Parametric Optimization: Response Surfaces Neural Networks

Abstract

This chapter will discuss one of the oldest simulation-based methods of parametric optimization — namely, the response surface method. For simulation-optimization purposes, the response surface method (RSM) is admittedly primitive. But it will be some time before it moves to the museum because it is a very robust technique that often works well when other methods fail. It hinges on a rather simple idea — that of obtaining an approximate form of the objective function by simulating the system at a finite number of points, which are carefully sampled from the function space. Traditional RSM usually uses regression over the sampled points to find an approximate form of the objective function.