计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (4): 238-247.DOI: 10.3778/j.issn.1002-8331.1711-0076

• 工程与应用 • 上一篇    下一篇

多工作日历下流水作业调度遗传优化方法

曾  强1,邓敬源1,张进春1,沈  玲2   

  1. 1.河南理工大学 能源科学与工程学院,河南 焦作 454000
    2.河南理工大学 安全科学与工程学院,河南 焦作 454000
  • 出版日期:2019-02-15 发布日期:2019-02-19

Optimization Scheduling Method Based on GA for FSP Under Multiple Work Calendars

ZENG Qiang1, DENG Jingyuan1, ZHANG Jinchun1, SHEN Ling2   

  1. 1.School of Energy Science and Engineering, Henan Polytechnic University, Jiaozuo, Henan 454000, China
    2.School of Safety Science and Engineering, Henan Polytechnic University, Jiaozuo, Henan 454000, China
  • Online:2019-02-15 Published:2019-02-19

摘要: 针对多工作日历下的流水作业调度问题,提出了一种遗传优化方法。首先,提出了基于多工作日历的时间推算方法,解决了多工作日历下流水作业调度的关键问题:以Excel为平台设计了“工作制”工作表和“设备”工作表,在“设备”工作表中为每台设备指定工作制并设定工作时段;在此基础上,以Excel VBA为平台设计了5个基于工作日历的时间推算函数。其次,以Excel VBA为平台设计了遗传算法用于求解问题:个体采用整数编码方式,交叉操作采用“交换交叉”方式,变异操作采用“交换变异”方式,解码过程采用基于多工作日历的时间推算方法准确计算各工序开工和完工时刻。最后,通过案例分析验证了所提方法的有效性。

关键词: 流水作业调度问题, 遗传算法, 多工作日历, 时间推算方法

Abstract: For the Flow Shop Scheduling Problem(FSP) under multiple work calendars, this paper proposes an optimization scheduling method based on Genetic Algorithm(GA). Firstly, a time reckoning method based on multiple work calendars is given as a critical technology to solve the researched problem:Worksheets “work system” and “machine” are designed by Excel. The work system and work time periods are specified to each machine in the worksheet “machine”. On above basis, five time reckoning functions based on the machine’s work calendar are designed by Excel VBA. Secondly, a genetic algorithm is developed to solve the problem by Excel VBA. In the algorithm, an integer coding method is used to code the individuals, a swapping method is used in the crossover and mutation operation, and the above time reckoning method based on multiple work calendars is used to calculate start time and end time of each operation accurately. Finally, the effectiveness of the proposed method is validated by case study.

Key words: flow shop scheduling problem, genetic algorithm, multiple work calendars, time reckoning method