动态定价与存货控制: 理论模型与实证分析

Our proposal is motivated by the innovative supply chain management practics of inventory-based pricing in the last two decades. The first topic of the proposed research is to extend the classical EOQ model to the case where the procurement cost varies over time. We hope to find the optimal economic order quantity for the simple case, the approximate optimal order quantity and the structure result for more general case. The result would be very useful for the upstream firm planning for its raw materials and production capacity. The second model considers a well-studied joint dynamic pricing and inventory control problem where the demand depends on price and varies with time. The existing literature considers two cases: lost sale case where unmet demand is lost, and backlogged case where unmet demand is backlogged to the next period. While the lost sale case is more realistic (at least for most applications), it is analytically more difficult. Building upon our recent work on the backlogged case, we propose to extend it to the lost sale case and establish more realistic condition for the optimality of a structured (which is known as (s,S,p)) control policy. The third area of our proposed research relates to the marketing literature on the new product diffusion. In this literature, the current demand (or the diffusion rate) depends on the cumulative past sales (in addition to price). The proposed research extends this literature by considering joint pricing and inventory decision, as well as considering a time-varying replenishment cost. This research complements the previous research on dynamic pricing and inventory control where the demands for different periods are assumed to be independent. We obtain the optimal dynamic pricing and inventory control policy that maximizes the total discounted profit and study how the product diffusion process and the optimal price and replenishment vary over the product lifecycle. Finally we propose to study empirically how the price, the inventory and the demand interact dynamically with industry data. This would extend the empirical literature in marketing which usually does not include the inventory data and the very limited empirical literature in operations which usually does not include the price data. Through the empirical analysis, we expect to estimate the model parameters and validate the predicted results in our theoretic models, and study how price, inventory and sales interact. The completion of the proposed research will provide a solid foundation for modeling and analysis of an inventory system with price- and time-sensitive demand and time-sensitive procurement cost. This extends the existing operations management literature on the pricing and inventory control of an inventory system with stationary environment, and complements the marketing literature on the new product diffusion, where the operations issue such as inventory control has not been considered.

所提议的研究课题关注动态定价与库存控制，包括理论模型分析及实证研究。首先，我们考虑一个包含时变采购成本的经济批量模型。相应采购成本时间函数的复杂性，我们将找最优订购数量的明确公式、近似公式、及结构属性。这将有助于指导上游准备产能或材料采购。第二项研究针对库存模型的定价和存货控制，将把有关积压模型的现有研究结果延伸至销量流失模型。这项延伸令模型假设更具现实性。另一个重要课题是针对新产品扩散的动态定价与库存控制。其特点是产品的当前销量依赖于价格及以往累计销量。这将对现有的产品扩散流程经典文献加以延伸和补充，因为后者并没有考虑生产成本及存货因素。最后一个研究课题是实证研究。包括两个方面：一是对以上的理论模型进行实证验证。二是研究价格、存货和销量之间的相互作用与依赖关系。这项研究将验证和支持定价和库存控制文献中的分析成果；更为重要的是，它将为未来相关模型假设和模型参数的研究提供适宜的基础和指导。

本研究课题主要是关注动态定价与库存控制和供应链竞争。首先、针对库存模型的定价和存货控制，我们将现有研究结果延伸至销量流失模型。这项延伸令模型假设更.具现实性。第二，我们研究了供应链中在需求不确定的情况下链条对链条的竞争。我们特别考虑了公司的多种不同战略定位：两个整合的供应链、两个分散的供应链、及混合供应链（一个整合、一个分散）。我们比较不同供应链结构和在平衡情况下对供应链的选择；这些选择取决与产品的特性（可替换产品或互补产品）。在另外一个研究中、我们检查了一些高知名的“工厂物理原理”。这些原理是关于变化性、利用率、返工、提前期、及固定在制品品的效率。我们指出这些“原理”是有边界的。我们给出一些反例、同时给出一些充分条件。在本课题的资助下，我还开始了金融领域的研究。特别是关注2007-2008年金融风暴后美联储直接投资的策略。我们检查那些策略可以达到社会效率。