新型教与同伴学习粒子群算法求解作业车间调度问题  

Novel teaching and peer-learning-based particle swarm optimization for Job-Shop scheduling problem

作　　者：董君 叶春明[1] Dong Jun;Ye Chunming(Business School,University of Shanghai for Science&Technology,Shanghai 200093,China;Henan Institute of Technology,Xinxiang Henan 453000,China)

机构地区：[1]上海理工大学管理学院,上海200093 [2]河南工学院,河南新乡453000

出　　处：《计算机应用研究》2019年第12期3764-3768,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(71271138);上海理工大学科技发展项目(2018KJFZ043)

摘　　要：针对最小化最大完工时间的单目标作业车间调度问题,提出了新型教与同伴学习粒子群算法。通过教学阶段融合多邻域搜索,采用多样性变异策略以及同伴学习阶段采用混合学习策略三个方面的改进操作,扩大了种群的多样性,避免算法陷入局部最优,算法收敛速度和寻优性能有了显著提高。通过作业车间调度问题FT、LA系列测试实例的对比实验,验证了新型教与同伴学习粒子群算法是解决单目标作业车间调度问题的有效方法。This paper applied a novel hybrid teaching and peer-learning-based particle swarm optimization( HTLBPSO) to minimize maximum makespan of single objective Job-Shop scheduling problem( JSP). The population diversity increased and also avoided being trapped in local optimum through adopting three improved operations,fusion multi-neighbor search,multiple mutation in teaching stage and mix learning in peer-learning stage. The optimization performance and convergence speed significantly improved. The contrast experimental results of FT and LA series test cases for Job-Shop scheduling problems demonstrate the effectiveness of the algorithm proposed on solving single objective JSP.

关 键 词：教与同伴学习粒子群算法 作业车间调度 多邻域搜索 多样性变异 混合学习策略 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]