The Programme Evaluation and Review Technique (PERT) is employed for planning and controlling the projects involving uncertainties.
PERT is an event-oriented technique. Its basis is a network of events in which the activities are derived by connecting the events. It lays stress on measuring the uncertainty in activity times by using the three-times duration estimation method. For computation of critical path, the PERT three-times probabilistic network is converted into a single-time deterministic CPM Model. PERT studies the implications of uncertainties on project time scheduling and slack of events by employing statistical tools.

Modelling PERT Network

Two special features of PERT distinguish it from the other network analysis techniques. These are emphasis upon events rather than activities and the use of three-times estimate for activity duration.
The PERT network-making approach is to identify the milestones necessary for successful completion of the project. These milestones are then depicated in the from of a key-events network showing their sequence and interdependence. After this, the events visualized between the milestones are added and their interrelationship established. These event nodes represent points in time which are generaly terminal in nature. The activities are derived by interconnecting the events. The event diagram thus obtained is converted into event-oriented PERT network using the network drawing rules.

The steps involved in this technique are:

a)     Identification of key events or milestones which must occur during the project execution.

b)    Determination of the sequence and interdependence of the key events.

c)     Incorporation of events (or activities) between the milestones, generally conforming to their sequence of occurrence.

d)    Connection of events, maintaining their interdependence.

This gives the project flow diagram from which the network can be drawn directly using the network drawing rules.
It may be noted that while determining the three-time estimate, the optimistic and pessimistic times should be assessed first. These two extremes bracket the most likely time, thus helping the planners to concentrate within this range. There is also a tendency of planners to either use the same figure for all the three times (say 9, 9 and 9) or arrive at the most likely time first and then arrive at the optimistic and pessimistic time by subtracting and adding an equal amount (say 6, 9 and 12). These trends should be avoided because such estimates are likely to be conservative.