Glossary
Monte Carlo Simulation
A computer-based technique that runs thousands of possible project scenarios to produce a probability distribution of cost or schedule outcomes.
Monte Carlo simulation is named after the famous casino — the idea being that it uses random sampling to model uncertainty, just as a roulette wheel produces a distribution of outcomes over many spins. In project controls, the simulation runs the project model thousands of times (typically 5,000 to 10,000 iterations), each time drawing a different set of values from the input uncertainty ranges. The result is not a single date or cost figure but a full probability distribution: you can see the likelihood of completing by any given date or within any given budget.
Monte Carlo is used in both Schedule Risk Analysis (SRA) and Cost Risk Analysis (CRA). In a schedule model, the simulation captures the combined effect of activity duration uncertainty and discrete risk events — including merge bias, which a deterministic network cannot show. In a cost model, it quantifies the range of possible outturn cost given uncertainty in quantities, rates, and risk allowances. The output — most commonly displayed as an S-curve — allows project teams and funders to make informed decisions about contingency and programme dates at defined confidence levels such as P50 or P80.
The simulation is only as good as the inputs. Garbage in, garbage out applies here more than almost anywhere else in project controls. Common pitfalls include three-point estimates that are too narrow (calibration bias), failing to model correlation between related risks, and treating every activity as independent when in reality many of them will be affected by the same underlying causes. Always interrogate the sensitivity outputs — tornado charts and criticality indices — to understand which activities and risks are actually driving the range of outcomes.
Frequently asked
- What is a Monte Carlo simulation in project management?
- A Monte Carlo simulation is a computational technique that runs a project model thousands of times, each time drawing random values from the probability distributions assigned to uncertain inputs (activity durations, costs, risk event impacts). The outputs are aggregated into a probability distribution showing the range of possible project outcomes — typically displayed as an S-curve of completion dates or total costs. The technique is a rigorous method for quantifying cost and schedule risk when multiple uncertain variables interact.
- How many iterations should a Monte Carlo simulation run?
- Most practitioners use 5,000–10,000 iterations as standard, with 10,000 being common on large programmes to ensure the tails of the distribution are well-sampled. Below 1,000 iterations, results can be unstable — running the same model twice may give noticeably different P80 figures. For high-stakes submissions such as gateway reviews or board papers, 10,000 iterations with a fixed random seed is good practice so results are reproducible.
- What is the difference between Monte Carlo simulation and PERT?
- PERT uses a single analytical formula — (O + 4M + P) ÷ 6 — to estimate expected duration and calculates variance analytically. Monte Carlo simulation runs thousands of random iterations and can accommodate any distribution shape, correlations between activities, and discrete risk events that PERT cannot model. For realistic programmes with interacting risks and risk-register events, Monte Carlo is significantly more accurate and the industry standard for formal QRA.
- What does P80 mean in a Monte Carlo result?
- P80 is the value at the 80th percentile of the output distribution — there is an 80% probability the project will complete within that cost or by that date, and a 20% probability of exceeding it. UK government guidance and most major infrastructure clients use P50 as the central planning estimate and P80 for funding approvals, on the principle that public money should be committed at 80% confidence. The choice between confidence levels is a risk-appetite decision: higher confidence requires more contingency.
Related terms
Putting these techniques into practice?
SOMA provides independent project controls consultancy for UK programmes. We can help you apply QRA, EVM, schedule risk analysis, and more.