Tracking Signal
There are two types of Tracking Signals in Servigistics.
For Supply Planning
A measurement that indicates whether the forecast average is keeping pace with any genuine upward or downward changes in demand. It monitors the forecast to see if it is biased high or low. As used in forecasting, tracking signal is the number of mean absolute deviations that the forecast value is above or below the actual occurrence.
Tracking Signal = (RSFE / count) / MAD
• RSFE = Running Sum of Forecast Errors include all slices, demand and archived forecast
• MAD = Mean Absolute Deviation
• Count = Forecast parameter "# of Slices for Forecast Error Calculation" or Best Fit parameter "Holdout Window".
 |
You will only see this field when Enable Tracking Signal is set to Yes for the Forecast Parameter of the selected stream. The Tracking Signal is computed for each SKU stream and the results are displayed as follows:
• On the Demand page
• On the Best Fit Management page
|
For Performance Analytics and Intelligence
A measurement that indicates whether the forecast average is keeping pace with any genuine upward or downward changes in demand. It monitors the forecast to see if it is biased high or low. Calculated as [Σ N observations (Demand of X periods - Forecast of X periods)] / [Σ N observations (Abs(Demand of X periods - Forecast of X periods))]
• The range is [-1, 1]
• A Tracking Signal value closer to zero indicates an accurate forecast.
|
|
Each demand or forecast record is formed by aggregating data of x forward-looking periods:
• For PAI Foundation dashboards, the forecast metrics are calculated using x = 1 slice
• For PAI Advanced dashboards, the forecast metrics are calculated using x = Lead Time, x = 3, or x = 6 slices
|