Matching supply with demand in decentralized multi-location systems.

Abstract

Efficient information systems and modern logistics networks facilitate the interactions among firms and endow them with enhanced capability to respond to actual demands. Many phenomena have emerged in real business settings because of the newly available advanced infrastructures. In this dissertation, we study several of them with the focus on how to better match supply with demand in decentralized multi-location systems. A responsive distribution system naturally exhibits a structure with locations close to customers and operating in a decentralized manner. Because of finite stock level and uncertain demand, a potential problem is that some locations may be understocked while others overstocked. An effective way to rebalance the network is through search for stock availability to reallocate unmet demands among locations with excess stocks. In Chapter I, we analyze two methods of search initiated respectively by customers and retailers and supported by certain information systems. Using a system of one manufacturer and two retailers, we characterize joint stock and price decisions of retailers, analyze the impact of the two methods on channel behavior, and discuss the preferences of different parties between them. Independent online agent, sustained by Internet and reliable online transactions, provides another channel for customers to search and purchase products. It functions as an outlet for products from a number of firms. Firms that sell through the agent compete and are required to respond to actual demand. Chapter II is devoted to a general duopoly where firms have responsive capabilities and face uncertain demand. We analyze the resource investment, stock allocation, and price decision to evaluate the performance of firms and the impact of various market assumptions. In Chapter III, we analyze the impact of online agent on channel structure. The online agent offers contract to firms which then make enrollment decisions that determine channel structure. We derive optimal terms for the contracts of interest, analyze optimal contract design, and conduct comparison of prices offered in different channels. With the results, we provide an alternative explanation for the phenomena from empirical studies on price comparisons. This dissertation is concluded with a summary and discussion of future research topics.