In this paper we present a two-echelon inventory system consisting of a warehouse that does not hold stock, and multiple retailers. The demand at the retailers is stationary and demand per period is normally distributed. The retailers forecast demand using single exponential smoothing, leading to dynamic demand forecasts. Thus the retailers use dynamic inventory control policies aimed at achieving a target service level. The dynamic policies are based on covering the forecasted demand over the lead time plus some safety time. The safety time is the control parameter to be determined, such that the target service level is met. This model allows us to study the impact of demand information distortion due to forecasting on the safety stock requirements of the retailers. We derive explicit expressions for the control policies and the safety time parameters. We introduce a new amplification factor ? that measures the relative increase in safety time requirements under demand information distortion due to the exponential smoothing with parameter ?. We derive quantitative and qualitative insights into the impact of the model parameters on the safety time amplification factor ?. We show that under demand distortion it may be favorable to reduce the review frequency.
|Number of pages||12|
|Journal||Review of Business and Economic Literature (ReBEL)|
|Publication status||Published - 2012|