APO DP disaggregation factors not being used correctly

[Former Member]

I am using APO DP V5.

I am forecasting at a level of product at customer, and the forecast is then disaggregated down to a 'despatch location' level via disaggregation type P (using disaggregation factors in APODPDANT key figure).

Suppose I have factors in APODPDANT which disaggregate the forecasts 100% to despatch location A.

I follow these steps:

- run a forecast, which is correctly disaggregated to despatch location A

- CHANGE the factors in APODPDANT to force disaggregation to despatch location B

If I now rerun the forecast, generating the <u>same</u> forecast quantities (because I use exactly the same forecast model), the system <u>does not</u> apply the changed APODPDANT factors. This is incorrect behaviour!

But if I rerun the forecast, generating <u>different</u> forecast quantities (because I now use a different forecast model), the system <u>does</u> apply the changed APODPDANT factors. This is correct behaviour!

Any ideas on how to 'correct' the incorrect behaviour?

Thanks,

Bob Austin

[Former Member · Accepted solution]

Hi,

If you are disaggregating using APODPDANT and giving small proportion with decimals, the rounding may be a problem. APODPDANT can take no decimals. This could be one reason.

It looks like the planning area is not getting updated with the APODPDANT

Also check these 2 notes...they may be helpful:

Note 671586 - New report for calculating new aggregates

Note 976991 - Incorrect proportional factors

Hope this helps.

[Former Member]

Bob

The behaviour you are seeing now is how Disaggregation is designed to work...

disaggregation based on APODPDANT works only if there is no pre-existing values for the detailed levels

In order to new APODPDANT values to take effect you need to zero out the details (or the total value ,in effect). This will help in disaggregation to be reset when you enter a new total

Since you already have a forecast and you are using the same model the pre-existing details stay on... and so do preexisting values.. so new prop factors are not used

When you run the Forecasts with a new model you are zeroing out the old forecast and redisaggregating

To correct this behaviour.. run a macro that will zero out the old forecast at a total level before re-running the forecast

[Former Member]

Yes. Harish is right. I haven't thought about it. Is that the reason?