考虑多技能水平工人分配的复杂产品装配调度问题  

Complex Product Assembly Scheduling Problem Considering Allocation of Workers with Multiple Skilled Level

作　　者：王培旭 裴凤雀 刘检华[1] 郭昊鑫 庄存波[1,2] WANG Peixu;PEI Fengque;LIU Jianhua;GUO Haoxin;ZHUANG Cunbo(School of Machinery and Vehicles,Beijing Institute of Technology,Beijing 100081;Yangtze River Delta Research Institute(Jiaxing),Beijing Institute of Technology,Jiaxing 314000;College of Electromechanical Engineering,Hehai University,Changzhou 213022)

机构地区：[1]北京理工大学机械与车辆学院,北京100081 [2]北京理工大学长三角研究院(嘉兴),嘉兴314000 [3]河海大学机电工程学院,常州213022

出　　处：《机械工程学报》2025年第4期389-402,共14页Journal of Mechanical Engineering

摘　　要：在卫星、导弹等航天复杂产品总装车间,许多装配工作由一组工人手工协作完成,且每个产品都需要经过相同的装配工序才能完成装配。由于不同工人的技能水平和装配效率不同,每道工序又有多个可选的工位,因此复杂产品装配车间调度问题可以被视为混合流水车间调度和工人分配的组合问题。为此,针对复杂产品总装生产调度问题,以最小化最大完工时间和最小化平均延迟时间作为优化目标,构建一个考虑多技能水平工人分配的多目标混合流水车间调度模型,提出一种改进的优化突变进化算法。首先,利用纳瓦兹-恩斯科-汉姆启发式算法产生初始种群以提高解的质量;其次,将非支配排序的种群增加一个突变生成的种群,以提高种群多样性。最后,基于某航天复杂产品装配车间实例,将所提算法与6种多目标优化算法(Multi-objective evolutionary algorithm,MOEA)进行了对比。结果表明,所提算法在逼近最优解的质量和算法稳定性上均优于其他6个MOEA算法。In the assembly production line of complex aerospace products such as satellites and missiles,many assembly jobs are manually completed by a group of workers collaborating together,and each product requires the same assembly process for completion.Due to variations in the skill levels and assembly efficiency among different workers,as well as multiple optional workstations for each process,the scheduling problem in the complex product assembly workshop can be regarded as a combination of hybrid flow shop scheduling and worker allocation.To address this issue,a multi-objective hybrid flow shop scheduling model considering the allocation of workers with multiple skill levels is constructed in this paper.An improved optimization algorithm called optimizing mutation evolution algorithms is proposed.Firstly,the Nawaz-Enscore-Ham heuristic algorithm is utilized to generate an initial population and enhance the quality of solutions.Secondly,a mutated population is added to the non-dominated sorted population to increase population diversity.Finally,the proposed algorithm is compared with six other multi-objective evolutionary algorithms(MOEAs)based on a case study of a complex aerospace product assembly workshop.The results demonstrate that the proposed algorithm outperforms the other six MOEAs algorithms in terms of approaching the quality of the optimal solution and algorithm stability.

关 键 词：混合流水车间调度 工人分配 复杂产品 装配车间 多目标优化算法 

分 类 号：TH165[机械工程—机械制造及自动化]