Phase out settings at a higher level such as brand or major customer

Question

Have any of you ever setup phase out assignments at a higher level than product and had it work correctly?  For example, we want to phase out a brand for a major customer.  In other words, a customer is dropping a brand and we don't want statistical forecast generated for that brand customer combination any longer.  I am able to setup the fields in phase out lifecycle settings for product, brand, major account but when i enter the brand and major account I am still getting forecast generated.  It appears to stop for some products within the brand but not all.  Another example is if a customer quits ordering from us I want to setup the major customer to phase out so no forecast is generated.
If you have done this successfully please let me know.  Or if you would handle these situations in a different manner other than phase out please let me know.  We can do historical adjustments each period but that is a lot of maintenance to do after each period before statistical forecast is generated.
Thanks
Steve

Answer 1

Hi Stephen,
Life cycle planning works only at the detail level (each CVC), the option of aggregate planning is helpful if you want to phase in or out a certain CVC when you are forecasting at the aggregate level.
One option is that you should have all products in the "profile assignment for life cycle " section which falls under that brand and customer. you can maintain a file and then automate the upload process in to the "assignment"
or
you can try to use the copy functionality in the realignment (/SAPAPO/RLGCOPY) , where you maintain copy factor as NIL and when the stat fcst is generated, you can use this as the next step to zero out -but you would need to maintain them manually.
or
easiest and safest wayy would be create a selection for those combination and do not include them in the planning job for stat fcst.
or
you can build a customised program to access the PA, PB, Data view and input the selection to zero out the stat fcst KF for that particular selection after the stat fcst run. here you would need to check if the diaggregated values are good enough.
hope it helps.