Quantifying the Bullwhip in Multi-location Supply Chain Network: The Impact of Demand Forecasting

This article focuses on quantification of the Bullwhip Measure (BWE)-the phenomenon in which information on demand is distorted as moving up a supply chain-for a simple two-echelon supply chain model with single distribution centre (DC) and n retailers that follow an (R, S) inventory policy, in which the Inventory Position (IP) at each retailer is reviewed each day (R = 1) and an order of a fixed batch size Q_r is placed on the DC to raise the IP to the desirable stock level, S. Assuming that the retailer employs an order-up-to level policy, and the demand forecast is performed via Moving Average model, MA (N), this paper investigates the BWE. Determining the orderup-to levels and the orders for the retailers’ demands in an isolated manner neglects the correlation of the demands, and the relevant risk-pooling effects associated with the network structure of the supply chains are disregarded. It is illustrated that the bullwhip effects are significantly reduced with consideration of potential correlation between the retailers’ demand.