Simulation Modeling and Analysis

Catalog description

This course provides a study of discrete-event systems simulation in the areas of queuing, inventory and reliability systems as well as Markov Chains, Random-Walks and Monte-Carlo systems. The course examines many probability distributions used in simulation studies as well as the Poisson process. Fundamental to most simulation studies is the ability to generate reliable random numbers and so the course investigates the basic properties of random numbers and techniques used for the generation and testing of pseudo-random numbers. Random numbers are then used to generate other random variable using the methods of inverse-transform, convolution, composition and acceptance/rejection. Finally, since most inputs to simulation are probabilistic instead of deterministic in nature, the course examines some techniques used for identifying the probabilistic nature of input data. These include identifying distributional families with sample data, using maximum-likelihood methods for parameter estimating within a given family and testing the final choice of distribution using chi-squared goodness-of-fit.