liveBook · Manning

Program evaluation and review technique

As explained in Statistics Every Programmer Needs

The program evaluation and review technique is a project-scheduling method that estimates activity durations using optimistic, pessimistic, and most likely durations.

PERT estimates project-activity durations when comparable historical data is unavailable. It combines optimistic, pessimistic, and most likely estimates into one expected activity time, accommodating uncertainty in project scheduling. section 13.2 section 13.6

Apply PERT by following this estimation workflow:

  • Define the optimistic time (a) as the shortest plausible completion time. section 13.2
  • Define the pessimistic time (b) as the longest plausible completion time. section 13.2
  • Define the most likely time (m) as the expected duration under normal conditions. section 13.2
  • Calculate the weighted expected activity time and assign that value to the task; retain the three estimates as inputs rather than schedule durations. section 13.2
  • Obtain estimates from the assigned resources or subject-matter experts, because input quality strongly affects output reliability. section 13.2

Automated reporting tool

For one activity, estimate 2 weeks optimistically, 3 weeks most likely, and 4 weeks pessimistically. PERT produces an expected duration of 3 weeks. Across activities A–L, expected times sum to 39 weeks, but parallel work means this sum is not the project duration. section 13.2

Figure 13.1 An illustration of the beta probability distribution assumed by PERT that displays the optimistic, pessimistic, and most likely activity times. The area under the curve represents the probability of different completion times, with the peak indicating the most likely duration.
figure

Visualizes the beta probability distribution assumed by PERT, including optimistic, pessimistic, and most likely times and the probability represented by the area under the curve. section 13.2

How Statistics Every Programmer Needs covers it

How deep it goes
Explains three inputs, weighted expected time, expert-input quality, and the beta-distribution assumption without extending into schedule-network calculations.
Examples
Uses an automated reporting tool and a twelve-activity table with expected times totaling 39 weeks.
What it assumes
Assumes project-task definitions and informed resource estimates while explaining the three duration categories from zero.
What you can do after
Apply PERT to produce defensible activity-duration estimates and support more resilient project schedules.
How it connects
Connects activity estimation with analogous estimating, work breakdown structures, critical path method, completion probability, and project crashing.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Statistics Every Programmer Needs. The section links point to the parts of the book it draws on.