一类多目标优化的半Flow Shop调度问题研究

摘要

半flow shop生产调度问题是一种类似于flow shop但又与之有本质区别的特殊调度问题，根据工件的工艺和客户要求，同一生产线上以批为单位的工件的加工可以跳过其中的某些工序。基于企业实际生产情况，建立了目标为产品的所有工件完工时间最小、拖期时间最少和设备的空闲时间最短的多目标优化模型。针对该问题，设计了一种自适应多目标和声搜索算法，算法中引入了信息熵的概念，使种群中个体的多样性的评价更加准确。通过对该算法与其他算法进行对比分析，说明该算法具有一定的优越性，在解决多目标优化问题上具有良好的应用前景。%The semi-flow shop scheduling problem is both similar to but also distinctively different in nature from the conventional shop scheduling problem, wherein according to work piece requirement and customer request, the processing of the batched parts on a same production line may skip some sequences. In this paper, we designed a multi-objective optimization model targeting at the shortest working time, least postponement and shortest equipment idling time. Next, we developed an adaptive multi-objective harmony search algorithm which incorporated the concept of information entropy to more accurately assess the diversity of the individuals in the population. Then, through comparing this algorithm and other algorithms, we demonstrated the superiority of this algorithm.