随机需求下的多阶段供应链库存问题研究

多阶段库存问题是供应链研究的重要课题，随着供应链思想的发展，控制各个存储环节的成本是整个供应链降低成本及规避风险的有效方法，对这一问题的解决方法主要有2种：(1)在分散决策下应用单阶段方法依次求优；(2)销售需求计划(类似MRP)。这两种方法适用于不同的企业环境，本文采用了第一种方法。产品需求数量的不确定性是多阶段库存和分销系统难以实施优化的主要原因之一．本文研究的是一个批发商，Ⅳ个零售商的销售系统，批发商对零售商的订单采取的是先到先服务机制。通过将各个水平的需求依据其概率特性进行随机的描述，然后运用单阶段的方法对其优化，进而得出各个阶段的最优库存。

Study on Inventory with Stochastic Demand in Multi-echelon Supply Chain

The problem of multiechelon inventory is an important task in supply chain study. With the development of supply chain thinking, The efficiency way to reduce the inventory cost and risks is to control the inventory cost on different level. Recently, there are two approaches: (1) apply the singleechelon approach to each echelon in the network.(2) use a distribution requirements planning (DRP) approach or a variation. These two approaches hold true in different environment of enterprise.The uncertainty of product demand is one of the most important causes for the difficult of optimizing the inventory system. This paper assumes a system with a wholesaler and Nretailers, the wholesaler serves the retailer with the firstcome, firstserved mechanism. Fist we describe demand on different level as stochastic variable according to their probability property. Then we can get the optimal inventory by using singleechelon inventory method.