Programme Evaluation Review Technique, or PERT Analysis is a technique which uses three estimates of activity duration rather than just one. This can be useful when there is a range of duration’s for any project activity. A calculation is used to determine the expected duration which is then applied to the project schedule.
Planning v Scheduling
It should be remembered that Planning and Scheduling are different things. Whilst the output of a schedule may look like a Gantt chart with resource names, a project plan (‘Planning’) consists of far more, including the project purpose, business case, contract strategy, risk management, communication management, stakeholder management etc.
Accurate estimates of activity duration are essential in order to create an accurate schedule. An accurate schedule can enhance the image of the project manager.
Triangular or Beta Distribution
To calculate the expected duration when using the PERT Analysis three point estimates, an equation is used as per this example:
The above equations ‘weight’ the result towards the Most Likely. It is called a Beta Distribution, using 4 times the most likely for the weighting. On any project activity that has been done several times before, it is a good method. This is because the ‘most-likely’ means just that. Most Likely.
On project activities that have not been done before, a simple average of the three estimates (or Triangular distribution) is used. This is simply because there is no experience of the duration of the task. Therefore, no ‘most likely’ information is available.
It can often be difficult to get one accurate estimate. If the Project Manager is not careful the estimate may be based on a simple guess by the task owner. Certainly, three estimates can give an idea of the confidence in the estimate based on the range between the optimistic and pessimistic estimates. If these are close, then there must be some confidence. Perhaps they are far apart, then this shows a lack of confidence.
If getting one accurate estimate is difficult, why should getting three estimates deliver a better result?
PERT Analysis is used to produce an accurate schedule based on accurate durations. The three point technique might be applied to just 10-20% of tasks in the project plan, because often, many tasks on projects have been delivered before.
Perhaps the biggest issue I have with PERT Analysis is the estimate of the pessimistic duration. How long is a piece of string? As long as you need it to be. What could go wrong with the task so that it takes longer? Everything. The duration could be nearly infinite if not carefully managed. This would not lead to an accurately calculated duration regardless of the methods used.
There is also one final risk to the method: The fact that the answer comes out of a computer does not make it any more accurate.