PERT Estimate Calculator

PERT (Program Evaluation and Review Technique) three-point estimation replaces a single guess at a task's duration with three: an Optimistic (O) best case, a Most Likely (M) realistic case, and a Pessimistic (P) worst case. Originally developed in 1958 by the United States Navy Special Projects Office with Lockheed Aircraft and Booz Allen Hamilton to schedule the Polaris missile program, and now taught as PMI's standard three-point estimating technique, the classic PERT formula weights the most likely estimate four times as heavily as either extreme -- Expected Duration = (O + 4M + P) / 6 -- on the reasoning that most tasks cluster near their realistic estimate and the two extremes are less probable outcomes worth acknowledging but not centering the plan on. This calculator also derives a standard deviation, (P - O) / 6, which is a common approximation for the spread of a beta distribution fit to the three points, and uses it to build a confidence range around the expected duration and a risk factor (the coefficient of variation) that expresses how much uncertainty exists relative to the estimate itself.

A task with a wide gap between optimistic and pessimistic carries more schedule risk even if its expected duration matches a tighter, more certain task exactly. Like all three-point estimation, the technique is only as good as the three inputs -- it does not independently validate that the optimistic and pessimistic bounds are realistic, and it assumes the underlying uncertainty is reasonably well described by clustering around the most likely value rather than, say, a long tail of rare but severe delays. The model assumes the three inputs are ordered Optimistic ≤ Most Likely ≤ Pessimistic; entering them out of order (for example a pessimistic estimate lower than the optimistic one) produces a negative standard deviation and an inverted range that has no real-world meaning.