摘要

A memetic algorithm, which combines globe search with local search strategies, is presented to deal with the multiprocessor scheduling problem(MSP). During the processes, an improved particle swarm optimization is employed to execute the globe search optimization, and the simulated annealing is adopted to improve the quality of the selected candidates based on a certain strategy. Simulations show that the proposed method performs well on the globe exploration. Experimental results based on MSP show that the algorithm achieved an efficient makespan.