Tracking Signal:

The tracking signal indicates how well the forecast is predicting actual values. It may be determined as

Tracking signal = Algebraic sum of the error /MAD

By expressing the cumulative deviation in terms of number of average deviations, tracking signal becomes a good indicator of the consistent difference between actual and forecast values of demand. Take for example a case when the cumulative deviation was 100 units, and the average MAD was 200 units. The tracking signal for the period is 1000 ÷ 200 = 5. Suppose that the values of tracking signal for the subsequent periods are 5.2, 5.8, 6.0. This would mention that demand was consistently greater than the forecast. If the tracking signal is negative, it would mean that demand is remaining less than the forecast. Good forecasts should have closely as much positive as negative deviation, thereby resulting in a low cumulative, or algebraic sum of forecast error.

In order to monitor the accuracy of their forecasting method some companies set up a control chart for tracking signal so that the forecast shall not be raised or lowered as long as the tracking signal remains within control chart limits. If the control limits are set very high, it will not provide timely response and shall limit the usefulness of the forecasting system in managing inventories. On the other hand if the control limits are very low it will require too often review for items that were being satisfactorily forecast. Experience recommends that an acceptable maxima for tracking signal values is 4 for high-value items and up to 8 for low-value items. When the signal crosses this range, investigation and corrective action should be initiated.