College of Business
AACSB International EQUIS - European Quality Improvement System

Research Seminar

Seminar: Optimal Capacity Under Stock-Out Probability Constraints: The Value of Responsive Allocation

Abstract: In many service systems, different customers may require different service levels defined by the maximum stock-out probability constraints. A key challenge in such systems is to find out the minimal capacity needed and how to allocate this limited capacity to customers in order to meet each customer’s service level requirement. In this talk, we will address these open questions. In particular, using a new approach motivated by Blackwell's Approachability Theorem, we obtain the minimal capacity needed and the corresponding optimal allocation policy. Our results hold for any demand distributions, and thus are appealing in practice. We show that, with a sufficiently large number of customers, our solution reaps significant savings from the case of fixed priority rule in which there is a need to maintain a safety stock that grows in the square root of the number of customers. Finally, our numerical studies demonstrate that moving from anticipative allocation policies to responsive ones can reduce the required capacity significantly in such systems. Joint work with Guodong Lyu and Chung-Piaw Teo (NUS), Daniel Zhichao Zheng (SMU), and Yuanguang Zhong - South China University of Technology.
Date: 24 March 2017 (Friday)
Time: 11:00AM - 12:30PM
Speaker: Dr Mabel Cheng-Feng Chou
Venue: Room 7-207, 7/F, Academic 3

[ Back ]