基于随机动态规划的有限库存ATO系统优化控制

发布时间：2018-06-03 15:42

本文选题：按订单装配(ATO) + 优化控制　；参考：《运筹与管理》2017年07期

【摘要】：本文研究n维组件单一产品,有限库存的ATO系统。通过建立马尔可夫决策过程模型(MDP),构造优化算法,研究组件生产与库存的最优控制策略。最优策路可以表示为状态依赖型库存阈值,系统内任一组件的控制策略受其它组件库存状态的影响。利用最优控制理论动态规划方法和数值计算方法对最优控制策略的存在性、最优值的数值计算进行研究,建立更符合实际生产的ATO系统决策模型,进行相应的理论和实验验证,研究系统参数对最优策略的影响。
[Abstract]:In this paper, we study the ATO system of n-dimensional component with single product and finite inventory. By establishing the Markov decision process model (MDP) and constructing the optimization algorithm, the optimal control strategy of component production and inventory is studied. The optimal strategy can be expressed as a state-dependent inventory threshold, and the control strategy of any component in the system is affected by the inventory state of other components. The existence of optimal control strategy and the numerical calculation of optimal value are studied by using the dynamic programming method and numerical calculation method of optimal control theory. The decision model of ATO system is established, and the corresponding theoretical and experimental results are verified. The influence of system parameters on optimal strategy is studied.
【作者单位】：天津工业大学管理学院;西南交通大学经济管理学院;
【基金】：国家自然科学基金资助项目(71602143) 教育部第49批留学回国人员科研启动基金项目(教外司留[2015]311号)
【分类号】：F273

【相似文献】