Monte Carlo Simulation Calculator

This calculator simulates a single number - one cost or one duration - from a triangular distribution. That already beats a point estimate, but a real project is not one number. Its risks interact: a supplier delay drives overtime, which drives cost, which strains the schedule further.

Incertive extracts every uncertainty from your actual plan, models how they correlate, runs thousands of scenarios, and returns a success probability with sensitivity analysis - so you know not just the range, but which risk to fix first. See a sample analysis.

What is a Monte Carlo simulation?

A Monte Carlo simulation runs a model thousands of times, each time drawing random values from the ranges you provide, to reveal the full distribution of possible outcomes instead of a single point estimate. This calculator draws 10,000 samples from a triangular distribution defined by your optimistic, most likely, and pessimistic values, then reports the percentiles.

What do P50, P80, and P90 mean?

They are confidence levels. P50 is the median: there is a 50% chance the real outcome is at or below it. P80 means an 80% chance of landing at or below that value, so P80 is a common figure to commit to because it builds in a buffer. P90 is more conservative still. The gap between P50 and P90 is your risk exposure.

Why not just use the most likely estimate?

Because a single estimate hides risk. Real projects have asymmetric downside - the pessimistic tail is usually longer than the optimistic one - so the most likely value systematically understates cost and duration. The simulation shows how far the outcome can drift and how likely each level is.

Can ChatGPT do this?

A chatbot can describe the idea of a range, but it cannot run the simulation, and it returns a different answer each time you ask. This calculator computes the actual distribution and is seeded, so the same inputs reproduce the same result. Incertive extends this to your whole plan: correlated risks, thousands of scenarios, and a success probability you can defend.

What distribution does this use?

A triangular distribution, defined by your three points (optimistic, most likely, pessimistic). It is the standard choice when you have a realistic minimum, mode, and maximum but no historical dataset. Incertive supports normal, beta, and triangular distributions and correlations between uncertainties for a more precise model.