The objective of this post is to explain how program evaluation and review technique (PERT) can be used to schedule a project with uncertainty.
The main difference between CPM and PERT is that PERT uses probabilistic durations instead of deterministic ones. In PERT, each activity has three estimated durations which are optimistic, most likely, and pessimistic. The purpose of using these three different durations is to model the uncertainty inherited in the activity durations.
The steps of scheduling a project using PERT will demonstrated through the following example. The information of the project is presented in Table 1. It should be noted that in this example, the finish to start relationship is used. The same procedure would be followed if different relationships do exist.
Step 1: Calculate the mean, standard deviation, and variance of each activity using the following equations. The results of this step are presented in Table 2.
Step 2: Construct the precedence diagram and preform CPM analysis using the mean duration as shown in Figure 1. Then find the critical path which in case is B-C-G-H-K-M-O-P-Q-R. Next, calculate the standard deviation of the project which is the square root of the sum of the variances of the critical path activities. In this case it is 3.75.
Figure 1 Precedence Diagram
Step 3: Probability analysis using the calculated information. Two main questions can be answered using probability analysis. The first question is “What is the probability that the project will be finished on or before 98 (you can choose the target duration that you are interested in) days?” To answer this question Z, which is the number of standard deviations the target date lies from the expected date, should be calculated using the NORM.DIST function in Excel 2010 (NORMDIST function in Excel 2007) or the following equation.
The next step would be finding the area under the normal distribution curve using normal distribution tables that corresponds to a Z value of 0.8. By doing that, we can find that there is 78.81 % chance that project can finish in 98 days or less. It should be noted that if the Z has a negative value, then the probability would be 1 – table value.
The second question is “What is the due date under which the project has 99 % (you can choose the confidence level that you are interested in) chance of completion?” In this question, we are given the area under the curve which 0.99. We can either find the answer by using NORM.INV function in Excel 2010 (NORMINV function in Excel 2007) or the following steps. What we need to find is the Z value from the normal distribution table. In this case, Z is 2.33.
I have also developed an Excel sheet that you can download from here.
If you have any questions, please do not hesitate to ask.
3 comments:
The trouble with PERT is that it considers only the one path which is critical based upon the mean durations. Generally there will be other paths which are nearly critical based upon mean durations and which could turn out to be critical if their durations happen to be towards the pessimistic end of the estimated range. Since the project is not complete until _all_ paths are complete, the expected project completion date (and all the percentiles of the project completion date distribution) estimated by PERT can be very biased towards the optimistic end of the scale.
A much better technique is schedule risk analysis, which uses Monte Carlo simulation to take into account all the uncertainty in all the paths. Cost effective tools for this are readily available. For example, if you use Microsoft Project there is Risk+ and my own Full Monte (see www.barbecana.com)
That is very interesting; you are a very skilled blogger. I have shared your website in my social networks! A very nice guide. I will definitely follow these tips. Thank you for sharing such detailed article.
Evaluation and Management Training in Hyderabad
This blog is very informative thanks for sharing and keep on posting .#CONSTRUCTION RELATED COURSE IN DUBAI
Post a Comment