Robust design through the use of a hybrid genetic algorithm

In this paper we present a hybrid optimization approach to perform robust design. The motivation for this work is the fact that many realistic engineering systems are mutimodal in nature with multiple local optima, and moreover may have one or more uncertain design parameters. The approach that is presented utilizes both local and global optimization algorithms to find good design points more efficiently than either could alone. The mean and variance of the objective function at a design point is calculated using Monte Carlo simulation and is used to drive the optimization process. To demonstrate the usefulness of this approach a case study is considered involving the design of a beam with dimensional uncertainty.