Ansicht
Dokumentation

Changes in Forecasting ( RELNAPO_401_FCS_FCST )

CPI1466 during Backup   SUBST_MERGE_LIST - merge external lists to one complete list with #if... logic for R3up  
This documentation is copyright by SAP AG.

Short text

Changes in Forecasting

Use

The following changes have been made in forecasting:

Aggregated lifecyle planning

See Lifecycle Planning.

BAdI for Measure of Error

A BadI (/SAPAPO/SDP_FCSTERR) is now available with which you can calculate your own measure of error.

Automatic Selection in Composite Forecasting

A composite forecasting profile contains several univariate profiles. You can now specify that based on one measure of error the system automatically selects the best profile. This is particularly of use when you are choosing and optimizing a profile for a particular forecasting task. You can also use the measure of error that you calculate with the new BadI for this purpose.

Exclude Erroneous Univariate Profiles in Composite Forecasting

You can set an indicator for univariate profiles in composite forecasting so that if errors occur when calculating the forecast results for this profile, the profile is not taken in account for the composite forecast. This means that the proportion that was intended for this profile is distributed amongst the other individual profiles.

Generic Forecast Profile Assignment

You can now specify when assigning a forecast profile to a selection ID that the the assignment is also valid for all related selections. For example, if the selection is PRODUCT = P1, the profile is also used for PRODUCT =P1, LOCATION = 0001. You can however specify more detailled assignments to other profiles without invalidating the more generic assignment. A BADi is available with which you define the priority for different charactaristics, if there are several possible selections which are equally possible.

Changes to Automatic Model Selection Procedure 2

In a univariate forecast profile using this method (strategy 56) you can specify which measure of error, including one defined in the BAdI, is used for parameter optimization. You can also optimize the the number of seasonal periods used in the model by specifying the margin in which the system should carry out the optimization.

Sporadic Forecast in Croston Model

Previously the Croston method (strategy 80) used sporadic historical data to produce a forecast that was constant over all time buckets in the forecast horizon. It is now additionally possible to produce a forecast with intermittent demand. The system detemines an interval after which the demand is scheduled, for instance 100 kg every 6 weeks instead of 100 kg every week.

The system now generates an ex-post forecast for the Croston method.

Additional Rounding Method

In forecasting with integer values the system only saves whole values, for example when using the unit of measure Pieces. This meant previously that if a forecast run were to calculate values of 0.4 for 10 time buckets, the system would write a value of 0 for all buckets. The new optional method enables you to specify a new rounding method in a univariate forecast profile.

The first value in a series of forecast values is rounded up to the next integer value. The remainder, which can be negative, is passed on to the next bucket and added to the value in the next bucket. The integer part is written to the screen and the remainder passed on the next bucket, where the process is repeated. As a result the overall error due to rounding effects is reduced.

Effects on Existing Data

Effects on Data Transfer

Effects on System Administration

Effects on Customizing

Further Information

ROGBILLS - Synchronize billing plans   rdisp/max_wprun_time - Maximum work process run time  
This documentation is copyright by SAP AG.

Length: 4020 Date: 20240416 Time: 142941     sap01-206 ( 86 ms )