Abstract
Abstract In general cases of inventory, we can find enough records to document the previous material. This has been intended via the history of that material, in contrary we suppose the rate of spare stock for the offer is non – zero. Whilst, when the demand is slow, it is difficult however to obtain the enough data to document the previous demand that identify the future demand on these materials with any degree on statistical confidence. Typical time series models (single exponential smoothing) are inadequate in the case of intermittent time series, because many of the series values are zero. Since these models are based on weighted - summations of past values, they bias forecasts away from zero. Unlike the single exponential smoothing that provide forecasts for future time periods, intermittent forecasting models provide recommended “stocking levels” or “estimated demand per period” that are used to satisfy future demand. Croston’s Method dissects the intermittent series into two components: a demand interval series and a demand size series. Both of these component series are indexed based on when a “demand” occurred (demand index) and not each time period (time index). The demand interval series is constructed based on the number of time periods between “demands.” The demand size series is Constructed based on the size (or value) of the “demands”. .
