考虑双目标的JIT生产系统看板数量设定

摘要

The past literature of Kanban number decision mainly focused on calculation under single objec-tive optimization for JIT production system. In this research, a calculation method for Kanban number un-der bi-objective optimization is proposed based on the design of experiment ( DOE ) and Witness simula-tion. The method considers both maximum order fill rate and minimum WIP inventory level. Taking the case of CR nozzle′s JIT production system of B Company, its witness simulation model is constructed to provide data for the design of experiment. The levels of the orthogonal experiment are set up based on the Kanban size and number of each loop of the CR nozzle′s JIT product system. Then, a full factorial experi-ment design is used to get nine groups of Kanban number Pareto-optimal solutions under the performance measurement of order fill rate and WIP inventory level. The Kanban numbers satisfying Pareto-optimal so-lutions form the Pareto front of optimal Kanban number combinations, from which B Company′s production planners can select appropriate Kanban number according to different production demand. The experiment results demonstrate the validity and the feasibility of the method.%针对现有JIT系统看板数量决策问题研究多以单目标为主的不足,提出了一种基于实验设计的双目标JIT生产系统看板数量设定方法.该方法同时考虑了高订单满足率和低系统平均在制品水平的双目标优化,以B公司CR油嘴JIT生产系统为例,建立了该JIT生产线的Witness仿真模型以实现数据的收集,以各看板循环回路的看板数量和看板容量进行水平设定,并进行正交实验设计及数据的直观分析处理,然后采用全因子实验的方法,基于帕累托最优的思想获得生产系统看板数量帕累托最优解,形成近似最优看板数量组合的帕累托最优前沿.生产管理人员可根据不同的生产计划和绩效目标从组合中选择合适的看板数量.最后的研究结果验证了该方法的可行性和有效性.