| Today's firms face unprecedented challenges due to increasing production variability, upstream supply disruptions, demand volatility, and recent surges in the transportation costs. Transshipments among firms supplying substitutable products can mitigate the impacts of uncertainty and enhance the overall supply chain efficiency. High-volume, two-way systematic transshipments, known as exchanges, are common in gas and chemical industries. In this work, we consider a two location system where systematic exchange transactions are controlled by a central agent. We model the system under consideration first as a finite-horizon expected discounted total cost dynamic program and then as an infinite-horizon expected discounted total cost dynamic program and characterize the optimal production/inventory and the exchange policies for such systems. We also extend this model to incorporate Markov-modulated demand where demand can be positively or negatively correlated. Using extensive numerical analysis, we present the sensitivity of our results to different factors such as capacity allocation, demand allocation, exchange transaction costs, exchange inventory holding costs and the demand volatility and provide managerial insights on the desirability of exchange transactions. Additionally, we introduce the generalized framework for transshipment systems, which accounts for traditional transshipment related costs, exchange transactions related costs and fixed costs for initiating transshipments. Finally, we summarize our findings and present several future research directions that can follow this work.;Keywords: Exchange transactions, transshipments, inventory management, make-to-stock system. Markov Decision Process. |
