NEW DEVELOPMENTS IN BUSINESS FORECASTING Larry Lapide, V.P. and Service Director, Supply Chain Strategies, AMR Research FORECASTING IS ABOUT UNDERSTANDING VARIATIONS (This is an ongoing column in The Journal which is intended to give a brief view on a potential topic of interest to practitioners of business forecasting. Suggestions on topics that you would like to see covered should be sent via email to llapide@amrresearch.com. Ed) As part of a consulting project I worked on for a major telecommunications manufacturer, one of the consultants did an analysis of the company's historical forecast error. He first computed the mean absolute percent error (MAPE) of the company's historical forecast errors for a sample of stock-keeping units (SKUs) and was astounded to discover that it was 70%. (MAPE, for the purposes of this article, is equal to the average absolute error divided by theme an demand expressed in percent). He thought this was high, but not too unrealistic since this was, after all, a company that sold very large telecommunications systems with many components having extremely slow-moving demand. He then computed the mean absolute percent variation (MAPV) of the historical ...

From Issue: Winter 1998
(Winter 1998-1999)

image
Forecasting Is About Understanding Variations