Capacitated Production Control with Virtual Lateral Transshipments

Abstract

We study the optimal control of a firm with two capacitated manufacturing plants situated in two distinct geographical regions. Demands from each region are mostly satisfied by the local plant. However, if necessary, some of the newly arrived demands can be designated to be served by the other, more remote, plant. The sources of the above virtual lateral transshipments, unlike the ones involved in the real lateral transshipments, do not need to have nonnegative inventory levels throughout the transshipment processes. We develop results about the simultaneous preservation of supermodularity and diagonal dominances of the cost functions and then use these results to derive structural results for the optimal policies. Specifically, to perform optimally, we find that each plant should adopt a modified base-stock policy in observance of its own capacity, while the base level should mildly decrease in the other plant's starting inventory level. Our computational results illustrate the benefits realizable by using transshipment.

Making Better Fulfillment Decisions on the Fly in an Online Retail Environment

“…Axsater (2003), for example, develops a decision rule dictating whether to transship or not, or whether to incur the backorder costs. Yang and Qin (2007) discuss a model that utilizes virtual lateral transshipments between two factories. This is similar to online retailing in that inventory need not travel from FC A to B, then to the customer to be considered a transshipment, but instead may be shipped directly from A to the customer in region B at increased cost.…”

Retail supply chain management: a review of theories and practices

“…Axsäter (2003) develops a decision rule dictating whether to transship, or whether to incur the back-order costs. Yang and Qin (2007) discuss a model that utilizes virtual lateral transshipment between two factories. Archibald et al (2009) develop a transshipment heuristic for a realistic multi-location inventory system.…”

Infinite horizon optimal policy for an inventory system with two types of product sharing common hardware platforms

“…With regards to lateral transshipment, our study of the literature reveals that Hu, Duenyas and Kapuscinski [10], and Yang and Qin [19] are two recent papers that study periodic review, infinite horizon transshipment model for discounted cost/profit by stochas-tic dynamic programming. Yang and Qin [19] provide detailed analysis on how they derive the optimal policy, while Hu, Duenyas and Kapuscinski [10] consider the infinite horizon scenario as an extension to the main results in their paper. Also, we note that there is a brief mention in Ignall and Veinott [13] on how the finite horizon results in Ignall and Veinott [13] can be extended to infinite horizon.…”