On the variance of intermittent demand estimates
On the variance of intermittent demand estimates
复制标题
DOI：
10.1016/j.ijpe.2010.07.005
复制
发表时间：
2010-12
影响因子：
12
通讯作者：
A. Syntetos;J. Boylan
中科院分区：
文献类型：
--
作者：
A. Syntetos;J. Boylan
间歇性需求是随机发生的，许多时间段根本没有需求。由于相关的双重变化源（需求间隔和需求规模），预测此类需求模式是一项具有挑战性的工作。近年来，该领域的研究发展迅速，新成果因其实际意义而被应用到供应链软件解决方案中。在库存环境中，预测的准确性及其可变性（平均值的抽样误差）对于服务水平实现和/或库存成本最小化而言具有同等重要性。尽管前一个问题已被广泛研究（主要基于 Croston 的模型，1972 年），但后者在很大程度上被忽视了。本文的目的是根据估计的方差来分析最常被引用的间歇性需求估计程序。提供了详细的推导以及对基本假设的讨论。因此，我们希望我们的贡献可以为该领域的进一步分析工作提供参考，并有助于更好地理解与间歇性需求建模相关的问题。
Intermittent demand occurs at random with many time periods showing no demand at all. Forecasting such demand patterns constitutes a challenging exercise because of the associated dual source of variation (demand intervals and demand sizes). Research in this area has developed rapidly in recent years with new results implemented into supply chain software solutions because of its practical implications. In an inventory context, both the accuracy of the forecasts and their variability (sampling error of the mean) have equal importance in terms of service level achievement and/or inventory cost minimisation. Although the former issue has been studied extensively (mainly building upon Croston’s model, 1972) the latter has been largely ignored. The purpose of this paper is to analyse the most well-cited intermittent demand estimation procedures in terms of the variance of their estimates. Detailed derivations are offered along with a discussion of the underlying assumptions. As such, we hope that our contribution may constitute a point of reference for further analytical work in this area as well as facilitate a better understanding of issues related to modelling intermittent demands.