【作者】 丁正平；

【导师】 汪克夷；

【作者基本信息】 大连理工大学 ， 企业管理， 2010， 博士

【摘要】 库存控制的有效性对企业运作效率具有非常重要的影响。单周期产品需求具有期限短、变化快、不确定因素多的特点,库存控制难度较大。随着供应链的发展,协调是供应链管理的主要方式。在供应链环境下,单周期产品的生产销售由多个企业完成的,库存协调成为运作成功的关键。由于销售期短、变化快,有限需求信息是单周期产品经营的常态。如何在有限需求信息下解决单周期产品的库存协调问题是很多企业迫切需要解决的问题。本文以有限需求信息下的单周期产品库存协调问题为研究对象。通过对单周期产品库存控制中不确定性的研究,探讨运用信息熵度量不确定性,并运用极大熵准则预测有限需求信息下的需求分布。为解决多层系统下单周期产品的库存协调问题,在研究库存协调策略的基础上,探讨了有限需求信息下的库存协调问题,提出了基于决策熵权的库存协调策略。在研究中,本文主要解决了以下几个方面的问题：(1)运用极大熵准则对需求分布进行预测。需求分布预测是有限需求信息下单周期产品库存控制的基础。极大熵准则是在有限需求信息下预测需求分布的有效方法。在已知需求均值和方差的情况下,运用极大熵准则可以确定所有可能分布中的最可能出现的分布(最可几分布)。通过与Scarf订货规则、正态分布假设的对比分析中可以看出,该预测方法有利于提高单周期产品的期望利润。(2)建立基于熵权的数量折扣机制。通过对由零售商和供应商组成的二层供应链的研究,发现独立决策时的结构缺陷,探讨关于批发价格的竞争,并在定义决策熵权的基础上,提出了基于决策熵权的数量折扣机制。该机制以联合决策为基础,从而使系统利益最大化,增加了协调机制的有效性；同时,该机制既考虑了协调前双方的可能收益,又考虑了双方的市场竞争能力,提高了协调机制的可行性。(3)研究不对称需求信息下的库存协调问题。多数情况下,供应链中关于需求方面的信息是不对称的。通常,零售商由于更靠近市场,拥有更多更准确的需求信息。本文通过研究不对称信息下的两阶段决策问题,提出了基于批发价格决策权有偿转让的思想,并运用决策熵权来协调利润分配。该机制既体现了各自的市场竞争地位,又体现了需求信息不对称性。(4)研究弹性需求下的库存协调问题。在弹性需求下,零售价格成为决策变量。通过对有限需求信息下需求量随零售价格波动方式的研究,运用极大熵准则推导了需求分布随零售价格的变化关系。在研究弹性需求下批发价格和零售价格之间联动关系的基础上,提出了基于决策熵权的协调机制。(5)研究基于退货策略的库存协调问题。在允许退货的条件下,退货价格也成为决策变量。在研究期末处理价值和惩罚成本对库存决策影响的基础上,探讨了退货策略对库存控制的影响,研究了基于决策熵权的退货价格决策方法。本文研究是信息论相关理论在库存控制中应用的一种尝试。在拓展信息论相关理论应用范围的同时,解决了有限需求信息下单周期产品的库存协调问题。更多还原

【Abstract】 The validity of the inventory control is very important to improve operational efficiency in an enterprise. The single-period products have some characteristics:the short sales period, quickly changes of the demand, and many uncertain factors to the inventory control. So it is very difficult to control the inventory of the single-period products.With the development of the supply chain, coordination is the main method in supply chain management. In a supply chain, the production and sales of the single-period products are implemented by many enterprises, and the inventory coordination is a key factor to successful operations. It is normally that the enterprises have partial demand information about the single-period products due to the short sales period and quickly demand changes of the single-period products. It is one kind of urgent problem for many enterprises to deal with the inventory coordination problem of the single-period products under partial information.The paper studies the inventory coordination of the single-period products under partial demand information. Based on the research on the uncertainty in the inventory control of the single-period products, the information entropy is employed to measure it, and the maximum entropy formulism is applied to forecast the demand distribution under partial information. In order to solve the inventory coordination in the system with more than one enterprise, the paper studies the inventory coordination under partial demand information, and puts forward the inventory coordination schemes based on the idea of decision-making entropy weight. Some problems have been solved in the study:First, the maximum entropy formulism is applied to forecast the demand distribution. The forecast of the demand distribution is the foundation of the inventory control of the single-period products under partial demand information. The maximum entropy formulism is an effective method to discover the demand distribution. Under partial demand information, where only the mean and the variance of the demand are known, the maximum entropy formulism can find out the most probable distribution among all possible distribution. The result of comparing the mothod with Scarf’s ordering rule and the assumption of normal distribution shows that it can increase the expect profit of single-period products.Second, a quantity discount scheme based on the entropy weight is established. In a supply chain with a retailer and a supplier, there is a structural limitation under independent decision. By discussing the competition of the retailer and the supplier in the wholesale price, it puts forward a quantity discount scheme based on the entropy weight. Based on the integrated decision, the scheme is an effective method to optimize the whole profit. The scheme is feasible because it considers both the possible profit of the two parties before coordination and the marketing competition of them.Third, it studies the problem of inventory coordination under asymmetric demand information. Usually, the demand information is asymmetric in the supply chain. Being more close to the retail marketing, the retailer has better knowledge of the demand. It studies the two stage decision-making problem under asymmetric demand information, and puts forward the idea of compensated transfer of the decision-making power of the wholesale price from the supplier to the retailer. The profit distribution is also studied by using decision-making entropy weight. The method considers both the marketing competition and the asymmetry of demand information.Fourth, the problem of inventory coordination under the elastic demand is studied. Under the elastic demand, the retail price is a decision variable. By studing the changes of the demand with the retail price under partial demand information, the maximum entropy formulism is applied to deduce the demand distribution at different retail price. After investigating the relationship between the optimal retail price and the optimal wholesale price, the coordination scheme based on the entropy weight is established.Fifth, it studies the inventory coordination method of return policy. When return policy is applied, buyback price is also a decision variable. After discussing the affection of the salvage value and the shortage cost, it studies the affection of return policy to the inventory control, and the decision-making method of return credit based on the entropy weight.The paper is an attempt to apply the information theory to the inventory control. It solves the inventory coordination problem of the single-period products under partial demand information, and also extends the application range of the information theory.更多还原