How can PERT help us at activity level?

At activity level PERT can help us to determine a more realistic estimate. Let us take the following estimates for five project activities as an example:

Let us assume that we obtained the following three estimates for Activity A. Optimistically we can complete the activity in five (5) days. Pessimistically we estimate that it can take us up to Fifteen (15) days, while most likely it will take us only seven (7) days. PERT calculates a weighted average as the PERT estimate by using the formula : Pert Estimate = (Optimistic + (4 X Most Likely) + Pessimistic)/6. That means that we are weighting the most likely estimate by a factor of four (4) and then determining the average of the weighted most likely time, the best case scenario and the worst case scenario. In the case of Activity A that will be (5 +(4 X 7) +15)/6. The pert estimate is therefore calculated as 8 days. If we compare the PERT estimate of 8 days with the normal average of the three estimates of 9 days (5 + 7 + 15 /3 = 9) we can see that the PERT estimated is weighted towards the most likely estimate. We use a weighted average calculation because statistically we find that the largest portion of a randomly occurring population of data will be found close to the mean as can be seen from the following bell curve, also called a normal distribution curve: