Program evaluation and review technique (PERT)

Revision as of 18:23, 18 September 2015

The Program Evaluation and Review Technique (PERT) is a method used in project management for planning, scheduling and coordinating the tasks of a project. It is based on a network diagram analysis of the different tasks that are involved. PERT is specifically suited for large, complex and nonrepititive projects, since the network diagram varies in each project. The aim of this article is to give an overview of this method with examples of applications and a comparison to other existing network based project management tools. It will also explore the usage of this method in todays projects.

General overview

Program Evaluation and Review Technique as a network analysis tool gives the possibility to analyze large projects task by task and to estimate the amount of time that will be spent. It was first introduced in the 1950s by the US Navy, during the works on the Polaris missiles project. It was a large and complex project involving multiple contractors and an effective planning tool was a necessity to finish the project successfully. The project has been finished long before the deadline and most of the people attributed this success to PERT, although it has not been really been proved.

Key concepts

- Definition of the key concepts that will be used

   - Activities
   - Events
   - Network diagram and analysis
   - Critical Path

PERT has several concepts that need to be defined, to understand the whole method. Those are :

• Activities
• Events
• Network diagrams
• Critical path

A network diagram is a visual representation of a project's schedule. It presents all the information about the project on a time basis like the tasks, their starting/ending times, total duration. Using this kind of representation makes the whole management process much easier and less error-prone (to be checked).

The activities are the tasks that need to be completed. Each task has a predefined duration, a priority, a cost and an amount of workers associated with it. On a network level, they are represented as nodes connected by arrows. They can have predecessors - other tasks that need to be completed before them - and successors - tasks that can start only once this one has finished.

The events are points in time marking a phase of an activity like the start or the completion. Events can relate to one or more activities.

The critical path is the path through the network that includes the most time-consuming sequence of events and activities. Hence it represents also the total time that will be needed to complete the project. Any delay that occurs in one of the activities on this path will affect the completion date.

The process

- Detailed step by step presentation of the method

   - building a network diagram
   - calculation of the times
   - possible modifications

The process of making a PERT analysis can be divided into 4 steps:

• Preparation of a list of all the tasks that will be involved in a given project
• Creation of a network diagram
• Estimation of the time required to finish each activity
• Calculation of the critical path
• Network analysis

Preparation phase

PERT starts with a standard step of defining the activities that will be involved in the project. This information is presented in a box representation. This box will later be filled with other time informations about the project.

Network diagram

Once the basic information about the activities has been gathered, the activities need to be put in order. The priority of the task and the dependencies between them are taken into consideration. In the network, activities are represented as arrows, whereas the events are the nodes.For instance, the preparation of the base of a building would be decoupled into "starting of preparation" and "end of preparations".

Estimation of the time

In this step we estimate the amount of time that can be taken by each activity. This time can be determined using a probabilistic or a deterministic approach. Three times need to be defined :

• the optimistic time (short time)
• the pessimistic time (shortest time)
• the normal time (most likely time)

The optimistic time (O) is the shortest time an activity will take if everything goes right. In general, it is highly improbable that the project will finish in this time, due to all the other factors.

The pessimistic time (P) is the longest time that an activity will take. During the calculations, several scenarios need to be considered with their respective probability in order to have a quite precise value.

The normal time (M) is the most likely time of an activity. Usually, it is determined using the time other similar activities took in other projects. For instance, the time needed to build a cockpit for one aircraft might be based on the average time it took to build cockpits for comparable aircraft in the past. Using these three times one can calculate the expected time of an activity with the formula below : Expected time = O + 4M + P / 6

It is basically an average of all of the times mentioned, but the normal time is given a higher weight, since it is the most probable. The expected time will be later used in the PERT diagram.

Critical path estimation

Once all the time information has been gathered about the different activities, it is possible to estimate the total duration of the project and the starting and ending dates. Determining the critical path comes down to identifying the path that takes the most time. A project cannot be completed as long as his longest component isn't finished. Here are the steps to find this path :

• Step 1 : Starting from the first activity, write the starting and ending times, by adding the duration to the ending time of the preceding activity.
• Step 2 : Make the same thing but from the end and substract the times instead of adding.
• Step 3 : Having the starting times and ending times calculated in two different ways, we can identify the tasks which have a null difference between the respective times. Those tasks constitute the critical path.

Time cost trade-offs and resource allocation

Having a first estimate of the time necessary to complete the project, it is possible to make a few changes to the plan and compress it a bit. In this step, the manager can decide which activities could be finished faster by reallocating resources properly or increasing them.

Project control

Having a clear plan of action, the project can finally be started. Every activity is monitored and checked on the schedule. Thanks to a clear representation of the project, in any case, the schedule can be reanalyzed and resources can be managed properly.

Applications and limitations

- Where is the method most useful and why - Why it cannot be used in other areas or is less efficient - Possible future ?

Advantages

The advantages of PERT are not only related to the calculations but to the whole process of planning. Here is a brief list :

• Compels to plan and predict project before start. The planning of the project is made in details, resulting in less surprises later.
• Better management of resources. Since we have a detailed plan, we can precisely manage all of the resources, sometimes leading to a reduction of costs of the whole project.
• Focus on the critical activities.
• Good control over the project from the beginning; weak points of the project located right from the start. Gives the possibility to adapt quickly to the situation
• Provides updated information about the project
• Clear view of the dependencies between tasks
• Accurate time estimation thanks to the 3-way computation of the total timing.

Limitations

Although PERT has a lot of advantages, it does also come with some disadvantages

• Emphasis only in time and not cost. This method does not present a model to estimate the costs of the project, therefore it is totally dependent on the managers skills and assumptions.
• Lots of analysis during control. Although a good plan can reduce the time of the whole project, making such a detailed plan is time-consuming, which leads to no time gain at all.
• Errors in time estimation might lead to miscalculations. Even if the 3-way estimate was considered an advantages, it is still based on assumptions and data given by the manager. Therefore
• No representation of repetitive tasks. Because the project is presented in a network form, a repetitive task would be represented as an inifinite loop, which could prevent us to determine the critical path.

Applications

PERT is a tool that can be useful in big and complex projects like in the construction industry. As mentioned in the introduction, it was created during the Polaris missile project and had a huge popularity afterwards. But soon, project managers and companies have found that it is too ineffective and error-prone to use it, causing lots of failures. The main reason behind it was the lack of IT tools, which had not the same abilities as now. Another major point was the update step in this method, which requires to reorganize the whole plan and in the case of large projects, this can be a time consuming and confusing step. Nowadays the software and hardware part is not a problem anymore. Many management software include PERT charts as part of their functionalities like Microsoft Project or Oracle Primavera.

Other modified versions

• PERT II
• PERTBN (Bayesian network)
• RPERT (repetetive PERT)
• Improved program evaluate review technique based on particle swarm optimization algorithm

Bibliography