Public Courses

This simulation course is for operations research analysts and systems & industrial engineers who would like to use simulation to design and optimize real-world systems. It presents methods for developing a simulation model, ensuring its validity, choosing simulation software, selecting input probability distributions, analyzing simulation runs, and project management. The importance of current “hot” topics in simulation modeling such as AI/machine learning, digital twins, and simulation-specific cloud computing will be carefully explained.

This simulation course discusses practical statistical techniques for comparing alternative system designs, the use of AI/machine learning to develop and analyze simulation models, experimental design techniques for determining important system factors, agent-based simulation, and simulation-based optimization.

Simulation models often have many input factors and determining which ones are really important can be quite difficult. We give a comprehensive presentation of design of experiments specifically for simulation modeling, whose major goal is to determine which factors have the greatest effect on the responses. We also discuss how to develop a surrogate for a simulation model that can be used to predict the model response in real time.

Simulation models are a surrogate for physical experimentation with a system, which is usually not cost-effective, disruptive, or simply impossible. Thus, if the model is not “valid,” any conclusions drawn from the model results are likely to be erroneous. In this seminar all major validation techniques are discussed and illustrated by examples based on actual simulation projects.

This one-day seminar provides a comprehensive introduction to discrete-event simulation, agent-based simulation, and system dynamics, and when to use each. For each type we discuss typical application areas and recommended software packages, and also situations where several types might advantageously be used together in one composite model. Numerous simulation models will be discussed and demonstrated.