Enter optimistic, most likely and pessimistic durations for one task or a whole list. The calculator applies the PERT formula to each task, adds them up and gives the expected duration, standard deviation, 68%, 90%, 95% and 99.7% ranges and the probability of hitting your deadline.
Last updated:
Three-point estimates
One task per line. Separate values with commas, tabs or semicolons. A single line with three numbers works too.
Shows the probability of finishing within it.
Enter your values to see the result.
Advertisement
Advertisement
How to use the PERT Calculator
List your tasks, one per line, as name, optimistic, most likely, pessimistic (for example Design, 4, 6, 10).
Pick the time unit that matches your estimates.
Optionally enter a target deadline to see the chance of finishing within it.
Read the expected duration and standard deviation, then the confidence ranges and percentiles.
Download the task table as CSV to put it in your project plan.
PERT formulas
Expected time: E = (O + 4M + P) ÷ 6
Standard deviation: σ = (P − O) ÷ 6, variance σ²
Project total: Etotal = ΣE and σtotal = √(Σσ²) for tasks done in sequence
Probability of finishing by T: z = (T − Etotal) ÷ σtotal, then look up z in the standard normal table
Example: a development task estimated at 10, 15 and 28 days has E = (10 + 60 + 28) ÷ 6 = 16.33 days and σ = 18 ÷ 6 = 3 days. Note how the long pessimistic tail moves the expected time above the most likely 15 days.
Confidence levels and z-values
Confidence
Two-sided range
One-sided "finish within"
50%
E ± 0.674σ
E
68.3%
E ± 1σ
E + 0.475σ
80%
E ± 1.282σ
E + 0.842σ
90%
E ± 1.645σ
E + 1.282σ
95%
E ± 1.960σ
E + 1.645σ
99.7%
E ± 3σ
E + 2.748σ
Tips for better three-point estimates
Ask the people doing the work for all three numbers, and ask for the pessimistic case explicitly. Most estimates are too optimistic.
Keep tasks small, roughly 1 to 10 days each. Large tasks hide uncertainty.
Commit to a percentile, not the mean. Promising the expected value means you will be late about half the time; P80 or P90 is a safer external date.
Remember the assumptions. Summing variances assumes tasks are independent; if one delay tends to cause others, real uncertainty is larger.
Frequently asked questions
What is the PERT formula?
The PERT (Program Evaluation and Review Technique) expected time is E = (O + 4M + P) ÷ 6, where O is the optimistic estimate, M the most likely and P the pessimistic. It is a weighted average that gives the most likely value four times the weight of the extremes, based on a beta distribution.
How is the PERT standard deviation calculated?
For one task, σ = (P − O) ÷ 6, so the spread between best and worst case covers about six standard deviations. For a sequence of tasks, add the variances (σ²) of each task and take the square root of the total; you do not add the standard deviations directly.
How do I get a 95% confidence estimate?
Take the expected duration plus or minus 1.96 standard deviations for a two-sided 95% range. If you only care about not running late, the 95th percentile is expected + 1.645σ, and 90% confidence is expected + 1.28σ. These use the normal approximation, which is reasonable for a chain of several tasks.
What is the difference between PERT and triangular estimates?
A triangular estimate is the simple average (O + M + P) ÷ 3. PERT puts more weight on the most likely value, so it is less pulled by an extreme pessimistic estimate. The calculator shows both so you can compare.
Should I add up all tasks or only the critical path?
Add the tasks that happen one after another on the critical path, the longest chain of dependent work. Tasks that run in parallel to it do not add to the total duration, so leave them out of the list or estimate them separately.
Why do the estimates need to be in order?
The optimistic value must be the smallest and the pessimistic the largest, with the most likely in between. If they are out of order the formula gives a misleading mean and a negative or wrong spread, so the calculator asks you to fix those lines.
Are my tasks saved anywhere?
Only in your own browser so they are there when you come back. Nothing is uploaded and all calculations run on your device.
Are all the tools on FreeToolCabin tested for quality and accuracy?
Yes! Every tool on our site undergoes automated Q/A testing in real browser environments to ensure high precision, correct output, and zero errors. If you happen to spot a bug or have a suggestion, please contact us at [email protected] and our team will review it right away.