China Mechanical Engineering ›› 2026, Vol. 37 ›› Issue (7): 1673-1685.DOI: 10.3969/j.issn.1004-132X.2026.07.017

Previous Articles    

A Hybrid Algorithm Based on Decomposition Strategy for DAJSP Solution

WU Zhengyuan1(), ZHAO Shikui1,2(), XIE Ruijian3, YU Fengshun1, LI Tong1, ZHAO Lin1   

  1. 1.School of Mechanical Engineering,University of Jinan,Jinan,250022
    2.Shandong Provincial Key Laboratory of Surface Treatment and Intelligent Equipment for Key Metal Components,University of Jinan,Jinan,250022
    3.JIER Machine-Tool Group Co. ,Ltd,Jinan,250022
  • Received:2025-04-10 Online:2026-07-25 Published:2026-08-18
  • Contact: ZHAO Shikui

基于分解策略混合算法的DAJSP求解

仵政源1(), 赵诗奎1,2(), 解瑞建3, 于丰顺1, 李彤1, 赵林1   

  1. 1.济南大学机械工程学院, 济南, 250022
    2.山东省金属关键构件表面处理与智能装备重点实验室, 济南, 250022
    3.济南二机床集团有限公司, 济南, 250022
  • 通讯作者: 赵诗奎
  • 作者简介:仵政源,男,2001年生,硕士研究生。研究方向为车间生产调度和智能优化算法。发表论文1篇。E-mail:1426095214@qq.com
    赵诗奎*(通信作者),男,1984年生,教授、博士研究生导师。研究方向为车间生产调度、智能优化算法。发表论文65篇。E-mail:me_zhaosk@ujn.edu.cn
  • 基金资助:
    国家自然科学基金(52275490);山东省自然科学基金(ZR2025MS766);山东省科技型中小企业创新能力提升工程(2025TSGCCZZB0845);中央引导地方科技发展资金(YDZX2024127)

Abstract:

This paper addresses the distributed assembly job-shop scheduling problem (DAJSP) by establishing a mixed-integer linear programming (MILP) model to minimize the makespan, and proposes a hybrid genetic-tabu search algorithm with a greedy strategy. The DAJSP is decomposed into two subproblems: job processing and product assembly. For the processing stage, a heuristic algorithm is applied; for the assembly stage, an efficient greedy algorithm is designed to provide a fast and effective scheduling solution. Experimental results on 40 instances show that, compared with a monolithic optimization method, the proposed decomposition strategy reduces computational time and improves solution quality, thereby validating its effectiveness and superiority.

Key words: distributed assembly job-shop scheduling, makespan, hybrid genetic-tabu search, greedy algorithm

摘要:

针对分布式装配作业车间调度问题(DAJSP),以最小化最大完工时间为目标,建立了混合整数线性规划(MILP)模型,提出了一种融合贪心策略的遗传-禁忌搜索算法(GTSAIGS)。将DAJSP分解为工件加工与产品装配的两阶段问题,在加工阶段采用启发式算法优化,针对装配阶段的调度问题设计了一种高效的贪心算法进行快速求解。40个算例的测试结果表明,相比于整体优化方法,GTSAIGS分解策略减少了计算耗时,提高了求解质量,进而验证了所提算法的有效性和优越性。

关键词: 分布式装配作业车间调度问题(DAJSP), 最大完工时间, 混合算法, 贪心算法

CLC Number: