Coordinating supply chain in decentralized environments: Optimization, auction, and bargaining-theoretic models

In this dissertation, we tackle three main topics in supply chain coordination as described below.; Integrating production and transportation logistics. We consider the integration of production and transportation logistics at the operational level. Our method allows the optimization model to reconcile the viewpoints from transportation and production planning. We introduce basic production and transportation models that are tailored to this particular integration and show the value of the integration using a Lagrangean decomposition scheme.; An auction theoretic model for production coordination. Motivated by the increasing needs to coordinate diverse decision processes and systems, we investigate an auction-theoretic mechanism for production coordination in a supply chain. Our efforts focuses on an iterative auction mechanism wherein agents compete based on their local utilities, announced conflict pricing, and production targets. Building upon the rich literature in optimization and auction, we investigate the advantages and limitations of this decision scheme on a large set problems in supply chain planning. Experimental results show that the proposed auction provides impressive improvement over traditional monolithic method without significant degradation to the solution quality.; A bargaining model for supplier coordination. We first propose a one-buyer one-supplier non-cooperative bargaining game for supply chain contracting, where the buyer negotiates with a sourcing supplier the order quantity and wholesale price. We show that in subgame perfect equilibrium, the channel coordinated optimal quantity is also optimal for the players, but the players must negotiate the surplus generated by the contract in a bargaining game. The model allows us to predict the negotiation outcome between a pair of buyer and supplier considering their outside options, the breakdown probability, and random proposers. Motivated by emerging applications in electronic marketplaces, we the propose a one-buyer multiple-supplier negotiation sequencing problem where the buyer could determine an optimal subset of sourcing suppliers to negotiate with, and the sequence to carry out the negotiations so as to maximize his expected gain. We show that the one-supplier version bargaining game serves as a building block and the negotiation sequencing problem can be solved as a network flow problem.