项目名称: 产品回流依赖于历史需求再制造系统的库存控制
项目编号: No.71301093
项目类型: 青年科学基金项目
立项/批准年度: 2014
项目学科: 管理科学
项目作者: 陶之杰
作者单位: 上海财经大学
项目金额: 20.5万元
中文摘要: 我们研究一个单产品多阶段库存系统。用于满足客户需求的新产品既可以从原材料制造生产也可以通过再制造回流的产品。我们假设产品的回流依赖于历史需求。由于这种相关性,传统的动态规划方法不太适合解决这类问题。于是我们提出了一种启发式方法用于管理这类系统。我们可以证明在这种启发式方法下,系统的总成本在最差的情形下也不会超过最优成本的两倍。与其形成对的是短视策略下的系统总成本可能会是最优成本的任意大倍。另外,很多实证研究指出产品回流与历史需求的相关性通常可以由一个自回归分布延时模型所刻画。对于这样的库存系统,我们提出了一个根据预测调整的基库存策略。在交通繁忙近似之下我们可以得到最优基库存水平的显性表达式。根据预测调整的基库存策略的优点在于非常易于实施。同时我们也将比较该策略下与传统的基库存策略下系统成本以及库存量变化的不同。
中文关键词: 产品回流;再制造;库存控制;近似方法;
英文摘要: We consider a single-product periodic-review inventory system with remanufacturable returned products, in which the serviceable product used to fulfill stochastic customer demand can be either manufactured from raw materials or remanufactured from the returned products. We assume the returns are dependent on the historical demands. Due to this correlation, traditional dynamic program framework is not suitable for the problem. We propose a new heuristic policy for managing such a system. We can show that this new policy will incur at most the twice of the cost as the optimal one, while in contrast, the myopic policy can be arbitrarily bad. Empirical study show that the correlation between product returns and demand can be in general characterized by a autoregressive distributed lag model. For such a system, we propose a forecast-adjusted policy and show that the optimal base stock level can be found in closed-form when the system goes to heavy traffic. The advantage of the forecast-adjusted policy is that it is very easy to implement. We also compare the system cost as well as the inventory level behavior under such policy and the traditional base stock policy.
英文关键词: Product Return;Remanufacturing;Inventory Control;Approximation Method;