57R-09 - Ntegrated Cost and Schedule Risk Analysis

57R-
09
INTEGRATEDCOSTANDSCHEDULE
RISKANAL YSI
SUSI
NGRISK
DRIVERSANDMONTECARLO
SIMULATIONOFACPM
MODEL
AACE International Recommended Practice No. 57R-09
INTEGRATED COST AND SCHEDULE RISK ANALYSIS USING

RISK DRIVERS AND MONTE CARLO SIMULATION OF A CPM
MODEL
TCM Framework: 7.6 – Risk Management
Rev. July 9, 2019

Note: As AACE International Recommended Practices evolve over time, please refer to web.aacei.org for the latest
revisions.
Any terms found in AACE Recommended Practice 10S-90, Cost Engineering Terminology, supersede terms defined in
other AACE work products, including but not limited to, other recommended practices, the Total Cost Management
Framework, and Skills & Knowledge of Cost Engineering.
Contributors:
Disclaimer: The content provided by the contributors to this recommended practice is their own and does not necessarily
reflect that of their employers, unless otherwise stated.
July 9, 2019 Revision:

Dr. David T. Hulett, FAACE (Primary Contributor) John K. Hollmann, PE CCP CEP DRMP FAACE Hon. Life
Waylon T. Whitehead (Primary Contributor) Donny Lai
James E. Arrow, DRMP John R. Schuyler, PE CCP DRMP
William A. Banks, Jr. Edward J. Thomas
David C. Brady, P.Eng. DRMP Warner W. Uhl
Larry R. Dysert, CCP CEP DRMP FAACE Hon. Life
June 18, 2011 Revision:

Dr. David T. Hulett (Primary Contributor) John K. Hollmann, PE CCE CEP
Christopher P. Caddell, PE CCE Donald F. McDonald, Jr. PE CCE PSP
Tommy Clarke Oscar A. Mignone
Dr. Ovidiu Cretu, PE Stephen P. Warhoe, PE CCE CFCC
Kevin M. Curran Robert F. Wells, CEP
Michael W. Curran Dr. Trefor P. Williams
Patrick B. Egger Ronald M. Winter, PSP
Ricardo Garcia da Roza David C. Wolfson
John M. Hale Rashad Z. Zein, PSP
Dennis R. Hanks, PE CCE
Copyright © AACE® International AACE® International Recommended Practices

Single user license only. Copying and networking prohibited.
This document is copyrighted by AACE International and may not be reproduced without permission. Organizations may obtain permission
to reproduce a limited number of copies by entering into a license agreement. For information please contact [email protected]
AACE® International Recommended Practice No. 57R-09
INTEGRATED COST AND SCHEDULE RISK ANALYSIS
USING RISK DRIVERS AND MONTE CARLO
SIMULATION OF A CPM MODEL
TCM Framework: 7.6 – Risk Management
July 9, 2019
TABLE OF CONTENTS
Introduction ...................................................................................................................................................................2
Scope .........................................................................................................................................................................2
Purpose ......................................................................................................................................................................2
Background ................................................................................................................................................................2
Recommended Practice .................................................................................................................................................3
Preconditions for a Schedule Risk Analysis................................................................................................................3
A CPM Schedule that Complies with Industry Practices ........................................................................................3
Risk Data Collected to Improve Data Quality ........................................................................................................4
Properties of Risk Drivers ..........................................................................................................................................5
Risk Drivers and Discrete Risks ..............................................................................................................................5
Effect of Duration Uncertainty with 100% Likelihood ...........................................................................................6
Effect of a Risk Driver with Less Than 100% Likelihood .........................................................................................7
Assigning Risk Drivers to Multiple Activities ..........................................................................................................8
Risk Drivers Inherently Model How Activity Durations Become Correlated .........................................................9
Assigning Risks in Parallel or in Series..................................................................................................................11
Case Study - Model of Offshore Gas Production Platform Project ..........................................................................12
Modeling Uncertainty ..........................................................................................................................................13
Modeling Systemic Risks as Risk Drivers ..............................................................................................................14
Schedule Risk Results Using Risk Drivers in a Monte Carlo Simulation ...............................................................15
Prioritizing Risk Drivers for Management Action ................................................................................................16
Risk Treatment Using Prioritized Risks ....................................................................................................................17
Integrating Cost and Schedule Risk Analysis........................................................................................................18
Summary......................................................................................................................................................................22
Contributors ................................................................................................................................................................24

57R-09: Integrated Cost and Schedule Risk Analysis Using Risk Drivers and Monte Carlo Simulation of 2 of 25
a CPM Model
July 9, 2019
INTRODUCTION
Scope
This recommended practice (RP) defines and explains the integration of cost and schedule risk analysis using a Monte
Carlo simulation of a critical path method (CPM) resource-loaded schedule. It explains in some detail the use of risk
drivers [1] to represent the identified risks to a project’s cost and schedule in an integrated approach. There are
generally three overall purposes of such an exercise:
• To estimate the probability of finishing on or before the schedule date and on or under the cost estimate.
• To determine the amount of cost and schedule contingency needed to provide a chosen degree of confidence
of hitting both targets.
• To identify the risks that cause any estimated overrun.
Using root-cause risks to generate the simulation results permits the prioritization of identified and quantified risks
by their impact on the schedule. This focus on risks facilitates identification of effective risk treatment options and
estimates of the post-treatment results.
Purpose
This document is not intended to be a standard. This document is intended to provide a guideline for integrated
project cost and schedule risk analysis using risk drivers in the context of conducting a Monte Carlo simulation-based
schedule risk analysis of a CPM project schedule that most practitioners would consider to be good practice. This RP
illustrates some of the most important features of risk drivers and compares the method to other risk analysis
methods also described in AACE International recommended practices. Alternative methods to evaluate contingency
include those relying on parametric estimation such as those described in Recommended Practice 42R-08 [2]. The
reader is also encouraged to read recommended practices 44R-08, Risk Analysis and Contingency Determination
Using Expected Value [3] and 64R-11, CPM Schedule Risk Modeling and Analysis: Special Considerations [4].
Background
In integrated cost and schedule risk analysis, the platform for the analysis is usually a summary analysis schedule
reflecting the project plan at a summary level of detail. The integration of cost and schedule is made possible because
the total cost, expressed without contingency, is loaded onto the activities as time dependent (labor and rented
equipment) and time independent (material and equipment to be installed) resources.
The risk driver method of representing the identified risks (project specific or systemic risks) characterizes those risks
by:
• Their probability of occurring with some tangible impact.
• A probability distribution of impact factors, to be applied to activities’ scheduled duration and cost variance
during a Monte Carlo simulation, representing each risk’s impact on activity durations if it occurs.
• The activities that the risk affects if it occurs.
Risks are represented by their probability of occurring, and their impact on activity duration and cost (burn rate for
labor, total cost for materials). The difference between risk drivers and discrete risks is in the use of multiplicative
factors to represent the impact of a risk on the duration of the affected activities.

a CPM Model
July 9, 2019
The method is consistent with Recommended Practice 40R-08, Contingency Estimating – General Principles, [5] that
specifies several first principles including: “starts with identifying risk drivers” and “links risk drivers to the activities
they affect.”
The risk driver described in this RP identifies risks’ importance to the cost and schedule objectives. The effect of
uncertainty (i.e. background noise reflecting inherent variability, estimating error and estimating bias if present) is
included to derive the total impact to the cost and schedule from these two sources.
An approach to examine the impact of risk upon a project schedule is the expected value method as described in
Recommended Practice 44R-08 [3] and 65R-11 for integrated cost-schedule risk analysis [7]. This method is based
on estimates of the probability of the risks occurring and of the overall impact (in days or months) on the final finish
date if the risk occurs. The expected value method often does not use the project schedule as a platform, which is a
drawback since the calculation of the risks’ influence on the final completion date is very difficult, particularly with
risks affecting parallel paths in the schedule. In contrast, Monte Carlo simulation uses the logic-driven CPM schedule
as its platform.
Risk drivers can be applied to all activities in the schedule that they influence. As a consequence, some risks will be
assigned to multiple activities and some activities will be influenced by multiple risks. This method models the holistic
effect of risks as they cause durations of all activities they affect to expand and contract. Costs also increase or
decrease because duration variation during simulation indirectly affects labor-type resource costs. Importantly,
during simulation the critical path may vary as some risks occur and others do not. The risk driver method uses
simulation to estimate the project schedule and cost using risk data to calculate the finish date according to dynamic
CPM principles.
The risk driver approach incorporates the interactions and interdependencies of the project plan. The analyst does
not have to resolve questions about whether the risk occurs on a risk-critical path or to estimate the impact of risks
occurring frequently, or infrequently, since the Monte Carlo simulation accounts for all these considerations.
Additionally, the analyst does not have to estimate and impose correlations between activity durations since the risk
driver method generates these during simulation, as shown below. If the risks are truly root causes of variations
rather than just symptoms of a more basic cause, those drivers are likely to be independent of each other and not
to be correlated. Finally, using the root cause risks as drivers of the simulation facilitates risk prioritization for timely
and effective risk treatment decisions by management.
The target audience for this recommended practice includes risk practitioners and project managers who need to
become aware of contemporary methods of using root cause risks, represented by risk drivers, to perform integrated
cost and schedule risk analysis. Others involved in making decisions about risk treatment actions should also be
aware of the benefits of using risk drivers.
RECOMMENDED PRACTICE
Preconditions for a Schedule Risk Analysis
A CPM Schedule that Complies with Industry Practices
Quantitative risk analysis that is based on Monte Carlo simulation of a CPM schedule has advantages and poses some
challenges. The schedule should be reviewed against easily-obtained scheduling practices such as that written by
the US Government Accountability Office.

a CPM Model
July 9, 2019
Project schedules are dynamic models of the project plan, since any risk event that affects any activity’s duration
may cause a delay in the finish date.1 Schedule uncertainty, represented by inherent variability, estimating error or
estimating bias, can be built upon by adding risk drivers on top of the uncertainty to illustrate the build up from the
deterministic schedule to a fully risk loaded schedule. Working with the project schedule allows detailed modeling
about each identified risk and assigning its impact to specific activities or to general categories of activities. The
impact may affect durations and/or costs. The simulation software working with the CPM schedule computes the
detailed implications for the finish date and cost of uncertainty and many risks potentially occurring simultaneously.
Frequently, the schedule available for analysis is a detailed contractor’s schedule. That schedule may have many
more activities than are needed to model the strategic effect of risk on the project’s finish date. This schedule also
may be incomplete or omit scope outside of the contractor that is needed for the project to be finished and total
cost to be considered. Often, schedules submitted for analysis are not compliant with project scheduling industry
practices (such as avoiding dangling activities, using constraints and lags, exhibiting realistic total float and having
continuous and believable critical paths). In such cases, analysis cannot proceed until these deficiencies have been
rectified. To perform schedule simulation for quantitative risk analysis, dynamic models, where the milestone dates
and total cost are determined by activity durations and schedule relationship logic, are required.
A new summary analysis schedule can be developed and validated by plenary stakeholders to ensure all project
scope is identified. The summary analysis schedule needs to be inclusive of all work since the analyst cannot tell in
advance of the analysis which paths will turn out to be critical when risks are considered. The main paths in the
summary schedule need to reflect the more detailed plan with realistic float values.2 That schedule needs to include
sufficient detail to represent the many interdependencies that occur in complex projects.
The analysis is carried out against the summary schedule. Note that time is needed to gather all the data, carry out
the analysis, prepare reports and the like. In that period, there may have been progress or changes, which needs to
be assessed and the analysis may need to be revised or revalidated with this latest data.
Risk Data Collected to Improve Data Quality
Collecting good-quality data about project risks is crucial to the success of the risk analysis. Typically risk workshops,
confidential risk interviews, or a combination of the two methods are used.
• Risk workshops gather knowledgeable people into a room to identify and quantify the risks’ probability and
impact on activities’ costs and durations. Under the leadership of a skilled facilitator the discussion may lead to
a consensus on a set of risks with usable data that benefits from the synergy of those in the room. Often,
however, people find that sharing honestly and openly in a workshop setting is difficult, particularly if there are
risks that cannot be discussed because they are unpopular, may conflict with management statements or
customer requirements, imply the project is in default of the contract terms, or for other reasons. These risks
are sometimes called “unknown knowns” since they are known but cannot be mentioned, as referred to in the
article “There are Known Knowns.” [6] Within organizations with a relatively low level of risk management
maturity, [7] discussing these risks in risk workshops may be difficult for some who fear reprisal from
management. Other issues that negatively affect workshop efficiency include: groupthink (suppressing dissent),
the “Moses factor” (i.e. an influential person such as the project manager who overwhelms others), and cultural
conformity (i.e. decisions that match the organization’s norms). [8]
1
In this RP the risk drivers will be described as potentially delaying the finish date or increasing cost. Of course, risks can be opportunities that
may conceptually shorten activity durations resulting in an earlier finish date. In general, many risks described by those on the project are either
all or mostly threats, so for ease of explanation most of the discussion is written as if risks are threats to project completion targets.
2
Float values may not replicate those in the detailed schedule if that schedule is not compliant with industry practices.

a CPM Model
July 9, 2019
• Confidential interviews provide the best opportunity for individuals to express their opinions openly, honestly
and without fear of retribution. These interviews usually identify and calibrate some risks that are captured in
the risk register, often identifying unknown knowns for the first time. Once the risks are identified in an
interview the risks can be commented on by other interviewees in confidence or brought up anonymously for
group buy-in but nobody knows what anyone else has said in other interviews.
• Review of existing data on comparable and recent projects should also be brought to the risk data collection
exercise. This is known in some organizations as the lessons learned register or database. AACE RP 114R-20,
Project Historical Database Development, provides further information about developing databases including
for scheduling purposes. [9] Comparing the data and results for the current project with past experience
represented by completed projects benchmarks may bring what is called the outside view to the discussion. [10]
Making reference to historic databases can often bring more realism to the risk discussion and provide a means
to corroborate identified risks with their likelihood and uncertainty ranges. As a way of learning from the
database of past projects a method of analysis based on statistical regression of overruns relative to systemic
risks contributes to, and provides a good check of, the simulation results. [2]
Properties of Risk Drivers
Risk Drivers and Discrete Risks
There are two approaches to representing individual risks. These are sometimes called discrete risks and risk drivers
and they share some attributes but differ in others. Both methods represent project-specific risks, are attached to
specific activities, and the probability that they affect these activities’ durations. The difference between these two
approaches is in how they represent and handle the impact of the risk if it occurs.
• Discrete risks represent the impact of the risk as a probability distribution of an estimated number of days that
is either added to or subtracted from the scheduled duration. If a risk is assigned to both short and long activities
the impact distribution must be adjusted by activity to be appropriate to the duration of those activities. In most
simulation software, because of the logic employed in solving the risk-impacted schedule, a discrete risk cannot
possess opportunity (fewer days) and threat (more days) characteristics in impact but must be either threat or
opportunity.
• Risk drivers’ impacts are expressed as a probability distribution of multiplicative adjustment factors from which
the simulation software chooses values at random for each iteration. For example, if a multiplicative factor of
107% is chosen for an iteration it multiplies the deterministic durations of the activities the risk driver affects
for that iteration by 1.07 in that iteration. Since the impact values are expressed as multiplicative factors, the
factor can be applied to both long and short activities and hence affect their durations proportionately.
Also, the risk driver can possess both threat and opportunity characteristics depending on whether the
multiplicative factor is greater than, or less than, 100%. Figures 1 and 2, below, illustrate these properties where
the range of multipliers is 90%, 105% and 130% is applied to the activity durations during simulation.
Some risks are more realistically represented as discrete risks since the impact is specified in days, not multiplicative
factors or is applied to a milestone that has zero days duration when multiplied by any factor. Risk analyses can
include both risk drivers and discrete risks.

a CPM Model
July 9, 2019
The choice of distribution type will affect the overall results. It is more important to get honest opinions about the
3-point ranges than to worry about the shape of the distribution to use.
Applying Risk Drivers to Activities
Risks may be assigned to multiple similar activities using filters or categories, and groups of similar activities may be
affected by the same risk(s). Even if the risk is described at a general level it must be assigned (or mapped) to
activities at the most detailed level of the schedule. The risk may appropriately impact many activities. Identifying
a group of activities that belong to a category, such as engineering, piping, or structural foundations, helps make the
assignment of risks to groups of activities easier during modeling. When the risk occurs during an iteration of the
simulation model, it affects all activities to which it is assigned, with the same chosen adjustment duration factor
applied to each. If the risk occurs during the next iteration, a new duration factor, randomly selected by the software
from the three-point impact distribution, is similarly applied across all affected activities. If the risk does not happen
in some iteration, the activities’ durations and costs are unaffected, at least by that risk.3
If the activities have time-dependent resources (labor-type) the cost of the activity will increase proportionately to
the increase in duration. In addition, if there is a cost impact component to the risk, then the expenditure (or burn
rate) of labor or the total cost of material resources may vary as well. If the risk driver has only a cost-impact
component its cost may vary during simulation even when the duration is not varied.
Implementing a process where a risk may affect several or even many activities recognizes that each risk’s
importance is determined in part by its effect on all activities to which it is assigned. This is a beneficial and realistic
way to model the holistic impact of risks that influence multiple activities. Whether those activities are on risk-critical
paths4 or not in iterations is accounted for by the model during simulation.
Effect of Duration Uncertainty with 100% Likelihood
An illustration of setting up a risk driver in a simple schedule is shown in Figure 1. The highlighted Risk 101, Duration
Uncertainty, has 100% probability of spanning all values from minimum to maximum. The duration multiplicative
factor is expressed as a triangular distribution in this case, although other distributions are available, with 90%, 105%
and 120% for MIN, MOST LIKELY and MAX values.5
Figure 1: Typical Specification of Risk Drivers Using a Simple Schedule
3
Risk drivers can be assigned to activity costs as well as to durations. In addition, activity costs will be influenced by variability in durations in an
integrated cost-schedule risk analysis. [18] [20] [19]
4
The risk critical path is the path that is most likely to delay the project finish date when risks are taken into account. The identity of the risk
critical path may change from iteration to iteration during the simulation as some risks occur and others do not.
5
All screenshots in this RP are created in Safran Risk® by Safran Software Solutions. Showing these screen captures does not imply a
recommendation of this software by AACE International.

a CPM Model
July 9, 2019
Because the risk is 100% likely it appears as a triangle in the simulation (10,000 iterations) as illustrated by the
histogram for the resulting duration of one activity, Fabricate CPP Topsides shown in Figure 2, with no other risk
assigned to that activity.
Figure 2: Duration Distribution when Affected by a 100% Likely Risk
Effect of a Risk Driver with Less Than 100% Likelihood
In figure 3, Risk 106, Installation may experience coordination issues, is specified to impact the installation activities
and to occur 60% of the time.
Figure 3: Specifying a Risk Driver with a 60% Likelihood on the Installation Activities
Figure 4 shows a histogram of activity duration results for the activity Install CPP Topsides, where the risk is 60%
likely to occur. Risk 106 occurs during only 60% of the iterations, chosen at random by the simulation software. In
the other 40% of the iterations, when the risk does not occur, the duration of the activity Install CPP Topsides would
be shown as originally scheduled if no other risks were present. The modeled probability distribution shown in Figure
4 may look strange, but after a little thought it can be appreciated that:
• 40% of the results result in the original deterministic duration of 30 working days, while
• The remaining 60% of results are impacted by Risk 106, represented by a range of durations that spread to form
a triangular distribution.

a CPM Model
July 9, 2019
Figure 4: Simulated Distribution of an Activity Influenced by a 60% Likely Risk
Assigning Risk Drivers to Multiple Activities
Often the risk is described at a high level so it may impact multiple activities such as all fabrication work, all electrical
work or all scaffolding work. In this case shown in Figure 5 Risk 107, Suppliers may be busy, is assigned to the two
procurement activities. Notice that Risks 101 and 102 representing duration and cost uncertainty variation are
assigned to the entire schedule. Notice as well that Risks 110 and 111 are also widely assigned – these are the
systemic risk and the stressor condition discussed below.

a CPM Model
July 9, 2019
Figure 5: Illustrating the Assignment of a Systemic Risk on Multiple Activities
Risk Drivers Inherently Model How Activity Durations Become Correlated
It may be difficult to estimate correlation coefficients between activities’ durations in complex situations where
there are many risks. Using the risk drivers method, the correlation between activity durations is created
automatically during simulation, avoiding having to estimate or guess correlation coefficients.
Activity durations become correlated when some uncertain factor (i.e. a risk driver) affects both of them. Modeling
how correlation is generated, using a risk that is common to both activities, will not produce an inconsistent
correlation matrix. [11] Figure 6 illustrates how one risk applied to two different activities causes them to be 100%
correlated, while Figure 7 shows the subsequent scatter plot of the two activities’ durations.
Figure 6: One Risk Applied to Two Activities will Cause their Durations to be Correlated

a CPM Model
July 9, 2019
Figure 7: With One Common Risk the Two Activities’ Durations are 100% Correlated
When at least one additional risk affects one of two correlated activities but not the other, the correlation between
these two activities’ durations is reduced. The model in Figure 8 shows one common risk affecting two activities,
and two other risks affecting one activity each. This results in a correlation coefficient of 38% between the two
activities, as shown in Figure 9. Using the model and simulation runs to manage correlations naturally, frees the
analyst from having to guess at the coefficients between many activities’ durations.
Figure 8: Effect of Three Risks assigned Among Two Activities
There is a possibility that risk drivers are correlated because they share and respond to a common more fundamental
risk. In this case, the lack of correlation may lead to an underestimate of schedule risk. Ideally the risks are described
at the root cause level so they are independent. The main goal is to identify and justify the degree of correlation
used through simulation rather than determining the correlation coefficients by guessing. There is an iterative nature
of getting to the root cause of risks, and adjustments should be made as more is learned about the risks.

a CPM Model
July 9, 2019
Figure 9: Adding Two New Risks Each Applied to Only One Activity but not to the Other Results in a Correlation
Coefficient of 38%
Assigning Risks in Parallel or in Series
Risks can be assigned to influence an activity’s duration in series or in parallel. When activities have several risks
assigned that occur on the same iteration, the activity’s duration will depend on one of these two assignment modes.
Whether risks should be modeled in parallel or in series depends on the nature of the risks, not the activities to
which they are assigned:
• Some risks may be more important to fully recover from than others. Recovery from these important risks may
require resources that are key to finishing the activity or may be fundamental to the objective of the activity. In
such cases, these risks are entered in series with other risks.
• Some risks that occur can trigger or make more likely other risks. An example would be “misalignment of rotating
equipment” in installation could trigger “compressor control system fails” during commissioning.
• Risks that are less important (in that they take fewer scarce or senior resources) and therefore can be recovered
from, if they occur, simultaneously with other risks if they occur are entered in parallel with other risks.
Figure 10 shows how duration multipliers are applied against risks that operate either in series or in parallel. Multiple
risks affecting activities in series can drive extreme schedule overrun predictions. To illustrate this point, Figure 11
highlights the finish date if all risks are specified in series versus all risks in parallel.

a CPM Model
July 9, 2019
Figure 10: A Comparison of Duration Multipliers Applied Either in Series or in Parallel
Risks in
Parallel
Risks in Series
Figure 11: Effect of Risk Assigned Either in Parallel or in Series on an Activity’s Cumulative Distribution or S-Curve
Case Study - Model of Offshore Gas Production Platform Project
The following sections illustrate the use of risk drivers on a simple schedule of an offshore gas production platform
construction project shown in Figure 12.

a CPM Model
July 9, 2019
Figure 12: Example Summary Schedule for an Offshore Gas Platform Construction Project
Modeling Uncertainty
In this recommended practice the term “uncertainty” is defined as relating to background variability originating in
inherent variability, estimating error and estimating bias if it exists. [14] This definition is similar to that of common
cause variation. [12] Modeling uncertainty in activity durations can be done by applying common or reference ranges
to groups of similarly uncertain activities or to the entire project, durations and costs, as illustrated in Figure 3 above.
Since uncertainty is 100% likely to happen, uncertainty is represented solely by its impact range and the activities
they affect, but not with its probability of occurring.
Using risk drivers to represent uncertainty causes the durations (or costs) to be correlated, since if the risk occurs
for an iteration it occurs for all activities and if some multiplier, say 1.07, occurs for that iteration that factor is applied
to all activity durations. In that way the range specified, in this case for durations 95%, 105% and 120% is carried
through the simulation. Notice that the minimum and maximum durations for schedule uncertainty for the entire
project, shown in Figure 13, when calculated relative to the deterministic duration of 1,189 days, closely match the
specified input durations, reflecting the inputs from the interviewees.

a CPM Model
July 9, 2019
Figure 13: Uncertainty Applied using Risk Drivers (Correlated) Results Match the Interviewees’ input Ranges
Modeling Systemic Risks as Risk Drivers
In this recommended practice, project-specific and systemic risks are defined similarly to special cause variation.
[12]
Modeling the impact of risk drivers on a project schedule involves applying some risks to multiple activities while
some activities have multiple risks assigned to them. An example of a systemic risk applied as a risk driver is shown
in Figure 14. Highlighted is Risk 110 “The project team may not be adequate for the complex tasks”. This systemic
risk associated with the entire project delivery team and as such is assigned to all activities. It is deemed to be only
20% likely to occur with some effect, and if it occurs it will have a MIN, MOST LIKELY and MAX schedule impact of
110%, 120% and 140% schedule.
Figure 14: Uncertainty, Project Specific and Systemic Risk Drivers with Probability and Impact Ranges
Assignment of risks to multiple activities is shown in Figure 5 above.

a CPM Model
July 9, 2019
Schedule Risk Results Using Risk Drivers in a Monte Carlo Simulation
Standard schedule risk analysis results are shown in Figure 16. These results are represented by histograms and their
associated cumulative distribution functions. The 80th percentile date from the cumulative distribution is highlighted
as a matter of convenience, although the project owner may choose a more aggressive or more conservative level
of confidence respectively. Often the P-5 and P-95 values are used to describe the full or realistic range of the
distribution.
The stakeholder team should determine the P-value (probability of completing at or before the corresponding P-
value finish date) that is applicable to their organization based on their risk tolerance. For example, the 80th
percentile, or P-80 date highlighted implies that there is an 80% likelihood that the finish date will be on that 25 July
2024 or earlier. Eighty percent is viewed by many users as a conservative target level of certainty that provides a
contingency of time to accommodate the schedule uncertainty and risk events that are known at the time of the
analysis, plus providing some room to handle those risks that were unknown at the time of the analysis but may
become known in the future or to account for underestimates of the probability or impact of identified risks.
Figure 15: Schedule Duration Probability Distribution Histogram and S-Curve Results
The results in Figure 15 reflect the impact of defined uncertainty and risks events as shown above in Figure 14 and
after running 10,000 Monte Carlo iterations. Notice the bi-modal distribution at about the P-80 level of confidence.
This second mode is the product of the systemic team weakness risk and the stressor factor of schedule pressure.
This second node represents a 40%-50% overrun from the scheduled duration similar to that reported by IPA and
Hollmann. [23, 24]
The deterministic target date is 4 April 2023, which is shown to be only 1% likely to occur in this case study. This low
level of confidence in the original schedule is fairly common because of interacting factors:
• It may happen that the schedule durations are set at or near their most optimistic values, perhaps because of
optimism, pressure from management or the customer or from competition at bidding, leading to a shorter
schedule than anyone believes is feasible. If this type of bias or strategic misrepresentation is discovered during

a CPM Model
July 9, 2019
risk data collection, such bias needs to be addressed in the risk analysis, usually by uncertainty ranges that are
skewed to the right.
• Most of the risks were specified to be mostly or completely threats, hence impact ranges would be right-skewed
toward schedule overrun. In addition, there is a systemic risk that the project team may not be up to the task of
managing a complex project, leading to a partially bi-modal distribution of finish dates.
• There are some merge points in the schedule. This fact of the project plan creates a phenomenon, referred to
as merge bias [13], in which the start of a successor activity or milestone is riskier than the finish of any of the
preceding merging paths. Examples of where this situation may arise in the case study schedule include the:
Install Long-lead Equipment, Install Central Processing Plant (CPP) Topsides and Hook-up and Commissioning.
At each of these activities, the two predecessors must both finish on time or earlier for the activity to start on
time.
Prioritizing Risk Drivers for Management Action
In Figure 15, the P-80 date (incorporating all uncertainty and pre-mitigated risk events) is 25 July 2024, approximately
15½ months later than the deterministic date. A delay of this magnitude may be unacceptable to the project
manager, end-use customer and other project stakeholders. The next step that is needed is risk management, often
by mitigation, which can be facilitated using iterative simulation methodology to prioritize the risk drivers.
The risk drivers should be prioritized in order to facilitate the mitigation effort. Using identified risk drivers and/or
discrete risks to drive the simulation permits the prioritization of root-cause risks, leading to timelier and more
efficient risk responses.
Risk prioritization using risk drivers is derived from an iterative program of simulations, where each risk is disabled,
one at a time, to determine the P-80 (or whatever target level of certainty is desired) date that occurs without that
risk, as if it were fully mitigated. [16] This disable-and-simulate process is repeated for each risk. At the end of the
multi-simulation exercise, the risk that produced the earliest P-80 date when removed is deemed the most important
risk to mitigate because it has the greatest marginal impact measured in days on the P-80 finish date if fully
mitigated.
After the most important risk is identified, the next set of simulations eliminates the previously identified highest
priority risk and simulates the system with the other risks each disabled in turn to determine the second most
important risk that results in the earliest P-80 date during multiple “delete and replace driver” simulations after the
most important risk is mitigated. The method is repeated until all identified risk drivers and discrete risks have been
chosen and disabled with multiple simulations. The resulting tornado diagram, like that pictured in Figure 16,
identifies the number of days of impact if the risks were fully mitigated in their priority order. Notice that background
uncertainty is included in Figure 16 even though it may not be reducible because its causes are not known.

a CPM Model
July 9, 2019
Figure 16: Uncertainty and Risks Prioritized for Schedule Impact at the P-80 Confidence Level
This risk prioritization method improves on standard sensitivity approaches. Standard tornado diagrams sort
activities based on the correlation between activities durations and the finish date during simulation. Because the
identified risks have been used to drive the simulation the prioritization method described focuses on risks, not
activities. Results expressed in terms of days saved if the risk were completely mitigated (which is a goal unlikely to
be reached) and measured at the desired level (e.g., P-80) are easier to comprehend by the project team, compare
with the cost of treatment actions, and support effective decision-making. [14]
To help facilitate risk treatment the details associated with these prioritized risks can be presented in a table. In this
case, there are a total of 477 days of schedule contingency, from the scheduled date of 4 April 2023 to 25 July 2025.
Of these, only 347 days may be subject to risk treatment because uncertainty is generally not thought to be reducible
and accounts for 130 days.
Risk Treatment Using Prioritized Risks
Using the prioritized risks, the project stakeholders should meet to develop proposed risk treatment actions. After
those actions are identified, a formal risk treatment (mitigation) workshop may be convened, to include all relevant
stakeholders. The purpose is to agree which risk response plans will be implemented including who will take
ownership.
To successfully conclude the risk treatment workshop, any risk response plans that are adopted should be:
1. Different from and not duplicate what has occurred on the project to date.

a CPM Model
July 9, 2019
2. Agreed and committed to by all parties involved.

3. Should assign a risk treatment owner to high priority risks.
4. Budgeted and staffed.
5. Monitored for mid-course corrections.
Approved and agreed to treatments can be used to generate assessments for probability and impact parameters
that represent post-mitigated or residual risk for each related risk since it is often impossible to fully eliminate the
risks for all stakeholders. This residual risk exposure can then be modeled to generate a realistic post-treatment
result.
Integrating Cost and Schedule Risk Analysis
Cost and Schedule are Related
It would be counterintuitive to expect to find large schedule overruns but significant cost underruns. The reason is
that if an activity takes longer than the project plan and schedule imply, any labor-type resources, including labor
and equipment rented by the day, assigned to that activity will cost more. This is not a surprise, since the project
cannot expect labor or heavy earth-moving equipment, tower cranes or drill rigs to work without compensation.
There could be exceptions to this relationship. If the project manager directed a slow-down because of cash flow
limitations a delay would delay schedule but only affect the indirect costs that could not be stopped. This
recommended practice is intended to put risks and uncertainty on the current plan as represented in the current
schedule.6
If there are risks that affect the project schedule, they can be expected to affect the project cost as well, to the extent
that labor-type resources are used. This statement does not make a judgment about which project participant will
actually pay labor for time overruns. Many owners sign fixed-price contracts with contractors, hoping to transfer
cost risk to them.
Some owner participants in risk analysis will claim that the contractor bears all schedule risk because of a fixed price
contract. However, there are usually changes to the scope, conditions unknown at the time of contract signing, and
external circumstances that are best thought of as owner-borne risks. Fixed-price contracts often do not effectively
transfer risk to the contractor, but they often lead to contractors’ including their assessment of cost risk in the fixed-
price bid. [25, 26] Fundamentally, a significant amount of the cost contingency is derived indirectly from schedule
risks’ lengthening activities, including hammock activities holding indirect costs, and causing labor-type resources’
costs to rise.
Time-Dependent (Labor) Resources
The platform for integrated cost and schedule risk analysis is a resource-loaded CPM schedule. The resource loading
can be done with as few as two resource categories, distinguishing between labor and materials. Any resources used
and contributing to the cost of the project need to be expressed without included contingency to avoid double
counting cost overruns since the Monte Carlo simulation will re-estimate the cost contingency. The total project cost
without contingency should be represented by resources placed on activities.
6
One improvement over the analysis only of the current plan as it exists in the schedule could be made using conditional branching, a capability
that is available with only a few simulation software packages at present. For instance, if some early activity, e.g., detailed engineering
complete, is significantly late, the project manager might be able to add resources to shorten the execution phase, thus preserving as much
schedule s possible but with more cost than planned. This test is implemented on each iteration.

a CPM Model
July 9, 2019
The risk analysis process handles labor-type resources including rented equipment (e.g., earth-moving equipment,
tower cranes, drilling rigs, etc.) as these resources will cost more than planned if the associated activities take longer
than planned.
In simulation modeling of resource-loaded schedules, labor and other time-dependent resources will cost more in
proportion to the increase of the duration of the activity. While this is a convenient assumption it is also intuitive; if
a crew is scheduled to work 10 days and instead, to get the job done, it works 20 days, it is logical that the labor for
that activity will cost twice as much as estimated. The elongation of the work is not due to management’s cutting
the resources but rather management’s staying with the plan to have one crew working but perhaps with lower
productivity. This represents the integration in integrated cost-schedule risk analysis.
In addition to the impact of schedule risk on cost there are risks such as variability in quantity take-off or price per
quantity that directly affect cost. These direct cost risks take different forms depending on the nature of the resource
applied, and are implemented by a probability distribution of multiplicative factors applied to the daily expenditure
or burn rate. If labor costs are higher because of labor market tightness there could be an increase in cost even if
the schedule is stable. It is equally intuitive that management will not stand idly by when delays happen. Often, they
will add resources as part of corrective action. This may be modeled with an advanced capability of conditional
branching, which is not included in many software packages.
Level of effort or hammock activities can be loaded with time-dependent resources to represent indirect cost such
as project management and controls, management systems assigned to the project, quality assurance and
commissioning activities. Those resources’ time on the job depends on the size of this project management office
and on the duration of the project or phases they support. They are loaded with time-dependent labor-type
resources.
Notice that there is a cross product of risks to costs that impacts labor resources since labor costs can be affected
by; (1) risks to cost (burn rate), (2) risks to schedule that affect labor costs (longer duration), and (3) the cross product
of higher burn rate and longer duration.
Time-Independent (Material) Resource Costs
For material resources including raw materials and fabricated equipment to be installed, direct cost risks affect the
estimate of the total cost. Since they are time-independent the cost of material will not be indirectly affected by risk
to schedule. If the suppliers deliver later than planned there will not necessarily be a cost impact since the labor is
the suppliers’ to deal with, but there may be a schedule impact of late delivery.
Representing the Risk to Both Cost and Schedule – The Joint Confidence Level (JCL)
With total project costs without contingency loaded onto the CPM schedule activities as labor and material resources
the simulation calculates both finish date and total cost for each iteration reflecting uncertainty and risks for that
iteration. The cost and schedule results for any iteration reflect the same consistent set of assumptions.
The simulation produces histograms and cumulative probability distributions for both cost and finish date, so the
user can review the P-80 (or whatever level of confidence is needed) on cost or finish date separately. The schedule
risk shows a P-80 date of 25 July 2024 (see Figure 15). The cost risk histogram from the same simulation reports on
the effect of risks directly applied to cost and indirectly arising from the effect of schedule risk on time-dependent

a CPM Model
July 9, 2019
resource costs. This result is $2,720 million with a contingency above the base cost of $939 million or about 53%
overrun, largely because of the systemic team weakness risk, as shown in Figure 17.
Figure 17: Pre-Mitigated Cost Risk Results
The problem with these P-80 results from individual histograms for cost and finish date is that they are not sufficient
to ensure that both targets will be met simultaneously on the project. There are many cost results associated with
25 July 2024 and some of these are not 80% likely to happen. There are many finish dates associated with $2,485
million and some of these are not 80% likely either.
Promising the P-80 cost and P-80 date will not represent a P-80 for both targets. This result is seen when the scatter
diagram produced by the simulation is shown as in Figure 18. When the two P-80 values are graphed (cross-hairs)
the result is 76% likely for both time and cost together. This is shown by the highlighted “76%” in the lower-left
quadrant indicating that about 7,600 of the 10,000 iterations are south-west of the P-80 values. Clearly this project
needs to promise a later date and a higher cost to achieve both targets simultaneously.

a CPM Model
July 9, 2019
Figure 18: Cost and Finish Date Scatter Diagram of Pre-Mitigated Risk Results
Figure 18 shows a color scheme following “JCL” bands. JCL stands for Joint Confidence Level, which is a name for
integrated cost and schedule risk analysis results coined by the US National Aeronautical and Space Administration
(NASA). There is some analysis showing that, since NASA adopted the JCL for their requests to US Congress, their
ability to achieve the stated targets has improved. [27] This result does not mean that NASA are suddenly better
project managers but that they are better at forecasting where their projects will end up. For large publicly-funded
projects this improvement in accuracy of budget requests is useful.
The scatter diagram in Figure 18 should be interpreted as a 3-dimension ridge of probability of duration and finish
date’s occurring. The project should try to find the most likely pair of cost-finish date results that achieve an 80%
chance (called the “JCL-80) of their both finishing on target or better. Finding the highest point achieving just 80%
likelihood or better is equivalent to following a trail on a topographical map at the same joint cost-schedule
probability and finding the highest such point. This is equivalent to identifying the spot of highest concentration
along the “JCL: 80-100%” shaded area of the scatter diagram. One such point is shown in Figure 19.

a CPM Model
July 9, 2019
Figure 19: Identifying the Most Likely Cost and Finish Date Pair that Exhibits a JCL-80
While there is no agreed-upon method to finding the most likely JCL-80 point, [26] the point chosen in Figure 19
indicates about $196 million more budget and 9 weeks more schedule. These values would be larger if cost and
schedule were not so highly correlated. In this figure they are correlated 86%. If cost and finish date were perfectly
correlated the P-80 and JCL-80 values would coincide, but if the correlation is lower, for instance after modeling
post-mitigated values, the JCL correction factor would be more than shown for this case study.
SUMMARY
In summary, using risk drivers for the identifiable project-specific risks to the schedule produces some significant
benefits for integrated cost-schedule risk analysis:
• Using risk drivers to represent individual, identified root cause risks to drive the Monte Carlo simulation is
consistent with Recommended Practice 40R-08, Contingency Estimating – General Principles. [5]
• Risk drivers allows the use of the project schedule that maintains integrity of the modeling consistent with the
project plan and facilitates capturing cumulative effects of risk through the project logic and correlations
between activity durations. Correlation between time and cost is one result of the simulation that can be seen
from the scatterplot and influences the need for analyzing the Joint Confidence Level of achieving both targets.
• Since risk drivers provide the variability in activity durations that drive the simulation, their individual impact on
the risked pre-treatment finish date can be identified at the margin using progressive simulation with risk
disabling methods leading to risk prioritization. Risk prioritization is important for effective risk mitigation.
• The prioritization method reports days of impact if the risk is mitigated, which is actionable information to
management because it contributes to cost-benefit analysis.
• Risk drivers can be assigned in parallel with other risks or in series. For risks applied in series their impacts
compound during simulations, sometimes causing very long activity durations.

a CPM Model
July 9, 2019
• Risk drivers can represent both opportunity and threat characteristics depending on whether the adjustment
impact factor is less than or greater than 1.0. Also, a risk driver’s impact range can span from opportunity to
threat.
• Assigning root cause risks to multiple activities is facilitated in a way that is intuitive. Since risk driver impact
ranges are compounding factors, they can be readily applied to both longer and shorter activities. Assigning risk
drivers to multiple activities allows the analyst to represent the holistic effect of a risk that affects many
activities.
• The risk driver approach inherently models the correlation between activity durations when they are affected
by a shared risk. During Monte Carlo simulation, the correlation coefficients between activity durations are
generated and can be viewed.
• Cost and schedule risks can be integrated in the sense that labor-type resources are time-dependent. If those
activities take longer to complete, the labor, rented equipment, and other resources will cost more than
planned. Some risks will affect the daily expenditure rate of time-dependent resources or total cost of time-
independent resources such as material and equipment to be installed, independently from the risk to activity
durations. In addition, there is a possibility that in schedule driven projects resources will be added if the
schedule finish date is seen to be in serious jeopardy, so a cost increase may be disproportionate to schedule.
• With integrated cost-schedule risk analysis a method called the Joint Confidence Level exists to determine how
much extra cost and duration will be needed to achieve both time and cost targets together. The less time and
cost are correlated, mainly through less labor and more material resource being planned, the more the JCL-80
values will be greater than the P-80 values for cost and time considered separately even with an integrated cost-
schedule model.
REFERENCES
[1] D. T. Hulett, Practical Schedule Risk Analysis (starting at chapter 8), Gower Publishing Limited, 2009.
[2] AACE International, Recommended Practice No. 42R-08, Risk Analysis and Contingency Determination Using
Parametric Estimating, Morgantown, WV: AACE International, Latest revision.
Expected Value, Morgantown, WV: AACE International, Latest revision.
[4] AACE International, Recommended Practice No. 64R-11, CPM Schedule Risk Modeling and Analysis: Special
Considerations, Morgantown, WV: AACE International, Latest revision.
[5] AACE International, Recommended Practice No. 40R-08, Contingency Estimating – General Principles,
Morgantown, WV: AACE International, Latest revision.
[6] D. H. Rumsfeld, Defense.gov News Transcript: "DoD News Briefing - Secretary Rumsfeld and Gen. Myers", US
Department of Defense, February 12, 2002.
[7] D. T. Hulett, "(RISK-2890) Journey-Map to a More Mature Schedule Risk Analysis (SRA) Process," in AACE
International Transactions, Morgantown, 2018.
[8] D. Hillson and R. Murray-Webster, Understanding and Managing Risk Attitude, Gower Publishing Limited,
2005.
[9] AACE International, Recommended Practice No. 114R-20, Project Historical Database Development,
Morgantown, WV: AACE International, Latest revision.
[10] B. Flyvbjerg, "From Nobel Price to Project Management; Getting Risks Right," Project Management Journal,
August 2006.
[11] S. Book, "A Theory of Modeling Correlations for Use in Cost-Risk Analysis," in Third Annual Project
Management Conference, Galveston, 2006.

a CPM Model
July 9, 2019
[12] AACE International, Recommended Practice No. 10S-90, Cost Engineering Terminology, Morgantown, WV:
AACE International, Latest revision.
[13] K. MacCrimmon and C. Ryavec, "“An Analytical Study of the PERT Assumptions,” Research Memorandum
RM-3408-PR," Rand Corporation, Santa Monica, 1962.
[14] E. Druker, G. Gilmer and D. Hulett, "(RISK-1662) Using Stochastic Optimization to Improve Risk Mitigation," in
AACE International Transactions, Morgantown, 2014.
[15] H. L. Stephenson, Ed., Total Cost Management Framework: An Integrated Approach to Portfolio, Program
and Project Management, 2nd ed., Morgantown, WV: AACE International, Latest revision.
[16] AACE International, Recommended Practice No. 65R-11, Integrated Cost and Schedule Risk Analysis and
Contingency Determination Using Expected Value, Morgantown: AACE International, Latest revision.
Range Estimating, Morgantown, WV: AACE International, Latest revision.
[18] D. T. Hulett, Integrated Cost-Schedule Risk Analysis (starting at chapter 8), Gower Publishing Limited, 2011.
[19] D. T. Hulett and E. Druker, "Integrating with Project Schedule Risk Improves Analysis of Cost Risk," Cost
Engineering, no. May/June, 2015.
[20] D. T. Hulett and M. T. Nosbisch, "Integrated Cost-Schedule Risk Analysis," Cost Engineering, no.
November/December, 2012.
[21] E. Uyttewaal, "Forecast Scheduling with Microsoft Project 2012," ProjectPro Corporation.
[22] US Government Accounting Office (GAO), "Schedule Assessment Guide: Best Practices for Project Schedules,
GAO-16-89G," US Government Accounting Office (GAO), Washington, 2015.
CONTRIBUTORS
Disclaimer: The content provided by the contributors to this recommended practice is their own and does not
necessarily reflect that of their employers, unless otherwise stated.
July 9, 2019 Revision:

Dr. David T. Hulett, FAACE (Primary Contributor)
Waylon T. Whitehead (Primary Contributor)
James E. Arrow, DRMP
William A. Banks, Jr.
David C. Brady, P.Eng. DRMP
Larry R. Dysert, CCP CEP DRMP FAACE Hon. Life
John K. Hollmann, PE CCP CEP DRMP FAACE Hon. Life
Donny Lai
John R. Schuyler, PE CCP DRMP
Edward J. Thomas
Warner W. Uhl
June 18, 2011 Revision:

Dr. David T. Hulett (Primary Contributor)
Christopher P. Caddell, PE CCE
Tommy Clarke
Dr. Ovidiu Cretu, PE

a CPM Model
July 9, 2019
Kevin M. Curran
Michael W. Curran
Patrick B. Egger
Ricardo Garcia da Roza
John M. Hale
Dennis R. Hanks, PE CCE
John K. Hollmann, PE CCE CEP
Donald F. McDonald, Jr. PE CCE PSP
Oscar A. Mignone
Stephen P. Warhoe, PE CCE CFCC
Robert F. Wells, CEP
Dr. Trefor P. Williams
Ronald M. Winter, PSP
David C. Wolfson
Rashad Z. Zein, PSP


57R-09 - Ntegrated Cost and Schedule Risk Analysis

Uploaded by

Copyright:

Available Formats

57R-09 - Ntegrated Cost and Schedule Risk Analysis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

57R-09 - Ntegrated Cost and Schedule Risk Analysis

Uploaded by

Copyright:

Available Formats

57R-

INTEGRATED COST AND SCHEDULE RISK ANALYSIS USING

Rev. July 9, 2019

July 9, 2019 Revision:

June 18, 2011 Revision:

Copyright © AACE® International AACE® International Recommended Practices

Copyright © AACE® International AACE® International Recommended Practices

Copyright © AACE® International AACE® International Recommended Practices

Preconditions for a Schedule Risk Analysis

A CPM Schedule that Complies with Industry Practices

Copyright © AACE® International AACE® International Recommended Practices

Risk Data Collected to Improve Data Quality

Copyright © AACE® International AACE® International Recommended Practices

Properties of Risk Drivers

Risk Drivers and Discrete Risks

Copyright © AACE® International AACE® International Recommended Practices

Applying Risk Drivers to Activities

Effect of Duration Uncertainty with 100% Likelihood

Figure 1: Typical Specification of Risk Drivers Using a Simple Schedule

Copyright © AACE® International AACE® International Recommended Practices

Figure 2: Duration Distribution when Affected by a 100% Likely Risk

Effect of a Risk Driver with Less Than 100% Likelihood

Copyright © AACE® International AACE® International Recommended Practices

Figure 4: Simulated Distribution of an Activity Influenced by a 60% Likely Risk

Assigning Risk Drivers to Multiple Activities

Copyright © AACE® International AACE® International Recommended Practices

Figure 5: Illustrating the Assignment of a Systemic Risk on Multiple Activities

Risk Drivers Inherently Model How Activity Durations Become Correlated

Copyright © AACE® International AACE® International Recommended Practices

Figure 8: Effect of Three Risks assigned Among Two Activities

Copyright © AACE® International AACE® International Recommended Practices

Assigning Risks in Parallel or in Series

Copyright © AACE® International AACE® International Recommended Practices

Figure 10: A Comparison of Duration Multipliers Applied Either in Series or in Parallel

Case Study - Model of Offshore Gas Production Platform Project

Copyright © AACE® International AACE® International Recommended Practices

Copyright © AACE® International AACE® International Recommended Practices

Modeling Systemic Risks as Risk Drivers

Assignment of risks to multiple activities is shown in Figure 5 above.

Copyright © AACE® International AACE® International Recommended Practices

Schedule Risk Results Using Risk Drivers in a Monte Carlo Simulation

Copyright © AACE® International AACE® International Recommended Practices

Prioritizing Risk Drivers for Management Action

Copyright © AACE® International AACE® International Recommended Practices

Risk Treatment Using Prioritized Risks

Copyright © AACE® International AACE® International Recommended Practices

2. Agreed and committed to by all parties involved.

Integrating Cost and Schedule Risk Analysis

Cost and Schedule are Related

Time-Dependent (Labor) Resources

Copyright © AACE® International AACE® International Recommended Practices

Time-Independent (Material) Resource Costs

Copyright © AACE® International AACE® International Recommended Practices

Figure 17: Pre-Mitigated Cost Risk Results

Copyright © AACE® International AACE® International Recommended Practices

Copyright © AACE® International AACE® International Recommended Practices

Copyright © AACE® International AACE® International Recommended Practices

Copyright © AACE® International AACE® International Recommended Practices

July 9, 2019 Revision: