批量生产集成调度问题的嵌套式混合遗传算法

Nested hybrid genetic algorithm for integrated economic lot and delivery scheduling problem

针对多级供应链环境中制造商利用混合流水车间方式生产时如何确定生产、采购及产品运送决策的问题,提出基于混合流水车间调度的批量生产与运送调度模型。考虑双重调度的复杂性,设计了两层嵌套式混合遗传算法,外层针对批量生产与运送调度,限制搜索最优运送频率及生产周期;内层解决混合流水车间调度,特别设计染色体的复式均匀交叉机制,求解产品生产排序及相应的原材料采购批量问题。通过算例验证了算法对解空间的搜索广度、准确度和全局收敛性能,为进一步研究较大规模的复杂供应链调度问题提供了可借鉴的方法。

Aiming at the problems of manufacturer in multistage supply chain to make optimal productionrelated de cisions by using Hybrid Flow Shop （HFS） way, an Economic Lot and Delivery Scheduling Problem （ELDSP） model based on HFS Scheduling Problem （HFSP） was proposed. The twolayer nested genetic algorithm was designed by considering the scheduling complexity, of which the outer hierarchy for limiting to search optimal delivery frequen cies and production cycle of ELDSP, and the inner hierarchy for solving production sequence and corresponding lot sizes by designing duplex uniform crossover method of HFSP, Through the numerical studies, the proposed method was proved to provide reference for researching scheduling problem of largescale complex supply chain.