
Answering a problem by simulating random draws from it — from detector response to lattice field theory.Monte Carlo simulation is a computational technique that uses random sampling to obtain numerical results for problems that might be deterministic in principle but are too complex for exact analytical solutions. The name comes from the Monte Carlo Casino in Monaco, reflecting the role of chance in the method. It is used across physics, from modeling the passage of particles through detectors to calculating properties of quantum field theories on a lattice, where direct analytical calculations are impossible.
The method was developed during the Manhattan Project in the 1940s by Stanislaw Ulam, John von Neumann, and Nicholas Metropolis, who coined the name. It was first applied to neutron diffusion problems.
The core idea is to set up a model of the system, then repeatedly draw random samples from the relevant probability distributions, run the model with those samples, and aggregate the outcomes. For example, to estimate the average energy of a particle in a potential, you would randomly sample positions and velocities according to the Boltzmann distribution, compute the energy for each, and average. The law of large numbers ensures that as the number of samples grows, the average converges to the true value. The accuracy improves only as the square root of the number of samples, so clever variance-reduction techniques are often needed.
Used in particle physics to simulate detector responses, in cosmology to model structure formation, and in condensed matter physics to study phase transitions via lattice models.
The first Monte Carlo calculation was done on the ENIAC computer in 1948, using a random number generator based on a roulette wheel.



