A multi-retailer decentralized distribution system with updated demand information

Abstract

In this paper, we consider a distribution system that consists of a manufacturer, a warehouse (or a distribution center), and n retailers.All retailers sell an identical product, made by the manufacturer.At the time their orders are placed, the retailers know their demand distribution but do not know the exact value of the demand.After certain production and transportation lead time elapses, the orders arrive at the warehouse.During this time, the retailers can update their demand forecasts.We first focus on cooperation among the retailers.Here, the retailers can reallocate their orders in the warehouse after their forecasts are updated, and they coordinate their initial orders accordingly.We investigate the associated cooperative game between the retailers and show that there exists a stable allocation of joint profit among the retailers (i.e., the core of the associated cooperative game is non-empty).Next, we concentrate on the relationship between the manufacturer and the retailers.We consider two contracting schemes -namely, the wholesale-price contract and the buy-back contract.Our buy-back contract differs slightly from the traditional one, in that we assume that the retailers return their extra or unwanted items from the warehouse (after forecast update) and not from their facilities (after demand realization).We analyze three cases (non-cooperating retailers, cooperating retailers, and manufacturer's resale of returned items), and we investigate the impact that each of these models has on the manufacturer's profit.Finally, we focus on coordination of the entire supply chain through buy-back contracts, which are known to coordinate the systems with full information.We show that, in our setting, buy-back contracts, in general, cannot coordinate the distribution system.Further, we illustrate through an example that the gap between the centralized solution and the decentralized output can be arbitrarily large.