Abstract: The single-factory manufacturing is gradually transiting to the multi-factory collaborative production with the economic globalization,and the distributed resource with diverse demand make it more challenged to schedule the production among multiple factories effectively.For this reason,the Distributed Flexible Job Shop Scheduling Problem (DFJSP) was studied.A bi-objective optimization model was formulated to minimize the earliness/tardiness and the total cost simultaneously.An Improved Differential Evolution Simulated Annealing Algorithm (IDESAA) was proposed,and two crossover and mutation operators were designed.With the strong robustness,simulated annealing was used to local search the best Pareto solutions.The greedy selection combined with the fast elitist Non-Dominated Sorted Genetic Algorithm's (NSGA-Ⅱ) selection was employed to select the offspring.The comprehensive experiments were conducted and the results showed that the proposed algorithm could solve DFJSP effectively and efficiently.

Key words: distributed flexible job shop scheduling problem, multi-factory collaborative production, improved differential evolution algorithm, bi-objective optimization model, cost, earliness/tardiness

摘要： 经济全球化使制造业从单工厂模式转变为多工厂协同生产模式,制造资源异地化、客户需求多样化使得多工厂的分布式调度难度急剧增加,为此研究了分布式柔性作业车间调度问题。首先建立了该问题的双目标优化模型,同时优化总成本和提前/延期惩罚。然后提出改进的差分进化算法,设计了两种变异机制以及两种交叉方式,结合模拟退火的鲁棒性进行局部搜索,并结合贪婪和带精英策略的快速非支配排序遗传算法的选择思想设计选择操作,产生下一代继续进行迭代进化。最后,通过综合实验证明了所提模型和算法能够很好地求解此类问题。

关键词: 分布式柔性作业车间调度问题, 多工厂协同生产, 改进差分进化算法, 双目标优化模型, 总成本, 提前/延期惩罚