What is PERT?
The Program Evaluation and Review Technique, commonly abbreviated as PERT, is a statistical tool used in project management. PERT was initially created in 1958 to simplify the planning and control of POLARIS submarine missile program by the Program Evaluation Branch of the Special Projects office of the U. S. Navy, and emphasis was on completing the program in the shortest possible time.
After PERT methodology was employed for this project, it actually ended two years ahead of its initial schedule. PERT is based on the probability of an event occurring at a specified time.
In addition, PERT had the ability to incorporate uncertain activity by making it feasible to schedule a project even when precise details and durations of all the activities are unknown. It is commonly used in combination with the CPM. Many managers, who use the PERT and CPM techniques, have claimed that these techniques drastically reduce the project completion time. While CPM is believed to be an easy technique to use and understand, it does not consider the variations in time which may have a huge impact on the completion time of a complex project. It has the potential to reduce both the time and cost required to complete a project.
PERT is actually an ‘event-oriented technique’ rather than an activity oriented technique, as CPM, and is mostly used in projects where time is the most important factor not the cost. It is applied to very large-scale, one-time, complex, non-routine infrastructure and research and development projects.
PERT uses three different time estimates, that is, optimistic, pessimistic and most likely, that help in establishing the prospects of a project completion within specific time and in calculating the risk before beginning a project. It has the ability to reduce both the time and cost required to complete a project.
Steps of PERT
PERT is a methodology for planning the diverse activities in either large or small projects. There are several variants of PERT which have been used in planning costs, scheduling manpower and machine time. Steps of the PERT technique include the following:
1.Identify activities and milestones: Define the project and identify all the significant activities or the tasks required for project completion. The milestones are the events spotting the beginning and end of one or more activities. It forms a work breakdown structure. The project might consist of many assignments. However, it should only consist of one start and one finish activity.
2. Determine activity sequence: This stage could be amalgamated with the first one as the series of activities is very apparent for some assignments. So in this step development of the relationships among the activities takes place and also decision about which activities should come first and which should follow others.
3. Construct the network diagram: Using the activity sequence information from previous step, a ‘network’ connecting all the activities is being developed. A network diagram is drawn to show the sequence of the serial and parallel activities. Each activity should be given a unique event number. To avoid giving the same number to two activities, dummy arrows are used. The activities of the original activity-on-arc model of PERT are characterized by arrowed lines and milestones, by circles. It can be done manually or there are software packages available to make this step simpler as they automatically convert the information from tabular activity into a network diagram.
4. Estimate activity times: In this step, time is being assigned and cost may be estimated for each activity which is called Activity Time Estimation which in turns helps in Network Time Calculation and Scheduling of a complete project. The most commonly used time unit for activity completion is ‘Weeks’. However, they can also use any other constant unit of time.
A distinguishing characteristic of PERT is its capability to deal with ambiguity in the time taken for completion of activities. Generally, this model includes three time estimates. In PERT, these three estimate times are calculated for each activity. In this way, a range of time is being obtained for each particular activity, with the most probabilistic value.
(a) Optimistic Time Estimate (TOPT): Optimistic time is represented by ‘to.’ Generally, the lowest time possible to complete the activity, being an optimist, the estimator assumes that all the necessary resources will be available and all previous activities will be completed in estimated time as planned. It is commonly practiced that while specifying optimistic times three standard deviations from the mean has been used so that there is more or less only one per cent chance of the activity to be completed within the optimistic time.
(b) Most Likely Time Estimate (TLIKELY): It is represented by tL, and is the highest probable completion time for a particular task. This time is in between optimistic and pessimistic times. It is important to
note here that this time is different from the expected time. Here the estimator expects that some sort of uncertainties may come across but most of the things will go right.
(c) Pessimistic Time Estimate (TPESS): This is represented by tP. It is the highest possible time that an activity might require. Three standard deviations from the mean are usually used for the pessimistic time. In
this case, it is being assumed that many things related to a particular activity will go wrong. While deriving this estimation a lot of rework and resource unavailability are assumed.
5. Determine the critical path: For every activity in the critical path, expected time and variance are calculated. By adding the expected times for the activities in each sequence computation of the longest time path in the project through the network is done. This is the overall expected completion time for the project and is called as the critical path. Activities occurring outside the critical path may speed up or slow down within certain limits; but cannot alter the total project time. The amount of time that can be delayed by a non-critical path activity without delaying the project is referred to as slack time.
The variance in the project completion time can be calculated by adding all the variances in the completion times of the activities occurring in the critical path. This is done only for the activities in the critical path as only the critical path activities can accelerate or delay the project duration.
With the help of this variance, one can calculate the probable date for the completion of the project by assuming a normal probability distribution for the critical path. This is usually referred as the probability aspect of PERT. Since the completion date of the project has been established by the activities of the
critical path, the project can pick up pace by amalgamating the necessary resources for reducing the time for the activities falling on the critical path. This is often referred to as ‘project crashing’.
6. Update as project progresses: This stage is being used to make adjustments in the PERT chart with the progress in project. Network plan is used to schedule, monitor and control the project. As the project unfolds, the estimated times can be replaced by actual times. In condition of delays, additional resources may be added to stay on schedule and modifications can be done in the PERT chart to reflect the new situation.
Advantages of PERT
- It is useful at various stages of project management.
- It is mathematically very simple.
- It provides critical path and slack time.
- The start date and end date of a particular activity.
- The estimated project completion time.
- Determine if the project is going as per schedule or not.
- Type and nature of risks involved
- The probability of project completion before a particular date at minimum cost.
- The financial aspect of the project is within estimated budget or not.
- Assess the activities having slack time and providing necessary resources to the critical path activities.
- Determine the critical path activities having direct impact on the project completion time.
- Project documentation becomes very easy if done on computer.
- It is also useful in monitoring costs and resources.
Limitations of PERT
- The estimation of activity time is subjective to some extent and depends on judicious decision making. In the cases when the person, who is estimating, has less knowledge and experience in executing an activity, the estimation may only be a guess. Whereas, in some other cases, if the person or group performing the activity estimates the time it may be partial in the estimate and again the estimation may be a prejudice.
- PERT takes on a beta distribution for such time estimates even though all the activity times are well-estimated. However, the actual distribution could vary.
- PERT is based on the assumption of beta distribution, it constantly underestimates the expected project completion time. It presupposes that the probability distribution of the time taken for project completion is similar to that of the critical path. However, another critical path could come up if the related activities are delayed due to some reason.
- There are over emphasis on critical paths in PERT.