Intermittent demand is characterized by demand data that has many time periods with zero demands. Intermittent demand is hard to model by conventional distributions as well as hard to forecast via traditional forecasting techniques. This paper compares traditional forecasting techniques, simple exponential smoothing (SES) and moving average (MA) and intermittent demand forecasting techniques Croston's approach and Revised Croston method, at different levels of intermittent demand scenarios, based on various weighted error policies. The paper also discusses the effect of temporal demand aggregation on forecast error.
展开▼
