Comments on Brightwork Articles on Forecast Error Measurement

Table of Contents: Select a Link to be Taken to That Section

Executive Summary

This article contains comments from articles on forecast error measurement.

Introduction

These comments are in response to the articles on forecast error measurement.

Comment #1: Lorraine Zungu

When calculating WAPE and either the Actual demand or the Forecast is zero, is that absolute error treated as 100 and Forecast accuracy zero

This is one of the biggest problems with MAPE. The error is not measurable in those circumstances. This is why I stopped using MAPE.

Comment #2: Michael Gilliland

“What you write about is, I think, consistent with how I would permit adjustments to the forecast inside supply lead times when I was managing forecasting at a processed meats manufacturer. There we needed 3 weeks to make changes to supply (we manufactured everything at our own factories). Forecasts (of shipments to customers, by item/DC/Week) were “locked” 3 weeks in advance for measuring forecast accuracy.
Our production plans were built around a target inventory for each item, which was about 2.5 weeks of supply. Pretty much every item was manufactured every week (in quantities approximately matching average weekly sales, adjusted up or down based on the projected inventory level, to make sure we maintained about the right weeks of supply for each item/DC). Because of short shelf life on the products, it was critical to maintain appropriate inventories.

If we realized there would be a huge bump (or drop) in sales within the 3 week lead time, we would adjust the shipment “forecast” accordingly, so we would get a better projection of future inventory. This would drive appropriate changes to the production plan as soon as it could be changed (3 weeks out).

Forecast accuracy was measured on the initial forecast that was set 3 weeks prior to shipment. But by adjusting the forecast within lead time, when necessary, it would allow inventory levels to more quickly recover to where they should be.

As an aside, I consider measuring forecast accuracy within supply lead times as “cheating” and also potentially dangerous — giving the organization a false sense of how well they can truly forecast their business.”

Thanks for the comment Mike, I completely agree. Changing forecasts inside of the lead time is really just a tool for supply planning to perform “housekeeping,” as you write — it’s not a forecasting approach.

Would You Like to Comment and Have it Added to This Thread?

Just provide your comment in the chatbox in the lower left of this screen.