Abstract: For the transformation of manufacturing enterprises to a distributed production structure with multiple factories,a multi-objective distributed flow shop scheduling problem was proposed considering order delivery time,and a  mixed integer programming model was built aiming at minimizing makespan and total weighted earliness and tardiness penalty.A multi-objective brain storm optimization combining with the characteristics of this problem was designed as a solution.To enhance its search ability,an adaptive clustering strategy was introduced according to the dominated rule,and two crossover methods as well as four neighborhood structures were employed based on the problem features.Experimental results indicated that the proposed approach could achieve superior performance.

Key words: distributed flow shop scheduling, order delivery time, brain storm optimization, multi-objective optimization

摘要： 针对制造企业向多工厂分布式生产结构的转变,提出一个考虑订单运输时间的多目标分布式流水车间调度问题,并构建了一个以最小化最大完工时间与总加权提前和拖期惩罚为目标的混合整数规划模型,进而结合问题特点提出一种多目标头脑风暴优化算法进行求解。为了提高算法的搜索能力,利用支配规则设计了一种自适应聚类策略,并基于问题特征采用两种交叉方法和4种邻域结构。实验结果验证了所提算法的优越性。

关键词: 分布式流水车间调度, 订单运输时间, 头脑风暴优化, 多目标优化