• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2021, Vol. 57 ›› Issue (5): 157-165.doi: 10.3901/JME.2021.05.157

• 数字化设计与制造 • 上一篇    下一篇

扫码分享

装配作业车间的JIT调度研究

吕海利1,2, 朱家涛1, 王正国1, 吴姝1,2   

  1. 1. 武汉理工大学物流工程学院 武汉 430063;
    2. 武汉理工大学港口物流技术与装备教育部工程研究中心 武汉 430063
  • 收稿日期:2020-04-20 修回日期:2020-12-06 出版日期:2021-03-05 发布日期:2021-04-28
  • 通讯作者: 王正国(通信作者),男,1975年出生,副教授。主要研究方向为物流系统决策支持与优化等。E-mail:zgwang@whut.edu.cn
  • 作者简介:吕海利,男,1982年出生,副教授。主要研究方向为物流和供应链系统仿真与优化。E-mail:lvhaili@whut.edu.cn;朱家涛,男,1997年出生,硕士研究生。主要研究方向为智能优化算法在作业车间调度中的应用。E-mail:1569779104@qq.com;吴姝,女,1986年出生,副教授。主要研究方向为供应链与质量管理等。E-mail:wushu0208@hotmail.com
  • 基金资助:
    十三五国家重点研发计划(2019YFB1600406)、国家自然科学基金青年科学基金(11701437)、武汉理工大学自主创新研究基金(195218008)和军工项目JGXM(202018HX02)资助项目。

Research of Just-in-time Scheduling for Assembly Job Shop

LÜ Haili1,2, ZHU Jiatao1, WANG Zhengguo1, WU Shu1,2   

  1. 1. School of Logistics Engineering, Wuhan University of Technology, Wuhan 430063;
    2. Engineering Research Center for Port Logistics Technology and Equipment of Ministry of Education, Wuhan University of Technology, Wuhan 430063
  • Received:2020-04-20 Revised:2020-12-06 Online:2021-03-05 Published:2021-04-28

摘要: 虽然作业车间调度问题在过去几十年里已经得到了广泛而深入的研究,但大部分研究都是以正规指标(Regular measures)最小化为目标。正规指标的调度只需要将工序尽量提前即可。相对于正规指标,最小化提前和延迟成本之和等则属于非正规指标(Non-regular measures)。对于非正规指标调度,将所有工序尽量提前并不能优化目标,其调度方案的生成过程必然更加复杂。针对提前和延迟成本最小化这类非正规指标问题,将研究对象从作业车间调度(Job shop scheduling,JSP)扩展到了装配作业车间调度(Assembly job shop scheduling,AJSP),以更广泛地探讨求解此类问题的一般规律。设计了一种三阶段调整的启发式算法以生成调度方案,并结合遗传算法的求解框架进行了分析。通过与最优解结果对比,证明该启发式算法的效率和有效性。另外,试验结果也揭示了不同求解因子对求解效果的影响,为后续研究提供参考。

关键词: 装配作业车间调度, 提前和延迟成本, 遗传算法, 启发式算法

Abstract: Although job shop scheduling problem (JSP) has been extensively and intensively studied in the past few decades, most research aims at minimizing regular measures. The scheduling of the problems with regular measures only needs to process as early as possible. Compared with regular measures, minimizing the sum of earliness/tardiness penalties is a non-regular measure. For the scheduling problems with non-regular measures, it is impossible to optimize the goal by processing all the operations as early as possible, and the schedule generation process must be more complicated. Aiming at the scheduling problem with earliness and tardiness penalties, the subject of research is expanded from simple job shop (JSP) to assembly job shop (AJSP) to explore the general law of earliness and tardiness scheduling problems more widely. A three-stage adjustment heuristic is designed to generate a schedule, and the heuristic is combined with genetic algorithm for experimental analysis. Compared with the results of optimal results, the efficiency and effectiveness of the heuristic is proved. In addition, experimental results also reveal the impact of different methods in the process of generating a schedule, which may serve as guidelines for future research.

Key words: assembly job shop scheduling, earliness and tardiness cost, genetic algorithm, heuristic

中图分类号: