Abstract: The reconfigurable assembly stations,the multi-skill workers and the multi-level assembly operations show a strong relationship when dealing with the assembly scheduling problem.In view of the lack of scheduling models that consider the above three key capabilities,a reconfigurable assembly shop scheduling model based on the hyper heuristic algorithm was proposed.The correlation of the reconfiguration switching time,regular/key process requirements and personnel/equipment correction factor were quantitative characterization.By using the hyper heuristic algorithm,the multi-scheduling rule would be fitted and the optimal scheduling scheme was iteratively optimized with evolutionary operators.The small/large batch instances and the case study of a satellite assembly workshop were drop in to verify the effectiveness and efficiency of the hyper heuristic algorithm,by reducing the mean/maximum value of the makespan,the delay time,the assembly time and reconfiguration time in the scheduling.

Key words: reconfigurable assembly shop scheduling, hyper heuristics algorithms, multi-skill, key process

摘要： 针对卫星、导弹等航天复杂产品装配车间的装配工位、装配工人、装配作业三者相互依赖与制约,给车间调度造成巨大障碍的问题,提出一种超启发式算法,创新地使用启发式拟合规则和迭代最优规则,实现了可重构装配车间调度研究。构建了考虑关键能力制约的可重构装配车间模型,以及重构切换时间、常规/关键工艺需求、人员/工站修正系数等相关的关联模型;在最佳装配工人技能、数量及其比例确定的基础上,采用超启发式算法,通过进化算子迭代寻优启发规则,实现了一种考虑工位可重构、工人多技能、常规作业/键性作业等多层次的装配车间最佳调度方案。通过小规模和大规模用例仿真验证,以及某航天企业卫星装配车间的实例应用表明,与传统启发式算法对比,所提算法具有效性和高效性。

关键词: 可重构装配车间调度, 超启发式算法, 多技能, 关键工艺