Monte Carlo Process

The Monte Carlo process is an effective means to propagate uncertainty in inputs to characterize the range of uncertainty in risk assessment model outputs. The Monte Carlo process begins by generating a set of random numbers. Each of these numbers is then transformed into a value that is useful in a risk assessment model. The randomly generated model inputs are used to complete the calculations of the model thus generating potential model outputs. Expected values of model outputs may stabilize with a 10² order of magnitude of iterations. Outcome probabilities may be estimated with a 10³ order of magnitude of iterations. Tails of output distributions may be defined by 10⁴, and extreme events may be estimated by an order of magnitude of 10⁵ or more. The Latin hypercube method of sampling is more efficient for smaller samples and may be preferred to Monte Carlo sampling.