考虑工序并行的差异工件批调度研究

A Research on Batch Scheduling with Non-identical Job Sizes Considering Parallel Processes

针对铸造车间差异工件组批多约束的问题,在工序可并行加工的前提下构建以最小化最大完工时间和最小化沙箱空置率为优化目标的并行工序批调度模型,设计一种改进和声算法求解该调度模型,提出一种单工序编解码方式和2种机器分配规则用于解决工件分批、沙箱选择、工序分配及机器选择的问题。在算法中提出一种新的和声产生方式和更新机制,同时为改善算法的局部搜索能力,加入模拟退火算法执行局部搜索过程。最后根据企业实际生产数据进行仿真实验,验证本文模型的有效性。

Considering that the batch process is multi-constrained with non-identical job sizes in foundry enterprises,a parallel process batch scheduling model is constructed,with the maximum makespan and sandbox vacancy rate as the optimization objective on the premise that the job can be parallel,an improved harmony search algorithm designed to solve the scheduling model,and a single process code method and two machine allocation rules proposed to solve the problems of batch division,sandbox selection,process allocation and machine selection.In the algorithm,a new harmony generation method is proposed.At the same time,the simulated annealing algorithm is added to make it jump out of the local optimal solution and tend to the global optimal solution.Finally,a simulation experiment is carried out based on the actual production data of the enterprise to prove the effectiveness of the proposed model.