Inventory Sharing and Rationing in decentralized dealer networks

Presenter: Hui Zhao
zhaoh@ecn.purdue.edu
Co-authors: Jennifer Ryan, Vinayak Deshpande
Purdue University


We analyze an inventory sharing and rationing problem in a decentralized dealer network, i.e., a network in which the dealers are independent of the manufacturer, in a continuous-review, infinite-horizon setting. The three key features of our model are: inventory sharing among dealers, multiple demand classes for each dealer, and decentralized decision making. We first use dynamic programming to characterize the structure of the optimal inventory sharing and rationing policy for an individual dealer, taking the other dealers' decisions as exogenous. We then consider a decentralized, two-dealer setting to study the impact of inventory sharing and rationing among independent dealers, considering the interactions between the dealers' decisions. A game theoretic approach is used to characterize the equilibrium behavior of the individual dealers. An extensive numerical study highlights the role of manufacturer incentives and penalties in a decentralized setting.