PERT Estimation: Three-Point Estimating on the PMP Exam

Three-Point Estimating: Why the PMP Exam Loves PERT

Single-point estimates are convenient but unreliable. They assume perfect knowledge about how long an activity will take, which almost never exists. Three-point estimating — and its PERT variant — addresses this by incorporating uncertainty directly into the estimate. The PMP exam includes calculation questions on this topic, so you need to know both the formulas and the reasoning behind them.

The Three Estimates

Three-point estimating uses three scenarios for each activity:

Optimistic (O): The best-case duration, assuming everything goes perfectly. This is not a fantasy number — it represents the fastest realistic completion under favorable conditions.
Most Likely (M): The duration expected under normal conditions, considering typical problems and resource availability.
Pessimistic (P): The worst-case duration, assuming significant problems but not outright catastrophes. It does not include force majeure events.

Two Formulas You Must Know

Triangular Distribution

The simple average of the three estimates:

E = (O + M + P) / 3

This treats all three estimates as equally likely, which is rarely accurate but is sometimes used for its simplicity.

PERT (Beta Distribution)

The weighted average that gives extra weight to the most likely estimate:

E = (O + 4M + P) / 6

This formula reflects the reality that the most likely estimate is usually closest to the actual outcome. The weighting of 4 on M makes the beta distribution more peaked around the most likely value.

Standard Deviation and Variance

The PMP exam also tests your ability to calculate the variability of an estimate:

Standard Deviation (σ) = (P - O) / 6

Variance = σ² = [(P - O) / 6]²

Standard deviation tells you how spread out the estimate is. A large standard deviation means high uncertainty; a small one means high confidence.

Worked Example

An activity has the following estimates: Optimistic = 4 days, Most Likely = 7 days, Pessimistic = 16 days.

PERT estimate: (4 + 4×7 + 16) / 6 = (4 + 28 + 16) / 6 = 48 / 6 = 8 days
Standard deviation: (16 - 4) / 6 = 12 / 6 = 2 days
Variance: 2² = 4 days²

This means the expected duration is 8 days, with a standard deviation of 2 days. There is approximately a 68% probability (one standard deviation) that the activity will take between 6 and 10 days.

Applying PERT to the Entire Project

To estimate the total project duration using PERT, calculate the expected duration for each activity on the critical path, then sum them. The project standard deviation is not simply the sum of individual standard deviations — you must sum the variances first, then take the square root:

Project σ = √(σ₁² + σ₂² + σ₃² + ...)

This is a critical detail that PMP questions test. You cannot add standard deviations directly; you must add variances and then convert back.

When to Use Three-Point Estimating

The PMP exam expects you to know when three-point estimating is appropriate:

When historical data is limited or unreliable.
When activities involve significant uncertainty or risk.
When stakeholders need to understand the range of possible outcomes, not just a single number.
When the project management plan requires probabilistic schedule analysis.

PMP Exam Tips for PERT Questions

Always check whether the question asks for the triangular or beta (PERT) distribution. The formulas are different.
Watch the units — days vs. weeks vs. hours. Ensure consistency.
If asked about confidence ranges, remember: ±1σ = ~68%, ±2σ = ~95%, ±3σ = ~99.7%.
Variance is additive across independent activities. Standard deviation is not.

Practice PERT calculations with our timed PMP practice exams that mirror the real exam format.