订单不确定性到达的混合流水车间调度

Hybrid Flow-shop Scheduling with Uncertainty of Order Arrival

摘要:

为了解决实际作业车间调度中由新订单中途到达插入而打乱原有计划的问题，在混合流水车间调度问题的基础上考虑订单到达时间与数量的不确定性，以质量鲁棒性和解鲁棒性为优化目标建立数学模型，并针对该模型设计具有双层决策结构的前摄性调度算法。算法外层运用改进的遗传算法框架搜索关键决策变量编码，内层通过分析订单到达不确定性形成的不同场景，制定带有缓冲时间设置的启发式规则对外层染色体进行解码。结合蒙特卡洛抽样和反应策略进行仿真评估，与Gurobi和多种现有模板计划生成机制进行对比，验证所提出的考虑订单不确定性到达的前摄性调度框架与算法的有效性。

Abstract:

To address the issue of disruptions in original schedules caused by the mid-arrival and insertion of new orders in practical workshop scheduling， the uncertainty of order arrival time and quantity was considered based on the hybrid flow shop scheduling problem. A mathematical model was established with the optimization objectives of quality robustness and solution robustness. A proactive scheduling algorithm with a two-layer decision structure was designed for this model. The outer layer of the algorithm used an improved genetic algorithm framework to search for key decision variable encodings， while the inner layer formulated heuristic rules with buffer time settings to decode the outer layer chromosomes by analyzing different scenarios formed by the uncertainty of order arrival. Simulation evaluation was conducted using Monte Carlo sampling and reaction strategies， and effectiveness was verified by comparison with Gurobi and various existing template schedule generation mechanisms.