Group for Research in Decision Analysis

A Newsvendor Model with Unreliable Suppliers

Nicholas Petruzzi

We consider the problem of a newsvendor that is served by multiple suppliers, where any given supplier may be unreliable. By unreliable we mean that the marginal amount received from a supplier is no more than, and typically is less than, the marginal amount ordered from the supplier. In this setting, the newsvendor needs to determine (1) whether or not to place an order with a given supplier, and (2) if so, then for how much? To address these questions, we develop a general framework in which the newsvendor can diversify its risk of inadequate delivery amounts by spreading its orders among any number and combination of available suppliers that differ in terms of cost and (delivery) reliability. Ultimately, we find that the newsvendor model with unreliable suppliers has the same structural properties as a newsvendor model in which all suppliers are reliable but have limited capacity. Our resulting contribution is two-fold: First, we establish properties of the optimal solution and develop corresponding insights into the trade-off between cost and reliability. Second, we perform comparative statics on the optimal solution, with a particular emphasis on investigating how changes in supplier cost or reliability affect the newsvendor's ordering decisions and customer service level.