具有行为效应的TFT-LCD模块组装多目标调度研究  

Research on TFT-LCD module assembly multi-objective scheduling problem with behavior effects

作　　者：吴思思 叶春明[1] 李瑞婷[1] Wu Sisi;Ye Chunming;Li Ruiting(Business School,University of Shanghai for Science ＆ Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学管理学院,上海200093

出　　处：《计算机应用研究》2018年第12期3716-3721,共6页Application Research of Computers

基　　金：国家自然科学基金资助项目(71271138);上海市高原学科项目(GYXK1201);上海理工大学科技发展基金项目(16KJFZ028)

摘　　要：为了优化调度方案,针对最小化最大完成时间、机器总空闲时间和工件总延期时间的具有学习退化效应的TFT-LCD模块组装多目标调度问题,提出改进混沌烟花算法。基于两段式编码,结合加工时间优先选择及动态淘汰锦标赛规则构建帕累托非劣解集。仿真实验表明:改进烟花算法求解质量优于烟花算法、粒子群算法,学习退化效应增强将减少可行解数量,学习率、退化因子增大使得非劣解远离原点。This paper applied the improved chaos fireworks algorithm to minimize makespan,total machine idle time and total delay time of the TFT-LCD module assembly multi-objective scheduling model with learning and deterioration effect. Based on two segment encoding,the algorithm could construct the Pareto model by combining processing time priority rules,dual tournament and dynamic elimination rule. Simulation experiments show that the improved chaos fireworks algorithm is better than fireworks algorithm and particle swarm algorithm. The reinforcement of learning and deterioration effects reduce the number of feasible scheduling. The increase of learning rate and deterioration factor make Pareto solutions away from the origin.

关 键 词：学习退化效应 薄膜晶体管液晶显示器 改进烟花算法 帕累托非劣解集 

分 类 号：TH165[机械工程—机械制造及自动化]