带恶化和学习效应的不相关并行机调度优化  

Optimization on Unrelated Parallel Machine Scheduling with Deteriorating and Learning Effects

作　　者：轩华 张寒 王薛苑 XUAN Hua;ZHANG Han;WANG Xue-yuan(School of Management Engineering,Zhengzhou University,Zhengzhou 450001,China)

机构地区：[1]郑州大学管理学院,河南郑州450001

出　　处：《控制工程》2023年第5期769-778,共10页Control Engineering of China

基　　金：国家自然科学基金资助项目(U1804151);河南省科技攻关计划项目(202102310310,232102321093,232102321026)

摘　　要：研究了以同时最小化makespan和总加权拖期为目标的具有恶化和学习效应的不相关并行机调度问题。针对此类NP-hard问题,设计了基于两段式编码的改进模拟退火算法,结合随机程序和均匀分配策略分别产生第一段的工件加工序列编码和第二段的机器加工信息编码,以获取问题初始调度解,进而提出了分段式交换和变异扰动操作以得到更新后的新解。通过仿真实验测试改进模拟退火算法,将其与一些启发式算法对比,结果表明,所提算法可获得更好的近优解。Industry 4.0 takes intelligent manufacturing integrating modern information technology with advanced manufacturing technology as the main line,and proposes a novel production mode where the application of intelligent machines with learning ability enhances the production efficiency of manufacturing enterprises.It also puts forward a new challenge for the production scheduling under the new environment.Based on these,an unrelated parallel machine scheduling problem with deteriorating and learning effects is studied,which aims to simultaneously minimize makespan and total weighted tardiness.To solve this NP-hard problem,an improved simulated annealing algorithm based on two-segment encoding is designed.A random procedure and a uniform assignment strategy are respectively used to generate job processing sequence coding of the first segment and machine processing information coding of the second segment so as to obtain the initial solution of the studied problem.Further,segmented exchange and mutation disturbance operations are proposed to achieve the new solution after updating.Simulation experiments are performed to test the improved simulated annealing algorithm which is compared with several heuristic algorithms.The experimental results show that the proposed algorithm can obtain better near-optimal solution.

关 键 词：学习效应 恶化效应 不相关并行机调度 改进模拟退火算法 两段式编码 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]