The distributed blocking flowshop scheduling problem (DBFSP) plays an essential role in the manufacturing industry and has been proven to be as a strongly NP-hard problem. In this paper, an ensemble discrete differential evolution (EDE) algorithm is proposed to solve the blocking flowshop scheduling problem with the minimization of the makespan in the distributed manufacturing environment. In the EDE algorithm, the candidates are represented as discrete job permutations. Two heuristics method and one random strategy are integrated to provide a set of desirable initial solution for the distributed environment. The front delay, blocking time and idle time are considered in these heuristics methods. The mutation, crossover and selection operators are redesigned to assist the EDE algorithm to execute in the discrete domain. Meanwhile, an elitist retain strategy is introduced into the framework of EDE algorithm to balance the exploitation and exploration ability of the EDE algorithm. The parameters of the EDE algorithm are calibrated by the design of experiments (DOE) method. The computational results and comparisons demonstrated the efficiency and effectiveness of the EDE algorithm for the distributed blocking flowshop scheduling problem.

分布式阻塞流水车间调度问题（DBFSP）在制造业中起着至关重要的作用，并且已被证明是一个很强的NP难题。本文提出了一种集合离散微分进化算法（EDE），以解决分布式制造环境中最小化制造周期的阻塞流水车间调度问题。在EDE算法中，候选对象表示为离散的作业排列。集成了两种启发式方法和一个随机策略，为分布式环境提供了一组理想的初始解决方案。这些启发式方法考虑了前端延迟，阻塞时间和空闲时间。重新设计了变异，交叉和选择运算符，以帮助EDE算法在离散域中执行。与此同时，在EDE算法框架中引入了精英保留策略，以平衡EDE算法的开发和探索能力。EDE算法的参数通过实验设计（DOE）方法进行校准。计算结果和比较结果证明了EDE算法在分布式阻塞流水车间调度问题上的有效性和有效性。