Limitations Of Averages When Forecasting Rates Of Sale
Many people use averages for forecasting (daily or weekly rate of sale), but it does mean you always under replenish sales as you go into a peak and over replenish sales coming out of a peak. This effect is shown below.
This graph shows the sales of an item going into and coming out of its peak. If you are at time A, looking back over the last 4 weeks and you calculate the simple average of the last 4 weeks or even a weighted moving average, the dotted line shows where the average will be and if you take that as your sales forecast, you will miss all the sales in the hatched area above the line.
If you are at time B, calculating an average of the last 4 weeks the result will be similar to the dotted line shown there. By using that as your forward forecast, you will overstock the store by the amount shown in the hatched area at B.
Hence you will under stock the store going into a peak and overstock it coming out of the peak.
In a situation like this, Martec recommends that you either use a profile based method or use a method with a seasonal trend built in.
If you’d like more best practice advice check out our forecasting e-learning course.
To keep up to date with our best practice posts register to receive notifications via our mailing list.
-
Posted by Brian Hume
15th August 2018