PERFORMANCE
QUANTITATIVE RISK .
ANALYSIS
PERFORMING QUANTITATIVE
RISK ANALYSIS
Quantitative risk analysis often follows qualitative risk
analysis, yet both processes can be done together or
separately.
Large, complex projects involving leading-edge
technologies often require extensive quantitative risk
analysis.
Quantitative risk analysis and modeling techniques are:
Decision tree analysis
Simulation and
Sensitivity analysis
DECISION TREE AND
EXPECTED MONETARY VALUE
A decision tree is a diagramming analysis technique
used to help select the best course of action when
future outcomes are uncertain.
A common application of decision tree analysis
involves calculating expected monetary value(EMV).
Example: Suppose Cliff Branch’s firm was trying to
decide if it should submit a proposal for Project 1,
Project 2, both projects, or neither project.
To create a decision tree, and to calculate expected
monetary value specifically, you must estimate the
probabilities or chances of certain events occurring.
the EMV for Project 1 is
$28,000.
.2($300,000)+.8($40,000
)
= $60,000-$32,000
= $28,000
The EMV for Project 2 is
$30,000
.2( $-50,000) +.1( $-
20,000) .+7($-60,000)
=$-10,000 -$2,000 +
$42,000
= $30,000
Because the EMV is positive for both Projects 1 and
2, Cliff’s firm would expect a positive outcome from
each and could bid on both projects. If it had to
choose between the two projects, perhaps because of
limited resources, Cliff’s firm should bid on Project 2
because it has a higher EMV.
Using EMV helps account for all possible outcomes
and their probabilities of occurrence, thereby
reducing the tendency to pursue overly aggressive or
conservative risk strategies.
SIMULATION
A more sophisticated technique for quantitative risk analysis is
simulation, which uses a representation or model of a system to
analyze its expected behavior or performance. Most simulations are
based on some form of Monte Carlo analysis. Monte Carlo analysis
simulates a model’s outcome many times to provide a statistical
distribution of the calculated results.
You can use several different types of distribution functions when
performing a Monte Carlo analysis.
The following example is a simplified approach. The basic
steps of a Monte Carlo analysis are as follows:
1. Collect the most likely, optimistic, and pessimistic estimates for the
variables in the model. For example, if you are trying to determine the
likelihood of meeting project schedule goals, the project network
diagram would be your model. You would collect the most likely,
optimistic, and pessimistic time estimates for each task. Notice that this
step is similar to collecting data for performing PERT estimates.
However, instead of applying the same PERT weighted average formula,
you perform the following steps in a Monte Carlo simulation.
2. Determine the probability distribution of each variable. What is the
likelihood of a variable falling 1 between the optimistic and most likely
estimates? For example, if an expert assigned to a particular task
provides a most likely estimate of 10 weeks, an optimistic estimate of 8
weeks, and a pessimistic estimate of 15 weeks, you then ask about the
probability of completing that task between 8 and 10 weeks. The expert
might respond that there is a 20 percent probability.
3. For each variable, such as the time estimate for a task, select a random value based
on the probability distribution for the occurrence of the variable. For example, using
the preceding scenario, you would randomly pick a value between 8 weeks and 10
weeks 20 percent of the time and a value between 10 weeks and 15 weeks 80
percent of the time.
4. Run a deterministic analysis or one pass through the model using the combination of
values selected for each of the variables. For example, one task described in the
preceding scenario might have a value of 12 on the first run. All of the other tasks
would also have one random value assigned to them on the first run, based on their
estimates and probability distributions.
5. Repeat Steps 3 and 4 many times to obtain the probability distribution of the
model’s results. The number of iterations depends on the number of variables and
the degree of confidence required in the results, but it typically lies between 100 and
1,000. Using the project schedule as an example, the final simulation results will
show you the probability of completing the entire project within a certain time period.
SEVERAL PC-BASED SOFTWARE PACKAGES THAT PERFORM MONTE CARLO
SIMULATIONS ARE AVAILABLE. MANY PRODUCTS DISPLAY THE MAJOR RISK
DRIVERS FOR A PROJECT BASED ON THE SIMULATION RESULTS. THIS ENABLES
YOU TO IDENTIFY THE CHIEF SOURCE OF UNCERTAINTY IN A PROJECT
SCHEDULE. FOR EXAMPLE, A WIDE RANGE FOR A CERTAIN TASK ESTIMATE
MIGHT CAUSE MOST OF THE UNCERTAINTY IN THE PROJECT SCHEDULE
SENSITIVITY ANALYSIS
It shows the effect of changing one or more variable on an
outcome.
"What if" analysis.
What change happen on the outcome when variables changes.
Focuses on which variable influences the outcome.
Example; Cliff’s team could develop sensitivity analysis models to
estimate their profits on jobs by varying the number of hours
required to do the jobs or by varying costs per hour.
BENEFITS
Determine break-even points based on different
assumptions
To make business decisions
Identifying Critical Factors
Resource Allocation
Assessing Risk Impact
Ranking of Variables
Ranking of Variables
PLANNING RISK RESPONSE
After identifying and quantifying risks, we must develop appropriate responses to them.
Developing a response to risks involves developing options and defining strategies for reducing
negative risks and enhancing positive risks
The five basic response strategies for negative risks
Risk avoidance : eliminating a specific threat, usually by eliminating its causes. Of course, not all risks
can be eliminated, but specific risk events can be.
Risk acceptance :means accepting the consequences if a risk occurs
Risk transference: means shifting the consequence of a risk and responsibility for its management to
a third party.
PLANNING RISK RESPONSE
Risk mitigation: means reducing the impact of a risk event by reducing the probability of its
occurrence.
Technical Risks Cost Risks Schedule Risks
Emphasize team Increase the frequency of Increase the frequency
support and avoid project monitoring of project monitoring
stand-alone project
structure
Increase project Use WBS and CPM Use WBS and CPM
manager authority
Improve problem Improve communication, Select the most
handling and understanding of project experienced project
communication goals, and team support manager
Increase the frequency Increase project manager
of project monitoring authority
Use WBS and CPM
PLANNING RISK RESPONSE
Risk escalation : notifying a higher level authority if the risk is outside the project managers authority
The five basic response strategies for positive risks
Risk exploitation :doing whatever you can to make sure the positive
risk happens
Risk sharing: allocating ownership of the risk to another party
Risk enhancement: changing the size of the opportunity by identifying
and maximizing key drivers of the positive risk.
Risk acceptance: applies to positive risks when the project team does
not take any actions toward a risk.
Risk escalation: notifying a higher level authority also applies to positive
risks
PLANNING RISK RESPONSE
Outputs of risk response planning
updates to the project management plan and other documents and
change requests
result changes in WBS and project schedule
provide updated information for the risk register by describing the risk
responses, risk owners and status information
Risk response strategies often include identification of residual and
secondary risks as well as contingency plans and reserves
• Residual risks: are risks that remain after all of the response
strategies have been implemented
• Secondary risks: are a direct result of implementing a risk response
OUTPUTS OF IMPLEMENTING
RISK RESPONSES
Implementing risk responses involves putting the
appropriate risk response plans into action
Change Requests: Possible changes to project scope,
timeline, or resources
Project Documents Updates: Updates to the issue log,
lessons-learned register, project team assignments, risk
register, and risk report.
prevent any negative events from happening
MONITORING RISKS
• ensuring the appropriate risk responses are performed.
• tracking identified risks, identifying and analyzing new risk, and evaluating the
effectiveness of risk management throughout the entire project
It may be necessary to alter a strategy that becomes ineffective, implement a
planned contingency activity, or eliminate a risk from the list of potential risks when
it no longer exists.
A redistribution of resources devoted to risk management may be necessary
because of relative change in risk exposure.
Tools and techniques for monitoring risks include data analysis, audits, and
meetings.
Outputs include work performance information, change requests, and updates to
the project management plan, project documents, and organizational process assets
workarounds—unplanned responses to risk events— when they do not have
contingency plans in place.
USING SOFTWARE TO
ASSIST IN PROJECT RISK
MANAGEMENT
If a risk is not identified, it cannot be managed, and intelligent,
experienced people are needed to do a good job of identifying
risks.
Software should be used as a tool to help make good decisions in
project risk management. Not when things go wrong.
project teams must be careful not to rely too heavily on software
when performing project risk management.
Example - Monte Carlo–based simulation
CONSIDERATION FOR
AGILE/ADOPTIVE
ENVIRONMENT
We use adaptive approaches for environments with high
variability environment .
While using agile approach , Risk is considered when selecting
the content of each iteration, and risks will also be identified,
analyzed, and managed during each iteration.
using adaptive approaches make use of frequent reviews of
incremental work products and cross-functional project teams
ensure that risk is understood and managed.
requirements are kept as a living document that is updated
regularly, and work may be reprioritized as the project progresses,
based on an improved understanding of current risk exposure.
THE END