PMP Primer Masterclass Online Training

Applying Monte Carlo Analysis – QSRA

Using Monte Carlo analysis on the first pass relies on you having created a schedule for your project, and that this includes the identification of your risks, their analysis and responses.

So that once the schedule has been created, reviewed and verified, a quantity of schedule risk analysis (QSRA) will apply statistical techniques to test the level of confidence in meeting your project completion date.

This analysis looks at the whole schedule and not just at the critical path.  The conclusion of this analysis may result in a completely different set of linked activities that determines a “most likely” completion date.

Monte Carlo Analysis

A QSRA relies on Monte Carlo analysis.  A Monte Carlo analysis is a set of calculations that rely on repeated random sampling to obtain a range of results – the probability distribution.

The sampling is based on information provided by the user.  There are two elements to QSRA:

Duration uncertainty.  This provides a minimum, most likely and maximum spread of activity durations

Risk impact.  This provides the minimum, most likely, and maximum impacts

In some organisations, QSRA is referred to as Timescale Risk Analysis, but I will continue to use the QSRA term.

The Five Key purposes of QSRA

QSRA  Purpose 1

Improve the quality of the project schedule

The best mitigation strategy for risk on a project is to have a robust and logically sound schedule.  It is important to ensure that the fall scope and working methods our accurately modeled in the schedule.

This modeling describes how to drive understanding and acceptance of the project schedule, and hence commitment to project delivery.

Running a deterministic schedule through a QSRA to all can derive benefit by exposing weaknesses in the original schedule which are not apparent in the static schedule.

The expected variations, or lack of them, provoked discussions, resulting in both an improved understanding of the schedule by the project team and a better finish schedule.

A QSRA will also highlight which activities and paths are perceived by the QSRA tool set as being key drivers.  These can often differ from the path associated with the critical path analysis, where more risk is associated with originally perceived sub critical paths.

Monte Carlo analysis Tools

There are many tools available to help you perform QSRA and Monte Carlo’s analysis, here are just two of them

  • Primavera Monte Carlo risk analysis tool.
  • @RISK using Excel spreadsheets

QSRA Purpose 2

Model duration uncertainty and risk

The two stages of running a QSRA are described in more detail below.  A key advantage of running a QSRA is the consideration of these two facets of the schedule and risk profile.

In the case of duration uncertainty, this allows the project team to express their confidence level in the schedule, which may in itself expose certain issues.

Reviewing the risks will ensure that these are considered as part of the schedule, and the implications will be more widely understood as a result.

QSRA Purpose 3

Assess the probabilities of completing the project on time

The fundamental purpose of QSRA is to determine the probability of finishing on or before a given point in time.  It will determine the confidence level of completing the project to the required date.

This when facilitate further strategic decision-making.

QSRA Purpose 4

Focus attention on areas requiring mitigation

As a result of undertaking a QSRA, it may be appropriate to consider medications that need to be put in place.  This will particularly be the case if the exercise has identified alternative critical path is to the one created during the original scheduling of the project.

QSRA Purpose 5

Setting baselines

The QSRA approach may be used to establish an appropriate target to be used as a project baseline.  Different organisations may have different risk appetites and establish targets based on the set probability of achievement.

These are often known as for example, “P70” levels, where P means probability and the number refers to the percentage confidence level – 70% in this case.

Go To Part 2 HERE!