Multi-population and Self-adaptive Genetic Algorithm Based on Simulated Annealing for Permutation Flow Shop Scheduling Problem

In order to solve the permutation flow shop scheduling problem, a multi-population and self-adaptive genetic algorithm based on simulated annealing is proposed in this paper. For the precocity problem of traditional genetic algorithm, the multi-population coevolution strategy is adopted. We introduce a squared term to improve traditional self-adaptive genetic operators, which can increase the searching efficiency and avoid getting into local optimum. A new cooling strategy is proposed to reinforce the ability of overall searching optimal solution. The algorithm is used to solve a series of typical Benchmark problems. Moreover, the results are compared with SGA, IGA, and GASA. The comparison demonstrates the effectiveness of the algorithm.