AACE International Recommended Practice No.

65R-11

INTEGRATED COST AND SCHEDULE RISK ANALYSIS AND CONTINGENCY

DETERMINATION USING EXPECTED VALUE
TCM Framework: 7.6 – Risk Management

Rev. May 2, 2012

Note: As AACE International Recommended Practices evolve over time, please refer to www.aacei.org for the latest revisions.

Contributors:
John K. Hollmann, PE CCE CEP (Author) Leonard Enger, CCE
Peter R. Bredehoeft, Jr., CEP Dennis Read Hanks, PE CCE
Christopher P. Caddell, PE CCE Carlton W. Karlik, PE CEP
Larry R. Dysert, CCC CEP Peter W. van der Schans, CCE ICC PSP

Copyright © AACE® International AACE® International Recommended Practices

AACE® International Recommended Practice No. 65R‐11
INTEGRATED COST AND SCHEDULE RISK ANALYSIS
AND CONTINGENCY DETERMINATION USING
EXPECTED VALUE
TCM Framework: 7.6 – Risk Management

May 2, 2012

INTRODUCTION

Scope

This recommended practice (RP) of AACE International (AACE) defines general practices and considerations for
integrated cost and schedule risk analysis and estimating contingency using expected value methods.

Purpose

This RP is intended to provide guidelines, not standards, for contingency estimating that most practitioners would
consider to be good practices that can be relied upon and that would be recommend for use where applicable.
There is a range of useful risk analysis and contingency estimating methodologies; this RP will help guide
practitioners in developing or selecting appropriate methods for their situation.

This RP is an extension of 44R‐08, Risk Analysis and Contingency Determination Using Expected Value, that
addresses using expected value methods only for cost. However, integrated cost and schedule methods are
generally recommended; this RP for expected value methods, or 57R‐09, Integrated Cost and Schedule Risk
Analysis Using Monte‐Carlo Simulation of a CPM Model, for CPM‐based methods.

Background

This RP is an extension of 44R‐08 which covers cost contingency only; both RPs are intended to be used together.
As described in 44R‐08, expected value has been in common use for both decision and risk management. Expected
value in its most basic form can be expressed as follows:

Expected Value = Probability of Risk Occurring x Impact If It Occurs

This calculation has long been a fundamental method used in decision tree analysis and risk screening. Its use is
common because it is quantitative, simple to understand, simple to calculate, and it explicitly links risk drivers with
their impacts so that the risks can be managed. However, its use for integrated cost and schedule risk analysis and
contingency estimating has been minimal, largely because of the popularity of Monte‐Carlo simulation software
using CPM schedules (see 57R‐09).

Expected value method has advantages and disadvantages that are described in 44R‐08. However, integrated
expected value has additional advantages over CPM‐based methods for specific situations. First, a precondition of
CPM‐based methods as described in 57R‐09 is having a “high quality CPM” schedule. Unfortunately, empirical
research indicates that high quality CPM schedules at the time of project authorization are the exception in
industry [3] (further, the study showed that high quality schedules correlate with 23% less schedule slip; i.e., poor
schedule quality is a systemic risk in itself.) However, high quality CPM schedules are not necessary for expected
value methods. Second, AACE’s 27R‐03, Schedule Classification System establishes that a critical path is not usually
available for Class 5 or 4 schedules that are developed using methods such as bar charts. Hence, expected value
methods can be considered more versatile in that they can be applied to any project plan of any Class or quality

Copyright © AACE® International AACE® International Recommended Practices

65R‐11: Integrated Cost and Schedule Risk Analysis and Contingency Determination Using Expected 2 of 7
Value

May 2, 2012

(albeit, the lower the quality of the base plan, the lower the quality of the risk analysis and contingency estimate).
AACE is not recommending toleration of poor practices, but practitioners must often work within less than ideal
circumstances.

Also, the effects of compounding, cascading, and/or dynamic logic risks are very difficult to model in deterministic
CPM schedules (which are typically based on fixed logic). These aggravating risk effects can be conceptually
identified by elicitation or through experience and applied in expected value (the method tends to force this
consideration). Also, expected value can be used for bounds testing of deterministic CPM models to factor in the
presence of strategic or chaos‐threatening risks.

The traditional expected value method (probability x impact) with Monte Carlo can be applied to both costs and
durations. However, unlike costs for which risk impacts are essentially cumulative, the time impacts to schedule
activities are not if there are multiple “paths” and the impact is to a non‐critical path. Also, a risk time impact to
one path can offset another risk that impacts a separate path providing that these impacts are also not cumulative.
The CPM method deals with these complications directly; the expected value method does not and hence requires
more subjective schedule analysis and intuitive understanding than CPM‐based methods. Therefore, this RP
highlights the importance of strong planning/scheduling knowledge and risk analysis facilitation.

The method in this RP integrates cost and schedule by using explicit assessment of combined cost‐schedule
impacts of a risk in consideration of potential risk responses (and their cost‐schedule tradeoffs in respect to project
objectives). This provides up‐front understanding, or at least consideration of possible future project team
behavior, as opposed to just simple mathematical derivations of schedule impacts. An additional outcome of this
practice is therefore a contingent response plan.

It is recommended that users consider application of both CPM and expected value methods where possible in
order to gain the benefits of both. For example, apply the expected value method to every project in a portfolio,
large and small, regardless of the quality of base plan. This gives some insight into potential project “system
behavior” including human factors, feedback loops, potential chaos, etc. Then, for major or strategic projects
where quality schedules are more typical, apply the CPM‐based methods for its insight into mechanical aspects of
schedule behavior.

Background – Risk Types

As discussed in 44R‐08, the expected value method of contingency estimating explicitly links risk drivers with their
impacts. This requires explicit understanding and treatment of the risk types. Risks fall into one of two categories;
risks that have systematically predictable relationships to overall project cost and schedule growth outcomes and
those that don’t. These categories have been labeled as systemic and project‐specific risks for contingency
estimating purposes. Parametric risk analysis methods are generally recommended for systemic risks such as the
impact of poor quality planning (see 42R‐08, Risk Analysis and Contingency Determination Using Parametric
Estimating). This RP explains how parametric and expected value contingency estimating methods can be used
together in a way that best addresses both systemic and project‐specific risks.

RECOMMENDED PRACTICE

The following steps assume that a formal risk management process is being followed and that risks have already
been mitigated in the project plans to some extent. This recommended practice then addresses the residual risks
that need to be funded, incorporated into plans, controlled and managed. Teams that skip the risk mitigation
effort, in the interest of saving time, and go directly to contingency estimating may miss many of the benefits of

Copyright © AACE® International AACE® International Recommended Practices

65R‐11: Integrated Cost and Schedule Risk Analysis and Contingency Determination Using Expected 3 of 7
Value

May 2, 2012

risk management.

RISK IDENTIFICATION

Identify Residual Project‐Specific Risks

See RP 44R‐08, Risk Analysis and Contingency Determination Using Expected Value. The identification of risks is
handled the same with integrated cost and schedule methods as with costs alone.

QUANTIFICATION/CONTINGENCY ESTIMATING

The risk identification step will result in a list of significant risks and opportunities for which probability of
occurrence and impacts need to be estimated.

Estimating the Probability of Occurrence

See 44R‐08, Risk Analysis and Contingency Determination Using Expected Value. The probability of occurrence of a
risk is handled the same with integrated cost and schedule methods as with costs alone.

Estimating the Impact if the Risk Occurs and Screening for Critical Risks

See 44R‐08, Risk Analysis and Contingency Determination Using Expected Value. Screening for critical risks is the
same as for costs alone; however, for integrated cost and schedule risk analysis, the significance threshold must
also consider schedule impacts. Significance can be judged using the same criticality criteria cited in 44R‐08 as
derived from 41R‐08, Risk Analysis and Contingency Determination Using Range Estimating as shown in Table 1
below:

Bottom Line Critical Variances

Bottom Line Conceptual Estimates (AACE Detailed Estimates (AACE
(Cost or Profit) Classes 3, 4, 5) Classes 1, 2)
Cost Δ ± 0.5% ± 0.2%
Profit Δ ± 5.0% ± 2.0%
Table 1 – Suggested Critical Variance Thresholds for Screening Risks

Schedule impacts of a risk tend to have associated cost impacts and therefore the cost critical variances apply to
that risk. However, even if a schedule risk has no capital cost impact, it can significantly affect profit which requires
consideration of revenue and expense cost streams (e.g., a >±5.0% change in net present value is significant).

There may also be other measures of significance that apply such as loss of reputation or contractual or regulatory
schedule commitments. In these cases, an explicit “schedule Δ” entry may be included in the table as appropriate.
It is more difficult to generalize on schedule than costs because it depends on project objectives. Each project must
consider what the significance thresholds are, but typically if more than about 15 residual risks are critical the
screening has not been disciplined enough, and/or risk treatment has been ineffective.

Copyright © AACE® International AACE® International Recommended Practices

65R‐11: Integrated Cost and Schedule Risk Analysis and Contingency Determination Using Expected 4 of 7
Value

May 2, 2012

Assessing Ranges of Impact

As described in 44R‐08, Risk Analysis and Contingency Determination Using Expected Value, the range of cost
impacts of each critical risk is estimated, and when using Monte‐Carlo simulation, probability distributions are
assigned. The same general practice is used for schedule duration impact of the given risk in that a range is
estimated and distribution chosen (which may differ from the schedule distribution if the cost impact is not
entirely time‐dependent). Figure 1 provides a simplified example of the estimate of cost and schedule impact for
one risk (that has a probability of occurrence of 50%, assuming a triangular probability distribution).

Project-Specific Risk Event Cost Impact Range (thousands $) Expected Value Schedule Impact Range (weeks) Expected Value
No. Description Low Most Likely High (prob x impact) Low Most Likely High (prob x impact)
4 >50 year rain storm occurs $ 100 $ 140 $ 200 $ 73 2.0 3.0 6.0 1.8
during site preparation Activity(s)? Site Preparation
Criticality? Impact shown is to project completion
Parallels? Risk #6 impact to eqpmt. procurement
Assumed Risk Response: Either accept (demobilize and allow to drain/dry), or mitigate (mobilize pumping, equipment and crews to recover)
Figure 1 – Cost and Schedule Impact Estimates for a Risk‐Event or Condition

Care must be taken that the cost and schedule impacts for a given risk reflect the same risk event and the same
assumed risk response (or range of potential responses) in consideration of project objectives. This attention to
potential risk responses is a key value‐added feature of the expected value method.

Considering Risk Response(s)

During project execution, for every risk that occurs, the project team will make some response (i.e., accept, repair,
recover, etc.). During the risk analysis, the team must discuss the expected response or range of responses
considering the project objectives. Because each response reflects a potentially business‐critical decision (these
are all critical risks), it is imperative that the project business sponsor or representative (i.e., stakeholder that sets
the overall objectives) be part of the analysis discussion.

Using Figure 1 as an example, if the risk was a >50 year frequency rain storm, the team may determine that there
are two likely responses (or bounds of responses): (a)‐accept the impact; do nothing but demobilize and minimize
cash flow until the site is accessible and dries out, or (b)‐mitigate the impact; mobilize resources to quickly drain
and dry out the site, including possibly changing logic to recover the schedule. The impact of (a) is assessed as low
cost and long duration; (b) is high cost and short duration. If the project is cost‐driven, response (a) would be
assumed, and the integrated cost and schedule impact considering this response would be estimated. If the project
is schedule‐driven, response (b) would be assumed and integrated impacts estimated. If the business was not clear
on project objectives (i.e., not sure about cost/schedule trade‐offs), then the impact estimate range and
distribution must accommodate either response; i.e., there will be more uncertainty and a wider range of impacts.
The example in Figure 1 reflects this type of inconclusiveness/indecisiveness.

This team dialogue regarding responses, including the business representative, provides for an immediate
understanding of impacts during a risk analysis. If the responses are noted as in Figure 1, a contingent response
plan is documented as a by‐product of the analysis. It is not unusual that this in‐depth dialogue results in the
business representative stopping the risk analysis and recycling the scope back through risk treatment to lessen
the residual risks or to clarify their own expectations and project objectives (i.e., if they can’t readily decide on
responses now, they won’t be able to when the risk occurs either; this lack of understanding of objectives and
decisiveness is a systemic risk). This discussion also requires time, so it is imperative to not encumber the analysis
with trivial risks.

Copyright © AACE® International AACE® International Recommended Practices

65R‐11: Integrated Cost and Schedule Risk Analysis and Contingency Determination Using Expected 5 of 7
Value

May 2, 2012

Addressing Cost and Schedule Dependency

By estimating the impacts with respect to assumed responses, basic cost and schedule integration is established;
however, there is one more integration consideration. Unless the cost impact is purely time‐dependent, it is not a
given that the boundary values of the cost range for a given risk correspond to the boundary values of the
schedule impact; they may be negatively correlated. For example, in Figure 1, the potential range of responses
included one that is low cost/high duration (accept), and the other is high cost/low duration (recover). Therefore,
when setting up the Monte‐Carlo simulation model, the correlation of the cost and schedule impact distributions
for each risk must be estimated. For example, the cost‐schedule correlation coefficient is 1.0 when costs are purely
time dependent; but may be closer to ‐1.0 when the impact ranges cover multiple responses representing
alternate cost‐schedule tradeoff scenarios.

Addressing Criticality Regarding Schedule Impacts

After the range and distribution of impacts are estimated, a Monte‐Carlo simulation will be run resulting in overall
cost and schedule impact distributions. In most cases, the schedule distribution of concern is the time impact on
the project completion milestone. So, when estimating the schedule impact of a given risk, the duration entered is
the impact on overall completion. However, if the activity(s) impacted by the given risk is/are not on or near the
critical path, the analyst must take care not to overstate the impact on the completion milestone. For example, if
the impact on an activity is 4 weeks and there is 6 weeks float in that path, there is no net impact to project
completion. In lieu of CPM probabilistic modeling, this requires subjective assessment and strong schedule
understanding by the project planner/scheduler of both the schedule logic and how that logic may be affected
(possibly significantly changed) by the risk response. It should be noted that for major risks, the post‐risk event
logic (including the new response activities and their logic) will not be the same as that in the original schedule.
Subjectivity is therefore always required.

This subjective input is both a challenge and strength of the method. It means the analysis is only as good as the
scheduler’s knowledge and expertise (but that applies to the base schedule as well). On the other hand, it also
forces the team to consider the fact that schedule logic changes after major risk events happen (and all risks being
quantified are critical). This dynamic thinking tends to push the team to consider that impacts could be worse than
a quasi‐deterministic, mechanical CPM model suggests. In any case, the risk analysis facilitator must push the team
to consider issues such as dynamic logic and compounding/cascading impacts with multiple risk occurrences
(which threaten chaos when plans and competencies are weak). As with range estimating (41R‐08, Risk Analysis
and Contingency Determination Using Range Estimating), this pushing or provoking of team thinking to include the
worst possible impact (short of absurdity), is a key element of good risk analysis and contingency estimating.

Addressing Impacts of Various Risks on Parallel Schedule Paths (Buying Time)

As mentioned, a Monte‐Carlo simulation will be run resulting in a distribution of the time impact on the project
completion milestone. The Monte‐Carlo model is based on adding the schedule impacts of each risk. However,
unlike costs, the schedule impacts may not be additive. For example, assume there are two parallel near‐critical
paths in the schedule (paths 1 and 2) and two risks (risks A and B). Assume that during a Monte‐Carlo iteration that
risk A occurs adding 4 weeks to Path 1 and Risk B also occurs at the same time adding 3 weeks to Path 2. In this
case, the approximate net impact will not be additive to the overall completion; the overall impact with both risks
occurring will be about 4 weeks (the Risk A impact bought time for Risk B). If you allow them to be added in a
Monte‐Carlo run, you might overstate the risk. In practical terms, unless these two risks are extreme in impact,
adding these risk impacts may be an acceptable (or even preferred) approach that balances the team’s tendency to
underestimate impacts when multiple major risk events occur at the same time (potential chaos trigger). If it is felt

Copyright © AACE® International AACE® International Recommended Practices

65R‐11: Integrated Cost and Schedule Risk Analysis and Contingency Determination Using Expected 6 of 7
Value

May 2, 2012

that the overstatement is egregious, this can be addressed in the Monte‐Carlo model by making the schedule
impacts of these two risks negatively correlated (the impacts will then not be added). As with criticality, this
consideration of parallel paths requires subjective input and can be both a challenge and strength of the method.

Coordinate with Contingency Estimates for Systemic Risks

See 44R‐08, Risk Analysis and Contingency Determination Using Expected Value, but with the added note that
parametric models of the schedule impact of systemic risks can be used in conjunction with cost models.

Assessing Overall Outcome using Monte‐Carlo

Having quantified and defined distributions to the probabilities and cost impacts, and having established
dependencies between the risks, the cost and schedule risk model can be run through a Monte‐Carlo simulation
using one of several commercial software packages available.

The cost risk model input includes the base estimate plus the parametric model outcome distribution (e.g.,
systemic risk impact) plus the products of the distribution of probability times the distribution of the cost impact
for each project‐specific risk. The same applies for the schedule side of the model, with the exception that for risks
affecting parallel paths, a correlation matrix may be included to make the schedule impact distribution of the risks
affecting parallel paths negatively correlated (this is optional and generally only for risks with significant schedule
impacts).

An advantage of the expected value method is that the cost and schedule impacts of each risk are quantified.
While it is recommended that there be only one contingency account in a project cost budget, it can be useful for
later risk management and contingency drawdown to have the potential impact of each risk explicitly quantified
(i.e., if the risk does not occur, it provides an indication of the potential contingency, pending ongoing risk analysis,
that could be returned to the business).

Application of schedule contingency in the final schedule plan is more complex than with costs. One approach may
be to include a single schedule buffer activity to accommodate the contingency duration; however, there are other
methods that could be considered [7].

Estimating Contingency

See 44R‐08, Risk Analysis and Contingency Determination Using Expected Value; the estimation method using the
probabilistic output of the Monte‐Carlo model is the same for cost and schedule duration.

P50 vs. Expected Value

See 44R‐08, Risk Analysis and Contingency Determination Using Expected Value; the fact that p50 and expected
value are generally not the same is true for both cost and schedule duration.

Evaluating Contingency (vs. Reserves or Other Treatment)

See 44R‐08, Risk Analysis and Contingency Determination Using Expected Value; the issues regarding treatment of

Copyright © AACE® International AACE® International Recommended Practices

65R‐11: Integrated Cost and Schedule Risk Analysis and Contingency Determination Using Expected 7 of 7
Value

May 2, 2012

the impacts as contingency or reserves are similar for both cost and schedule. However, schedule contingency
management is a less defined area of practice.

SUMMARY

This RP is intended to guide practitioners in developing or selecting appropriate methods for their situation. The RP
provides processes that users can incorporate into more complex or refined models for risk analysis and
contingency determination. As discussed, the cost impacts could be assessed for major cost accounts and not just
in total. Users are encouraged to study the reference materials including the RPs for alternate methods and seek
ways to apply the methods that work best in their situation.

REFERENCES

1. AACE International Recommended Practice No. 44R‐08, Risk Analysis and Contingency Determination Using
Expected Value, AACE International, Morgantown, WV, (latest revision).
2. AACE International Recommended Practice No. 57R‐09, Integrated Cost and Schedule Risk Analysis Using
Monte‐Carlo Simulation of a CPM Model, AACE International, Morgantown, WV, (latest revision).
3. Griffith, Dr. Andrew F., Scheduling Practices and Project Success, Cost Engineering Journal, AACE International,
Morgantown WV, September 2006.
4. AACE International Recommended Practice No. 27R‐03, Schedule Classification System, AACE International,
Morgantown, WV, (latest revision).
5. AACE International Recommended Practice No. 42R‐08, Risk Analysis and Contingency Determination Using
Parametric Estimating, AACE International, Morgantown, WV, (latest revision).
6. AACE International Recommended Practice No. 41R‐08 Risk Analysis and Contingency Determination Using
Range Estimating, AACE International, Morgantown, WV, (latest revision).
7. Douglas, Edward E., Managing Schedule Contingency, 2010 AACE International Transactions, AACE
International, Morgantown, WV. 2010.

CONTRIBUTORS

John K. Hollmann, PE CCE CEP (Author)

Peter R. Bredehoeft, Jr., CEP
Christopher P. Caddell, PE CCE
Larry R. Dysert, CCC CEP
Leonard Enger, CCE
Dennis Read Hanks, PE CCE
Carlton W. Karlik, PE CEP
Peter W. van der Schans, CCE ICC PSP

Copyright © AACE® International AACE® International Recommended Practices