Game theoretic analysis for supply chain coordination: Modeling electronic intermediation, coalition formation, and capacity reservation

We examine supply chain coordination mechanisms that attempt to improve supply channel efficiency. We focus our attention on game theoretic models that could generate insights of significant practical importance. The dissertation is a collection of three research papers. The first paper focuses on supply chain contracting in electronic marketplaces. The current supply chain literature considers the one-supplier, one-buyer system as the basic building block. In emerging electronic market transactions, situations often arise where it is more appropriate to model the system as multiple suppliers and one buyer. Moreover, it is necessary to consider the role of electronic intermediaries who implement the coordination mechanisms. While auction often serves as a price-determination mechanism in this environment, we show that it could also serve as coordination mechanisms in the supply chain. The second paper addresses supplier coalition in an auction market. We propose a profit distribution scheme that allows suppliers to form coalitions with one another for the purpose of enhancing profitability. We identify basic requirements for a valid coalition mechanism including characteristics such as individual rationality, society welfare compatibility, maintaining competition, and financial balancedness. We verify that the proposed profit distribution mechanism satisfies all validity requirements and will increase the market efficiency as a whole. The third paper examines supply chain capacity coordination via reservation contracts. We consider various capacity reservation contracts in a high-tech device market. We show that various reservation contracts can be implemented by the manufacturer to share the risk of capacity expansion with their main customers. We show that these contracts not only improve the profitability of the manufacturer, they also improve the customer's profitability, and the system's overall performance. We consider both one-manufacturer, one-customer systems, and multiple-customer cases.