Abstract:  Aiming at the distributed flexible job shop green scheduling problem with transportation time， a mixed-integer programming model for minimizing makespan and total energy consumption was established， and an INSGA-Ⅱwas proposed. A double-layer encoding scheme was adopted based on operation and machine， and a greedy insertion decoding method was designed considering transportation time. Considering processing time and energy consumption, an initialization method was designed to improve the quality of the population， and the multi-parent crossover and new mutation operations were used to update the population. A variable neighborhood search strategy was embedded to improve the quality of Pareto front. The effectiveness of the proposed method was verified by a series of experiments. 

Key words: distributed flexible job shop, transportation time, makespan, total energy consumption, improved non-dominated sorting genetic algorithm Ⅱ（INSGA-Ⅱ）

摘要： 针对考虑运输时间的分布式柔性作业车间绿色调度问题，建立了最小化完工时间和总能耗的混合整数规划模型，并提出了一种改进的非支配排序遗传算法。采用基于工序和机器的双层编码方式设计了考虑运输时间的贪婪插入解码方法。为提高种群的质量，设计了同时考虑加工时间和能耗的初始化方法，应用多父代交叉和新的变异操作更新种群，并嵌入一种变邻域搜索策略以提高Pareto前沿的质量。多组实验验证了该方法的有效性。

关键词: 分布式柔性作业车间, 运输时间, 完工时间, 总能耗, 改进的非支配排序遗传算法