Time：16:00-17:00, March 18th, 2016

Venue：1002

Speaker：Shen Xiaobei post-doctoral,  Department of Management Sciences, City University of Hong Kong 

Host：Hua Zhongsheng 

Topic：Optimal Policies for Capacitated Multi-Echelon Inventory Systems

Summary：

In this paper, we consider a serial inventory system with capacity constraints of ordering. When the most downstream stage has the smallest capacity, we show that a multi-tier generalized modified echelon base stock (GMEBS) policy is optimal: There are multiple sets of echelon base stock levels and the echelon inventory state determines which set is optimal. In particular, the optimal echelon base stock level for each stage is nondecreasing in the echelon inventory state. We derive these results by proposing the notion of (strict) weak decomposability and subsequently showing that the value function is strictly weakly decomposable. With these results, we not only fully characterize the structure of the optimal policy of this problem, but also dramatically reduce its corresponding computational complexity. Finally, we extend our results to systems with fixed leadtimes and Markov modulated demands.

About the Speaker：

Xiaobei Shen is a postdoctoral fellow in the Department of Management Sciences at City University of Hong Kong. She received her PhD degree in Industrial Engineering from the Hong Kong University of Science and Technology in 2013. She conducts research in the areas of operations management and statistical analysis, with an emphasis on inventory management, healthcare operations, marketing-operations interface, and statistical process control.