The Value of Demand Sharing in Supply Chains Facing ARMA Demand

• Activity Number: 502
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Vladimir Kovtun*+ and Avi Giloni and Clifford Hurvich
• Companies: Yeshiva University and Yeshiva University and NYU Stern School of Business
• Keywords: supply chains ; ARMA ; information sharing ; seasonality ; forecasting demand

Abstract

We develop several novel results on the topic of information sharing in K-stage supply chains where the most downstream player (the retailer) observes demand that could be described by an autoregressive moving average process (ARMA). Our first result provides the value of a demand sharing arrangement between two adjacent players. We demonstrate a potential reduction in an upstream player's mean square forecast error (MSFE) of its best linear forecast when its downstream player shares its demand information. In doing so we provide a framework for comparing demand sharing with other types of sharing arrangements. This also yields conditions under which demand sharing is valuable.

We proceed to consider a two-player supply chain to study whether demand sharing is valuable to the upstream player (supplier) when the retailer observes demand that is generated by a process in the class of seasonal ARMA models or second-order autoregressive models. We find that demand sharing is valuable to the supplier only in the presence of strong seasonality in the retailer's demand process and when the retailer and supplier observe different peak and low seasons.