A PERT calculator turns three estimates for a task, the best case, the most likely case and the worst case, into one expected value with a standard deviation. Enter them for a single task, or list many tasks and the calculator adds the expected values and combines the standard deviations to give the expected total with likely ranges. It also shows the simpler triangular average for comparison. It works for time, cost or any other quantity. It runs in your browser.
How to use the PERT calculator
- Choose one task or several tasks.
- Enter the optimistic, most likely and pessimistic estimates, or one task per line.
- Press Calculate.
The PERT formulas
The three-point estimation method gives the expected value as E = (a + 4m + b) ÷ 6 and the standard deviation as SD = (b − a) ÷ 6, where a is the optimistic estimate, m the most likely and b the pessimistic. It describes E as a weighted average that takes into account the optimistic and pessimistic estimates.[1] A university operations research text gives the same formulas for PERT as μ = (o + 4m + p) ÷ 6 and σ = (p − o) ÷ 6, and it puts most of the weight on the most likely estimate.[2] It also states the variance as the square of that standard deviation.[2] A task with a = 2, m = 4 and b = 12 has an expected value of (2 + 16 + 12) ÷ 6 = 5 and a standard deviation of 10 ÷ 6 = 1.67.
A worked example
Take two example tasks of our own. Task A has the estimates 2, 4 and 6, which give an expected time of 4.00. Task B has 3, 5 and 9, which give 5.33. Enter them in the calculator to check: for task B, (3 + 20 + 9) ÷ 6 = 5.33, and the standard deviation is 6 ÷ 6 = 1. The university text applies the same formulas to each activity of its project.[2]
The triangular alternative
The three-point method also gives a simpler average, E = (a + m + b) ÷ 3, for a triangular distribution.[1] It counts the most likely value only once. For the task above, it is 18 ÷ 3 = 6, compared with 5 from the PERT formula. The calculator shows both, so you can see how much the choice matters. When the most likely value is close to the middle, the two agree. The PERT value is nearer to the most likely estimate when the range is wide.
Likely ranges
The three-point method relates the range around the expected value to the confidence. About 68 percent of outcomes fall within E ± SD, 90 percent within E ± 1.645 × SD, 95 percent within E ± 2 × SD, and 99.7 percent within E ± 3 × SD.[1] The calculator gives these ranges for a task or for a project. For the task with E = 5 and SD = 1.67, a range of 3.33 to 6.67 covers about 68 percent. The source notes that the figures are only as accurate as the three initial estimates.[1]
Adding up a project
For several tasks, the three-point source says that the expected value of the project is the sum of the expected values of the tasks, and its standard deviation is the square root of the sum of the squares of the task standard deviations.[1] The university text gives the same rule for the activities on the critical path: the mean of the project duration is the sum of their means, and its variance is the sum of their variances.[2] This assumes that the tasks do not depend on each other. The calculator lists each task with its expected value and standard deviation, and then the total with the ranges. For tasks A and B above, the total is 4 + 5.33 = 9.33, and the standard deviation is the square root of 0.44 + 1, which is 1.20.
Limits of the method
The source warns that the estimate is no more accurate than the three starting numbers, and that there are dangers in assuming a shape for the distribution. It says that the normal ranges need enough tasks, in the order of 20 to 30, for the total to be close to a normal curve, and that the task estimates should be unbiased and not correlated.[1] The calculator therefore shows a note when you enter fewer than 20 tasks. The university text bases the normal shape of the project duration on a central limit theorem argument and calls it an approximation.[2] Use the results as a guide, and keep your own judgement.
Giving good estimates
Estimate the best case as a result that you would be surprised to beat, and the worst case as one that you would be surprised to exceed. The most likely case is the one that you would bet on. Ask people who will do the work, not only managers, and write down the reasons for the worst case. If you have past data, compare your estimates with what happened, and adjust.
Common mistakes
- Putting the numbers in the wrong order. The optimistic estimate is the smallest.
- Adding standard deviations. Add the variances, then take the root.
- Using it for a few tasks and trusting the 95 percent range. The ranges need many tasks.
- Mixing units. Use the same unit for all estimates.
What next
Use this often? Add PERT to your Home screen.
Frequently asked questions
What is a PERT calculator?
A tool that turns optimistic, most likely and pessimistic estimates into an expected value and a standard deviation, for a task or a whole project.
What is the PERT formula?
E = (a + 4m + b) ÷ 6, with a standard deviation of (b − a) ÷ 6.
How do I add up several tasks?
Add the expected values, and take the square root of the sum of the squared standard deviations.
What range does one standard deviation cover?
About 68 percent. Two standard deviations cover about 95 percent.
What is the difference from the triangular average?
The triangular average is (a + m + b) ÷ 3 and gives the most likely value less weight than PERT does.
Sources
Was this tool helpful?
What went wrong?
We only receive the reason you pick, never what you typed into the tool.
Thanks for your feedback.