Improving Supply Chain Performance by Imposing Simple Restrictions and Using Information

• Main Authors: Zhu, Wanshan, Gavirneni, Srinagesh, Kapuscinski, Roman
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2007
• Subjects: Management Information Systems
• Online Access: https://ink.library.smu.edu.sg/lkcsb_research/612; https://doi.org/10.1080/07408170903394314
• Institution: Singapore Management University

We study a two-stage serial supply chain in which a retailer and his supplier are op- erating in a make-to-stock environment and the retailer faces uncertain demands from the end customers. When this supply chain is centrally managed, we show that the optimal policy is an extension of the classic Clark and Scarf echelon base stock policy. Recognizing that these supply chains are usually operated in an ine±cient decentral- ized environment, we analyze a simple operational change that signi¯cantly reduces the detrimental impact associated with decentralization. The strategy we analyze restricts the retailer to placing an unrestricted order only in every alternate period and requires that the retailer receives a ¯xed shipment in the other periods. We characterize the opti- mal policies and their associated costs for the non-stationary inventory control problems faced by the retailer and the supplier under these strategies. With the total supply chain cost as the primary objective, a detailed computational study shows that the described strategies improve the supply chain performance by about 11% on the average. This im- provement is a 43% (on the average) reduction in the e±ciency gap between centralized and decentralized control.