An enhanced RSM algorithm using gradient-deflection and second-order search strategies

This paper presents a new methodology for improving the search techniques currently being used in standard Response Surface Methodology (RSM) algorithms. RSM is a collection of mathematical and statistical techniques used for experimental optimization. Our improved RSM algorithm incorporates certain gradient deflection methods, augmented with appropriate restarting criteria, as opposed to using the path of steepest descent as the only search direction. In order to empirically investigate our new RSM algorithm in comparison with the standard RSM techniques, a set of standard test functions is used. We consider two cases for each test function; with a random perturbation added to the function and without a random perturbation added to the function. Computational results exhibit the improvements achieved under the proposed algorithm.