1. Home
  2. Glossary
  3. Monte Carlo Simulation

Monte Carlo Simulation

A technique that runs thousands of randomized trials to produce a probability distribution of outcomes.

  • Project Management
  • Work Measurement

Monte Carlo simulation models uncertainty by sampling input variables from their probability distributions thousands of times and aggregating the results into a distribution of possible outcomes, not a single point estimate.

In project management it turns PERT three-point estimates into a probabilistic completion forecast ("80% chance of finishing by week 14"); in operations it stress-tests schedules, capacity, and inventory policies against variability.

Example. A project has three tasks in sequence with optimistic, likely and pessimistic durations of (8, 10, 15), (4, 5, 9) and (10, 12, 20) days. The single-point plan says 27 days. Sampling each task from a triangular distribution 10,000 times gives a mean around 29.5 days, an 80% chance of finishing within 32 days and a 5% chance of exceeding 35. The plan at 27 days has roughly a 25% chance of being met.

It needs honest input distributions: a three-point estimate from the person who will do the work, with the pessimistic value including things that have gone wrong before. Correlated risks (one late supplier delays three tasks) must be modelled as correlated or the spread is understated. The Monte Carlo guide sets one up in a spreadsheet; the project risk guide covers what to do with the result.

Go deeper

Related terms

Browse all glossary terms