最优需求预测下预测与处理过程中的牛鞭效应

研究了在AR(1)市场需求模式下,生产商与零售商所组成的供应链系统中,当零售商采取不同预测技术时,在信息预测,处理及传递过程中产生的牛鞭效应问题.证明当存在订货提前期时,零售商采用移动平均法及一次指数平滑法预测会导致在需求预测,信息处理及传递过程中产生牛鞭效应;而采用最优预测仅在需求相关性很强时存在有限值的牛鞭效应.并对二级供应链进行拓展,证明在信息传递过程中,非最优预测将导致上游需求模式复杂化,牛鞭效应逐级递增,而最优预测可使上游需求模式简化,遏制误差的传递.

Bullwhip effect in demand forecasting and processing based on optimal forecasting technology

In this paper a two-stage supply chain consisting of a single retailer and a single manufacturer is presented,within which the problem of bullwhip effect produced in demand forecasting and processing owing to different forecasting technologies is studied.The paper proves that with AR(1) process for the end demand,if retailer adopts moving average or exponential smoothing to forecast the non-zero lead time demand,then bullwhip effect in demand forecasting and processing will occur;if retailer adopts the optimal forecasting method,then bullwhip effect will occur conditionally.The supply chain mode is further extended to study the upstream demand mode,for which it is proved that the bullwhip effect is not always amplified under different forecasting technologies.