistic approaches at

all phases in a project’s life cycle. Both owners and contractors indicate that using
probabilistic simulation outputs offers the following benefits:

• strengthens decision making on which projects to pursue

• establishes contingency levels and realistic schedule durations to ensure

success

• identifies and prioritizes the risks to mitigate.

Project teams that have risk management at status meetings also benefit through
enhanced communication and team alignment. For effective implementation, project
teams should conduct probabilistic risk analyses on a regular basis, e.g., monthly or
quarterly, or whenever a significant change is identified.

33
5. Conclusion

The RT 280 risk register tools will help project teams with risk management in the
following ways:

• communicate the benefits and limitations—and, thus the

appropriateness—of each level of risk management

• explain how to properly apply each of the risk management levels

• generate the inputs needed to successfully complete a Monte Carlo

simulation.

This guidance will also help teams interpret the outputs from the simulations to
aid decision making.

34
 Glossary

Following are definitions for the terms used by RT 280.

Cost Estimating Terms

Allowance: An amount included in the base estimate for items that are known but
the details of which have not yet been determined.

Base Estimate: The most likely project estimate, exclusive of project contingency,
for known costs for all known design, engineering, cooperative agreements, right-
of-way, environmental, utilities, pre-construction, and construction work.

Contingency: An estimate of costs associated with identified uncertainties and

risks, the sum of which is added to the base estimate to complete the project cost
estimate. The expectation is that contingency will be expended during the project
development and construction process.

Cost Estimate: A prediction of quantities, cost, and/or price of resources required

by the scope of a project. As a prediction, an estimate must address risks and
uncertainties. The cost estimate consists of the base estimate for the known costs
associated with identified uncertainties and risks.

Estimate Basis: A documentation of the project type and scope for each cost
estimate, including estimates for items such as available drawings (i.e., defining
percent engineering and design completion), project design parameters, project
complexity, unique project location characteristics, and disciplines required to
prepare the cost estimate.

Total Cost Estimate: The sum of the base cost estimate and contingency.

Cost Management Terms

Baseline Cost Estimate: The most likely total cost estimate, serving as the
approved project budget and the basis for cost control. The approved budget
must correspond to an approved scope of work, a work plan, and an approved
project schedule.

Cost Control: The process of controlling deviations from the estimated project costs
and monitoring the risks and contingencies associated with change. Two principles
apply: 1) there must be a basis for comparison (e.g., the baseline cost estimate); and
2) only future costs can be controlled.

35
Glossary

Cost Management: The process for managing the cost estimate through reviews
and approvals, communicating estimates, monitoring scope and project conditions,
evaluating the impact of changes, and making estimate adjustments as appropriate.

Scope: Encompasses the elements, characteristics, and parameters of a project

and the work that must be accomplished to deliver a product with the specified
requirements, features, and functions.

Scope Changes: Changes in the requirements, features, or functions on which the

project design and estimate is based.

Scope Creep: The result of incremental minor scope changes that cumulatively
change the scope, cost, and schedule of a project.

Risk Analysis Terms

Cumulative Distribution: A visual representation that shows the probability that
a variable (e.g., a cost or schedule estimate) takes a value less than or equal to
X. For probabilistic project controls applications, the cumulative distribution from
a Monte Carlo cost or schedule simulation can quickly show the not-to-exceed
values at the 50-, 80-, or 90-percent confidence levels.

Monte Carlo Simulation: Risk simulations that sample probability distribution for
each variable to produce hundreds or thousands of possible outcomes; the results
can be analyzed to get probabilities of different outcomes.

Probability: A measure of how likely a condition or event is to occur, ranging from

0 to 100 percent (or 0.00 to 1.00).

Probabilistic Controls: Methods for managing project uncertainty in cost, scope,

and schedule, taking into account known or unknown events or conditions that
cannot be adequately defined.

Probability Density Graph: A visual display of the various properties of data to

identify probability distribution.

Qualitative Risk Analysis: A process for prioritizing the effects of risks and
conditions on project objectives. It involves assessing the probability and impact
of project risk(s) and using methods such as the probability and impact matrix to
classify risks as high, moderate, and low for prioritized risk response planning.

36
 Glossary

Quantitative Risk Analysis: A process for measuring the probability and

consequences of risks, and estimating their implications for project objectives. Risks
are characterized by probability distributions of possible outcomes. This process
uses quantitative techniques such as simulation and decision tree analysis.

Risk: An uncertain event or condition that, if it occurs, has a negative or positive

effect on a project’s objectives.

Risk Acceptance: This risk planning technique indicates that the project team has
decided not to change the project plan to deal with a risk, or is unable to identify
any other suitable response strategy.

Risk Allocation: The placement of responsibility for a risk on a party through a

contract. The tenets of risk allocation include assigning risks to the party that is best
able manage them, assigning them in alignment with project goals, and assigning
them to promote team alignment with customer-oriented performance goals.

Risk Assessment: The component of risk management that bridges risk identification
and risk analysis in support of risk allocation. It involves quantitative or qualitative
analysis that assesses the impact and probability of a risk.

Risk Avoidance: This risk planning technique involves changing the project plan to
eliminate the risk or to protect the project objectives from its impact.

Risk Documentation: The recording, maintaining, and reporting of assessments,

the handling of analysis and plans, and the monitoring of results. Includes all plans,
reports for the project manager and decision-making authorities, and reporting
forms that may be internal to the project manager.

Risk Event: A discrete occurrence that may affect the project, for better or worse.

Risk Identification: The determination of which risks might affect the project and
the documentation of their characteristics.

Risk Management: All of the steps associated with managing risks: risk
identification, risk assessment, risk analysis (i.e., qualitative or quantitative), risk
planning, risk allocation, and risk monitoring control.

Risk Management Plan: A document detailing how risk response options and
the overall risk processes will be carried out during the project. This plan is the
output of risk planning.

37
Glossary

Risk Mitigation: This risk planning technique seeks to reduce the probability and/
or impact of a risk to a level below an acceptable threshold.

Risk Monitoring and Control: The capture, analysis, and reporting of project
performance, usually as compared to the risk management plan.

Risk Planning: The analysis of risk response options (acceptance, avoidance,

mitigation, or transference) and the decision making on how to approach and plan
risk management activities for a project.

Risk Register: A document detailing all identified risks, including description,

cause, probability of occurrence, impact(s) on objectives, proposed responses,
owners, and current status.

Risk Transference: This risk planning technique seeks to shift the impact of a risk
to a third party together with ownership of the response. (See also, Risk Allocation.)

Sensitivity: When the outcome is dependent on more than one risk source, the
sensitivity to any specific one of those risks is the degree to which that specific
risk (i.e., event or condition) affects the outcome or value.

Sensitivity Analysis: The results of systematically changing a model variable to

determine the impacts of the changes on the outcome.

Simulation: A computer model that translates the uncertainties specified at a

detailed level into their potential impact on objectives at the project level. Project
simulations typically use the Monte Carlo technique.

38
 References

Association for the Advancement of Cost Engineering International (2012). Integrated

Cost and Schedule Risk Analysis Using Monte Carlo Simulation of a CPM Model.
AACE International Recommended Practice No. 57R-09, Morgantown, PA.

Association for the Advancement of Cost Engineering International Risk Committee

(2000). “AACE International’s Risk Management Dictionary,” Cost Engineering
Journal, 42 (4), 28-31.

International Organization for Standardization (2009). ISO 31000 Risk Management —

Principles and Guidelines, Geneva, Switzerland.

Construction Industry Institute (2010). Management of Project Risks and Uncertainties,

Research Summary 6-8, Version 1.2. Austin, TX.

Project Management Institute (2004). A Guide to Project Management Body of

Knowledge, Newton Square, PA.

Wideman, R.M. (1992). Project and Program Risk Management: A Guide to Managing
Project Risks. Newton Square, PA.

39
Research Team 280, Methods for Dealing with Uncertainty:
Applying Probabilistic Controls in Design and Construction
Jeffery Bormann, Zachry
Chris Curow, Southern Company
* Karen Furlani, CH2M HILL
Daniel Hogan, U.S. Department of State
* Amy Javernick-Will, University of Colorado
† Douglas Kaiser, Exxcel Project Management
Kathleen Marwitz, Abbott
* Keith Molenaar, University of Colorado
Scott Penrod, Walbridge
* Craig Relyea, Eli Lilly and Company
Brendan Robinson, Architect of the Capitol
Marie Robinson, Jacobs
Ralph Rodriguez, Kaiser Permanente
* Christopher Senesi, University of Colorado
Michael Shirey, Tennessee Valley Authority

* Principal Authors
† CII wishes to thank the Project Management Institute for generously sponsoring the
participation of this team member and for providing a significant amount of data for
this study.

Editor: Jacqueline Thomas

 Construction Industry Institute
The University of Texas at Austin
3925 W. Braker Lane (R4500)
Austin, Texas 78759-5316
IR 280-2, Version 1.2