摘要
针对航天结构件生产调度过程中工序加工时间随操作人员能力级别而变化,关重件和关键工序特定设备和人员要求约束等问题,构建综合考虑关键设备和人员双资源约束的柔性车间作业调度数学模型,解决航天结构件生产设备资源维、人员资源维、工序信息维的描述。为实现双资源约束下的柔性调度,提出嵌套式蚁群-遗传混合算法,针对结构件加工工序设备和人员双资源选择问题,建立资源选择问题与蚁群并行搜索的映射关系,设计蚁群算法实现资源选择;为提高算法搜索性能,利用遗传算法求解当前蚂蚁资源选择下的调度问题,在避免人员资源使用冲突前提下实现设备加工工序的优化排序,使得关键设备利用率最大,总完工时间最短,并通过综合考虑蚁群和遗传算法阶段的求解目标进行蚁群算法信息素的更新,大大提高算法获得较优解的性能。最后,通过航天结构件车间实际案例进行算法测试,测试结果表明,嵌套式蚁群-遗传混合算法能有效地求解双资源约束的航天结构件车间生产调度问题,可降低资源总负荷,提高关键设备利用率,避免人员资源冲突,具有良好的综合调度性能。
The production scheduling method of aerospace structural parts with dual resource constraints of critical equipment and operator is investigated, where structural parts’ processing time varies with the operator’s ability, and the key components and processes have specific equipment and operator requirements constraints. A mathematical model of the flexible job shop scheduling with dual resource constraints(FJSSDR) is constructed to describe the equipment、operator and process information of the parts. A nested ant colony-genetic hybrid algorithm(NACGHA) is designed to solve the FJSSDR problem. Aiming at the dual resource selection problem, the mapping relationship between resource selection problem and ant colony parallel search is structured. Genetic algorithm is used to solve the optimal scheduling problem by taking the selected double resource constraints as the processing constraint information of the job. The equipment processing sequence is optimized with avoiding conflict of key operator resources, where the objectives of fullest utilization of critical equipment and makespan minimization are considered. The ant colony algorithm pheromones are updated by integrating the objective value obtained by ant colony and genetic algorithm to improve the performance. At last, a scheduling case of aerospace structural parts production shop is analyzed. The case study demonstrates that the proposed NACGHA method has good performance on FJSSDR in terms of reducing resource total load, avoiding resource conflict and improving the utilization of key equipment.
出处
《机械工程学报》
EI
CAS
CSCD
北大核心
2018年第9期55-63,共9页
Journal of Mechanical Engineering
基金
国家自然科学基金资助项目(U1637211)
关键词
航天结构件生产
柔性调度
双资源约束
嵌套式蚁群-遗传混合算法
production of aerospace structural parts
flexible scheduling
dual resource constraints
nested ant colony-genetic hybrid algorithm
