中文摘要：

针对半导体晶圆制造系统（SWFS）调度中的多目标优化和目标时变性问题，提出了一种时变多目标（TVMO）调度算法。基于模糊理论研究了时变模糊集合与多目标重要程度的复杂非线性量化关系，进而研究了SWFS的优化调度方法。所提出的方法兼顾最大产品交期率、最大生产移动量和瓶颈机台最大产能利用率三个目标，根据时变目标权重计算出在制品的加工优先权序列号。大量的仿真实验数据证明，该调度算法能够改善系统多个绩效目标，可为大规模复杂重入型制造系统的科学生产控制与调度提供有效支持。

英文摘要：

The paper presents a fuzzy theory-based time-variable-multiple-objectives （TVMO） scheduling algorithm for achieving complex semiconductor wafer fabrication systems （SWFS）＇ multiple-objective optimization and objective time-variation. The fuzzy membership was assigned to describe the complex non-linear relations between the time-variable membership set and weight of different objective. Therefore, it provides a potential way to dealing with the scheduling optimization in the SWFS. Through above method of integration, each WIP wafer lot in the fab can get a distinct lot priority index of TVMO feature which is taking account of three multiple objectives such as maximum on-time delivery, maximum operation moves and high equipment utilization. The results of a large number simulation experiment prove that the TVMO scheduling approach can improve those production indices of multiple objectives successfully. It also provides supports to the related researches of production scheduling and control for a typically large-scale complicated re-entrant manufacturing system （LSCRMS）.