Intermittent demand: Linking forecasting to inventory obsolescence

摘要: The standard method to forecast intermittent demand is that by Croston. This available in ERP-type solutions such as SAP and specialised forecasting software packages (e.g. Forecast Pro), often applied practice. It uses exponential smoothing separately update the estimated size interval whenever a positive occurs, their ratio provides of per period. Croston has two important disadvantages. First foremost, not updating after (many) periods with zero renders unsuitable for dealing obsolescence issues. Second, positively biased this true all points time (i.e. considering forecasts made at an arbitrary period) issue only following occurrence only). second been addressed literature proposal estimator (Syntetos–Boylan Approximation, SBA) approximately unbiased. In paper, we propose new overcomes both these shortcomings while adding complexity. Different from method, unbiased (for time) it updates probability instead interval, doing so every comparative merits are assessed means extensive simulation experiment. results indicate its superior performance enable insights be gained into linkage between obsolescence.