Two-stage hybrid flowshop scheduling in a single factory has been considered fully; however, the distributed two-stage hybrid flowshop scheduling problem (DTHFSP) is seldom studied in a multi-factory environment. In this study, a DTHFSP with sequence-dependent setup times is investigated and a shuffled frog-leaping algorithm with memeplex grouping (MGSFLA) is proposed to minimize makespan and the number of tardy jobs. After an initial population is generated by an heuristic, two phases are executed sequentially with a new population division. In the second phase, all memeplexes are categorized into three groups, the different search processes are implemented in the different groups and the best memeplex is excluded from population division. A number of experiments are conducted on many instances and computational results validate the effectiveness of the main strategies and the promising advantages of an MGSFLA.

中文翻译：

---

使用带有 memeplex 分组的 shuffled 蛙跳算法解决分布式两阶段混合流水车间调度

充分考虑了单厂两级混合流水车间调度；然而，在多工厂环境中很少研究分布式两级混合流水车间调度问题（DTHFSP）。在这项研究中，研究了具有序列相关设置时间的 DTHFSP，并提出了一种带有 memeplex 分组（MGSFLA）的混洗青蛙跳跃算法，以最小化完工时间和延迟作业的数量。在启发式生成初始种群后，两个阶段按顺序执行，并进行新的种群划分。在第二阶段，将所有 memeplex 分为三组，在不同的组中执行不同的搜索过程，并将最佳 memeplex 排除在种群划分之外。