Forecast Analytic System Settings
In order to calculated the forecast statistics, two Advanced Planning System Settings must be configured:
- MAPE Calculation Time Frame: Allows you to configure the number of periods in the past to use in calculation of the error and BIAS measures
Lag Distance: Allows you to configure the number of periods of lag between the forecast and the actuals. It is recommended to select a lag measure that closely aligns to the average lead time of your products. At present there are 4 Lag options:
- Lag 0: Last month forecast for the period versus last period actual for the same period
- Lag 1: The forecast for last period from 1 period ago. For example, the forecast for the month of March as determined in the month of February
- Lag 2: The forecast for last period from 2 periods ago. For example, the forecast for the month of March as determined in the month of January
- Lag 3: The forecast for last period from 3 periods ago. For example, the forecast for the month of March as determined in the month of December
Forecast Statistics Tab Structure and Metrics
The forecast statistics tab is organized with two statistics grids - Parent and Child.
Parent grid is the current level and node in which the forecast is being viewed. Child grid are the entities within the context of the Parent Node. For convenience, the number of children and grid pages are also provided.
Please note that if you are at the lowest level of a hierarchy, there will be no child nodes listed.
Please note that all calculations are based on approved plans and their calculations are based on the values that populate the waterfall.
- Rolling 52W Units: Sum of last 52 weeks of units
- Rolling 52W Revenue: Sum of last 52 weeks of revenue
- Percent: The percentage contribution within the sum of the revenue of all listed nodes in the grid (which are within one parent).
Cumm %: The cumulative percentage is the cumulative sum of all Percentage values for this node
- For example, T-Shirts which are Red, Yellow, Blue and Green, and their revenue is Red - 40%, Yellow - 30%, Blue - 20% and Green - 10%. Thus their cumulative percentage will be Red - 40%, Yellow - 70%, Blue - 90% and Green - 100%.
- Algorithm: The selected forecast algorithm. Available only on the forecasting level(s).
- ADI: Average Demand Interval measures the regularity of a demand in time by computing the average interval between two demands.
- CoV: Coefficient of Variation over the last 52 weeks period of time calculation. To compute the coefficient of variation, we will only consider the non-zero values of the forecasting history.
- Demand Profile: This is based on ADI & CoV, defining the type of demand as Smooth, Intermittent, Erratic or Lumpy.
- Best Fit Or Plan: Determines which demand plan value has been more accurate. The best-fit forecast or the user define demand plan override.
- Plan Mape: ABS(Actual - Plan) / Actual against the Demand Plan value
- Best Fit Mape: ABS(Actual - Plan) / Actual against the Best-Fit statistical forecast
- Plan WMape: Calculates plan errors based on the context nodes below the current level. The error of each node is weighted based on the shipment history. Provides an understanding of the degree of error within the level that the current level MAPE may not identify alone.
- Best Fit WMape: Calculates best-fit errors based on the context nodes below the current level. The error of each node is weighted based on the shipment history. Provides an understanding of the degree of error within the level that the current level MAPE may not identify alone.
- Plan Bias: Positive value means the demand plan is consistently less than actual. Negative means that the demand plan is consistently more than actual.
- Plan Bias %: Above value represented as a percentage
- Best Fit Bias: Positive value means the best-fit forecast is consistently less than actual. Negative means that the best-fit forecast is consistently more than actual.
- Best Fit Bias %: Above value represented as a percentage
- Stability Index: A measure of how much the total demand deviates from the recent past. Stability Index is Average of Last 3 months forecasting history vs Average of Current + Next 2 months total demand plan
Run Analytics Prompt
Yellow color coding indicates when the Analytics need to be recalculated. Please note that you will need to refresh the browser to remove color code after running.