PRODUCTION AND OPERATIONS MANAGEMENT Vol. 17, No. 6, November–December 2008, pp. 614–625 issn 1059-1478  eissn 1937-5956  08  1706  0614 POMS doi 10.3401/poms.1080.0068 © 2008 Production and Operations Management Society Value of Sharing Production Yield Information in a Serial Supply Chain Hyun-cheol Paul Choi Mihaylo College of Business and Economics, California State University, Fullerton, Fullerton, California 92834, hcpchoi@fullerton.edu James D. Blocher Operations and Decision Technologies, Kelley School of Business, Indiana University, Bloomington, Indiana 47405, dblocher@indiana.edu Srinagesh Gavirneni Johnson Graduate School of Management, Cornell University, Ithaca, New York 14853, sg337@johnson.cornell.edu N ew developments in corporate information technology such as enterprise resource planning systems have significantly increased the flow of information among members of supply chains. However, the benefits of sharing information can vary depending on the supply chain structure and its operational characteristics. Most of the existing research has studied the impact of sharing downstream information (e.g., a manufacturer sharing information with its suppliers). We evaluate the benefits of sharing upstream yield information (e.g., a supplier sharing information with the manufacturer) in a two-stage serial supply chain in which the supplier has multiple internal processes and is faced with uncertain output due to yield losses. We are interested in determining when the sharing of the supplier’s information is most beneficial to the manufacturer. After proposing an order- up-to type heuristic policy, we perform a detailed computational study and observe that this information is most beneficial when the supplier’s yield variance is high and when end-customer demand variance is low. We also find that the manufacturer’s backorder-to-holding cost ratio has little, if any, impact on the usefulness of information. Key words : inventory control; information sharing; random yield; supply chain management History : Received: August 2005; Revised: April 2007 and November 2007; Accepted: January 2008 by Jayashankar Swaminathan. 1. Introduction The recent explosion in the availability of information (due in part to the popularity of enterprise resource planning systems) is giving supply chain decision makers many opportunities for improving their sup- ply chain performance. Information “provides the decision maker the power to get ahead of the com- petition, the power to run a business smoothly and efficiently, and the power to succeed in an ever more complex environment. Information plays a key role in the management of the supply chain” (Nahmias 2001, p. 337). In the literature on this topic, the members at the supplier’s end are called upstream members and the members at the manufacturer’s (or buyer’s) end are called downstream members. A common example of sharing downstream information is when grocery retailers (downstream members) share end-customer point-of-sale (POS) demand data with their suppliers (upstream members). An example of upstream infor- mation sharing is when a supplier shares its cost and lead-time information with its customers. A large body of literature addresses how the vari- ous participants in supply chains should react to the information available to them. First, the seminal work of Clark and Scarf (1960) showed that the inventory of a multistage system can be managed optimally as long as the central planner has knowledge of the inventory levels at all stages of the system. Lee and Whang (1999) showed that even a decentralized mul- tistage supply chain can perform as well as a cen- tralized one when an appropriate transfer-payment contract is in place between the various members of the chain. In a somewhat similar context, Cachon and Zipkin (1999) considered a game theoretic approach to the problem consisting of one supplier and one retailer. They showed that the equilibrium solutions to competitive (inventory managed independently) and cooperative (inventory managed as a system) set- tings are different, but that with proper transfer pay- ments, a common solution can be found that mini- mizes system-wide costs. Lee et al. (2000) analytically quantified the bene- fits of sharing downstream information and identi- fied the drivers of the benefits. They studied a simple two-stage serial supply chain with nonstationary end 614