Research On Multi-objective Flexible Flow Shop Scheduling Problem Based On Customers’ Satisfaction

This paper studies the multi-objective flexible flow shop scheduling problem with minimizing makespan,minimizing the ratio of processing delay to improve customer satisfaction,and minimizing the total electricity cost under time-of-use electricity price.First,stand on the position of customer’s and based on reality,then design an objective function that can better measure user satisfaction—the ratio of processing delay based.Based on these objectives and the characteristics of FFSP,a multi-objective FFSP mathematical model is constructed.Then,according to the characteristics of the problem,chooses the NSGA-Ⅱ to solve the multi-objective FFSP.In the preliminary experiments,it was found that NSGA-Ⅱ has the problems of premature convergence,crossover between repeated solution,and selection operator consuming too much time.Aiming to these shortcomings,a hybrid algorithm combining GRASP and NSGA-Ⅱ is designed to solve the problem of repeated solutions in NSGA-Ⅱ;an adaptive crossover probability based on the similarity of parents is designed to reduce the possibility of the crossover between close relatives then improve the diversity of the population;based on the concept of Pareto,an elective selection is designed to reduce the time cost of NSGA-Ⅱ iterative in the selection phase.Meanwhile,designs 12 sets of test cases for numerical experiments which designed by the predecessors in FFSP.The test results show that GRASP and adaptive can effectively improve the exploitation ability of NSGA-Ⅱ when finds solutions,so that the solution set has more and more diverse non-dominated solutions;elective selection iteration speed is increased at the cost of a small reduction in the quality of the solution.Finally,the optimized hybrid NSGA-Ⅱ and Wang’s INSGA-Ⅱ algorithms are used in 12 sets of test cases to conduct numerical experiments with time as the stopping criterion.The results show that the optimized HNSGA-Ⅱ can obtain a solution set with better quality,quantity and diversity than NSGA-Ⅱ and Wang’s INSGA-Ⅱ.