A parallel island model genetic algorithm for the multiprocessor scheduling problem

Abstract

In this paper we compare the pe~Jormance o] a serial and a parallel island model Genetic Algorithm for solvin 9 the Multiprocessor Scheduling Problem. We show results usin 9 fixed and scaled problems both using and not usin9 migration. We have found that in addition to providing a speedup through the use o] parallel processing, the parallel island model GA with migration finds better quality solutions than the serial (.;A.