A robust design tutorial

Describes a framework for discrete event simulation which synthesizes Taguchi\´s (1986, 1987) robust design philosophy and a response surface metamodeling approach. We show how the use of a loss function which incorporates both system mean and system variability can be used to efficiently and effectively carry out system optimization and improvement efforts. The results can yield new insights into system behavior, and may recommend system configurations which differ substantially from those selected by analysis solely of the mean response. Issues of model validation and model complexity can also be addressed. The tutorial is meant for both practitioners and researchers. We assume a knowledge base at the level of Chapters 11 and 12 of "Simulation Modeling and Analysis" (Law and Kelton, 1991), but review essential elements and give illustrative examples.