Abstract
In this paper we study how the time-series structure of the demand process affects the value of information sharing in a supply chain. We consider a two-stage supply chain model in which a retailer serves auto-regressive moving-average (ARMA) demand and a manufacturer fills the retailer's orders. We characterize three types of situations based on the parameters of the demand process: (i) the manufacturer benefits from inferring demand information from the retailer's orders; (ii) the manufacturer cannot infer demand, but benefits from sharing demand information; and (iii) the manufacturer is better off neither inferring nor sharing, but instead uses only the most recent orders in its production planning. Using the example of ARMA(1,1) demand, we find that sharing or inferring retail demand leads to a 16.0% average reduction in the manufacturer's safety-stock requirement in cases (i) and (ii), but leads to an increase in the manufacturer's safety-stock requirement in (iii). Our results apply not only to two-stage but also to multistage supply chains.
Original language | English (US) |
---|---|
Pages (from-to) | 961-969 |
Number of pages | 9 |
Journal | Management Science |
Volume | 51 |
Issue number | 6 |
DOIs | |
State | Published - Jun 2005 |
Externally published | Yes |
Keywords
- Electronic data interchange
- Information sharing
- Nonstationary demand
- Single-item inventory model
- Supply chain management
ASJC Scopus subject areas
- Strategy and Management
- Management Science and Operations Research