Question on Dis-aggregation Calculation types in Demand planning

[former_member323340]

Hi All,

I have some queries on Dis-aggregation calculation types APO Interactive Demand planning. I have gone the SAP document on the same. It clearly explains what are the different calculation types and how does each work. However I would like to know the below point,

1. Why is dis-aggregation necessary.

I believe the forecast accuracy would be more if planning is done at a aggregated level. To get the values at the lower level, dis-aggregation is necessary. Also planning at detailed level consumes lot of load on the system because of which dis-aggregation is necessary. Is my understanding correct.

2. Under which business condition, which calculation type would be used.

I would like to know the practical scenario where each of the calculation types would be ideally suitable.

Can anyone help me understand the above.

Thanks

[Former Member · Accepted solution]

Hi

There are multiple scenarios in this aggregate forecasting plays a vital role.

For example, in our scenario we do forecasting at product group level at region level in the initial time.

As you mentioned, we could be more accurate at this i.e. aggregate level for forecasting in demand planning. But for supply planning we require information at more detail level like article and plants.

Hence it is important to dis-aggregate the FCST on the level which is also important for supply planning to have integrated business planning.

I hope this gives high level understanding. Please let me know if it helps to provide you more information.

Thanks

Amol

[Former Member]

Lakshmanamurthy,

Disaggregation/aggregation is required to maintain internal consistency within the planning area.  In most cases, the data you see in a Planning book is created and displayed at a summary level.  You have to define how you want the data to actually be stored.

Assume your characteristics are Region, Customer, Material number.  Further assume that in Region 1, your Marketing group creates their forecasts at the Region/Material level; but for region 2 your Marketing group creates forecasts at the Customer/Material level.

Both of these forecasts have common elements (e.g. they are forecasting the same materials and the same customers).  You need to have a way to ensure that each of these forecasts is stored in a way that allows these different forecasting methods to co-exist in the same data structure.

SAP DP does this my storing ALL forecasts at the lowest level (the most granular level).  In the example above, the forecasts are ALL stored at the Region/Customer/Material level.  Since noone is actually creating forecasts at this level, DP creates the missing data using aggregation/disaggregation.

You can define how the data is disaggregated.  A very common method is proportional.  The system evenly distributes your forecasts at the higher levels down to the lower levels, using a simple proportion.  Another method commonly used is to disaggregate by using another key figure.  For instance, if you create a statistical forecast at one of the higher levels, you may want to disaggregate to the lower levels by using, say, the ratios found in your sales backlog, rather than using ratios calculated using a simple proportion.

With respect to your question 2, there is no simple answer.  You have to look at each key figure; you have to know what each key figure represents and how it is to be populated in the Planning area, you have to know how the business intends to enter their data, you need to know how they intend to report their data, you have to know how any exported data will ultimately be used should you decide to release a forecast to a planning system, and you then must decide which disaggregation method will best suit all of the business processes that must be supported.

Best Regards,

DB49