TY - UNPB

T1 - Forecasting intermittent demand

AU - Teunter, R H

AU - Duncan, L

PY - 2006

Y1 - 2006

N2 - Methods for forecasting intermittent demand are compared using a large data-set from the UK Royal Air Force (RAF). Several important results are found. First, we show that the traditional per period forecast error measures are not appropriate for intermittent demand, even though they are consistently used in the literature. Second, by comparing target service levels to achieved service levels when inventory decisions are based on demand forecasts, we show that Croston's method (and a variant) and Bootstrapping clearly outperform Moving Average and Single Exponential Smoothing. Third, we show that the performance of Croston and Bootstrapping can be significantly improved by taking into account that each lead time starts with a demand.

AB - Methods for forecasting intermittent demand are compared using a large data-set from the UK Royal Air Force (RAF). Several important results are found. First, we show that the traditional per period forecast error measures are not appropriate for intermittent demand, even though they are consistently used in the literature. Second, by comparing target service levels to achieved service levels when inventory decisions are based on demand forecasts, we show that Croston's method (and a variant) and Bootstrapping clearly outperform Moving Average and Single Exponential Smoothing. Third, we show that the performance of Croston and Bootstrapping can be significantly improved by taking into account that each lead time starts with a demand.

KW - Forecasting

KW - Inventory

KW - Intermittent demand

M3 - Working paper

T3 - Management Science Working Paper Series

BT - Forecasting intermittent demand

PB - The Department of Management Science

CY - Lancaster University

ER -