Impacts of demand forecasts and supply chain coordination on bullwhip effect

Abstrcat In this paper, the bullwhip effect in a multistage supply chain is analyzed using self-adaptive filtering, moving average and exponential smoothing demand forecast technniques. We find out that the bullwhip effect does not always exist when the self-adaptive filtering forecasting is used, and present the existence condition. However, the bullwhip effect always exists when the moving average and the exponential smoothing forecasting are used. Based on these three forecasting tools, it is concluded that suitable forecast parameters and supply chain coordination will alleviate the bullwhip effect. In addition, we compare the change behaviors of the bullwhip effect with respect to demand parameter under three different forecasting scenarios when a supply chain is coordinated. Our analysis shows that these change patterns of the bullwhip effect are different due to different forecasting tools.