Stage 1 Define a distribution of possible inputs for each input random variable: Requires recognition of the underlying probability distribution of the process. This may be directly apparent or may require empirical observation of the process under investigation. |
Stage 2 Generate outputs randomly from those distributions: Requires the selection of an appropriate random number generator to model the observed probability distribution. Random number generators are generally available in most statistical software packages and Micro Soft Excel. |
Stage 3 Perform a deterministic computation using that set of outputs: Involves computing the desired output variable or variables from the generated random numbers. |
Stage 4 Aggregate the results of the individual computations into the final result: The aggregation process is dependent on the specific simulation but could be as straightforward as computing the average of the simulated results. |