In APO, the forecast model is maintained in the forecast profile. You can assign a particular material to the forecast profile and the forecast will be carried out using the parameters in that profile.
Basically same setup as R/3 - but you have an addtional
profile - which contains the parameters and you can reuse that same profile
for multiple products.
Q2) Using the Forecast Model 35 (Seasonal & Linear Regression Model) for Statistical Forecast generation. Facing the Issue that Model 35 is consistently Under Forecasting. What Formula or Logic is used by the Model to generate the Statistical Forecast. What are the Parameters that can really improve the Under Forecasting trend for Model 35?
Couple of quick checks that you should do:
1. Make sure there is no Like Profile active for that planning object that is interfering.
2. Do a Strategy 56 to test for season and trend. See if streatgy 56 indeed recommends regression model.
3. If strategy 56 fails for season or trend, and you enforce seasonal or linear regression model - I would expect to see the
behaviour you are seeing.
4. Make sure there are adequate time buckets for forecast initialization.
5. Make sure the forecast initialization buckets are representative
of the forecasts and are not outliers.
Q3) How can I set up selections of the product/customer combinations so that a selection contains combinations which will be forecast by the same forecast strategy?
If understood correctly, you want to assign Forecast profiles to selections ( CVC's at which you want to forecast ) in an easy way.
We have also faced this problem, so I have got some suggestion for this.
1. Create a Forecast Master profile with "Forecast with Automatic Model Selec" as Forecast Strategy in Univariate profile which is attached to that master profile.
2. Create an activity using Tcode /SAPAPO/MC8T for Forecasting with parameters in Forecast tab : Master Profile - Created in earlier step, and in control parameters check ' Generate profile and assign it & Generate new profile when errors'.
3. Create a backeground job with the activity defined in previous step and in aggregation level select characterstics on which you want to Forecast ( in your case it could be Product & Customer i guess ) and execute.
4. Now goto Tcode: /SAPAPO/MSDP_FCST2 and give plang area and execute.
5. You can see all possible combinations got created in the table and assigned to some dummy profiles, there you can assign Forecast profiles manually.
6. Otherwise goto Tcode: SE16 and enter table : /SAPAPO/DP_FCST2 and execute with your selection and take dump into excel sheet.
7. If you really good in excel then it wont take much time to assign Forecast Profiles to your combinations.
8. Once your sheet is ready with all the data you can load that into table /SAPAPO/DP_FCST2 using LSMW ( Tcode: LSMW). Dependency for the above is that you have to create LSMW program using these characterstics. Its one time job but can be used when ever you want to chage Forecast profiles.
So far, LSMW is the easiet way to do it.
