ADAPTIVE STOCHASTIC OPTIMIZATION USING MULTIPROCESSORS

A study is presented of a multiprocessor-based adaptive random search optimization algorithm that is suited for application to high-order sy stems with many parameters. Guidelines are established for optimum per formance of the search technique in regard to the interaction of the o ptimization parameters and the influence of the number of processors b eing used. Analytical results are presented to show the influence of t he random number distribution characteristics on the probability of su ccessful improvement in the cost function, thereby suggesting more eff icient search strategies. Furthermore, the relationship between the to tal number of processors and the corresponding rate of improvement in speed relative to sequential machines is presented. The class of probl ems likely to benefit from parallel processing is discussed. The compa rative efficiency of the adaptive random search algorithm for differen t levels of parallelisms is illustrated through simulation studies.