China Mechanical Engineering ›› 2024, Vol. 35 ›› Issue (02): 260-267,279.DOI: 10.3969/j.issn.1004-132X.2024.02.010

Previous Articles     Next Articles

Scheduling in SAT in Multi-operation Mode Based on Artificial Hummingbird Algorithm with Twin Population

WANG Hong1,2;WU Lihui3;CHEN Da1,2;ZHANG Jie2   

  1. 1.College of Mechanical Engineering,Donghua University,Shanghai,201620
    2.Institute of Artificial Intelligence,Donghua University,Shanghai,201620
    3.College of Mechanical Engineering,Shanghai Institute of Technology,Shanghai,201418

  • Online:2024-02-25 Published:2024-04-12

基于孪生人工蜂鸟算法的多作业模式半导体封测环节调度

王洪1,2;吴立辉3;陈达1,2;张洁2   

  1. 1.东华大学机械工程学院,上海,201620
    2.东华大学人工智能研究院,上海,201620
    3.上海应用技术大学机械工程学院,上海,201418

  • 通讯作者: 张洁(通信作者),女,1963 年生,教授、博士研究生导师。研究方向为大数据驱动的智能制造系统、制造系统的建模仿真调度,发表论文100余篇。E-mail:mezhangjie@dhu.edu.cn。
  • 作者简介:王洪,男,2000年生,硕士研究生。研究方向为复杂制造系统调度、制造大数据分析。E-mail:2221044@mail.dhu.edu.cn。
  • 基金资助:
    国家重点研发计划(2022YFB3305003)

Abstract:  In order to solve the scheduling problem of SAT in multi-operation mode, an artificial hummingbird algorithm with twin population was proposed with the goal of minimizing the maximum completion time. The twin population mechanism was designed to improve the solution accuracy. By double decoding, twin population generation and cooperation methods, the searching space for solutions was expanded, the quality of the initial population solution was improved, and the diversity of population solutions was increased in optimization processes. By the bidirectional-guiding foraging strategy, the relationship between algorithm diversity and convergence was balanced, and algorithm stability was enhanced. By the strategy of four-variable neighbor searching, the local optimization ability of the algorithm was enhanced. The test results show that the proposed method may effectively shorten the maximum completion time of the SAT.

Key words: production scheduling, semiconductor assembly and test(SAT), multi-operation mode, artificial hummingbird algorithm with twin population(AHA-TP)

摘要: 针对多作业模式的半导体封装测试环节调度问题,以最小化最大完工时间为目标,提出了孪生人工蜂鸟算法。设计了孪生种群机制,通过构建双解码、孪生种群生成与协作方法,扩大解的搜索空间,提高初始解的质量,增加优化过程中解的多样性,进而提高求解精度。通过双向引导觅食策略,平衡算法多样性与收敛性,增强算法稳定性。通过构建四邻域搜索策略,增强算法局部优化能力。实验结果表明,该方法能有效缩短半导体封测环节的最大完工时间。

关键词: 生产调度, 半导体封装测试, 多作业模式, 孪生人工蜂鸟算法

CLC Number: