Flexible Planning: New Forecasting Functions


Release 4.0A of Flexible Planning offers you the following new forecasting functions:

To base the forecast on previously corrected historical values, and not on the original historical values, set the Corrected values indicator in the forecast profile . The system then calculates the forecast results using the corrected historical values of the forecast version that was last copied to, and is currently displaying in, the planning table.
The period indicator on which you base the forecast can be different than the storage periodicity. You set the storage periodicity in Customizing by choosing Set parameters of info structures and key figures . You enter the forecast period in the Period ind. field of the forecast profile.
You can base the forecast on an average number of days per period. For example, if your forecast period is "month", you can run the forecast on the assumption that the number of workdays in each month is 20 by entering "20" in the No. days field of the forecast profile. The system then:
a) Corrects the historical values using this standard number of workdays.
b) Runs the forecast using the actual number of workdays in the period. The system takes the actual number of workdays in the period from the factory calendar.
If you do not have any historical values for a material (for example, because it is new), you can still run the forecast for this material by using the consumption quantities of a "reference material." You define this reference material in the material master record. You must also set the Consumpn. indicator in the forecast profile. This function is supported for level-by-level planning.
To delete one or more forecast versions using mass processing, choose Planning -> Mass processing -> Reorgan. forecast. This function allows you to delete forecast versions in either Flexible Planning or the material forecast.
In Flexible Planning, you delete forecast versions in a particular planning version of the information structure; moreover, you can opt to delete forecast versions based on a particular forecast model (that is, a particular combination of forecast parameters).
In Flexible Planning, you can also delete individual forecast versions online in the planning table. Do this via the pushbutton Versions in the forecast dialog box.
If you do not intend to work with forecast versions, you can switch off the use of them by setting the indicator Do not save to forecast DB in Customizing (choose Set parameters of info structures and key figures ). The forecast results are then saved in the information structure but not in the forecast database. Therefore, if you run the forecast and copy the results to the planning table, any previous forecast results are lost. The advantage of setting this indicator is that you achieve performance improvements, especially when forecasting in the background.
To have the system automatically correct outliers in the historical data on which the forecast is based, you select Outlier control in the forecast profile. The system calculates a tolerance lane for the historical time series, based on the Sigma factor. Historical data that lies outside the upper or lower band of this tolerance lane is corrected so that it corresponds to the upper or lower tolerance limit for that point in time.
The width of the tolerance lane for outlier control is defined by the Sigma factor . The smaller the sigma factor, the greater the control. The default sigma factor is 1.0. SAP recommends that you set a sigma factor of between 0.6 and 2.0.
If you run the forecast online, historical data that has been automatically corrected by this function is indicated in column "C" of the Forecast: Historical Values dialog box.
Alternatively, you can automatically correct historical data online, without having selected Outlier control in the forecast profile. To do this, select "C" for each historical value you want to correct and choose Correct. The system then calculates the average of all the corrected historical values and displays it as the new corrected historical value.