cancel
Showing results for 
Search instead for 
Did you mean: 
Read only

IBP average demand calculated at different time period than input key figure

adamcohen7
Explorer
0 Kudos
1,749

Hi all

We have a demand key figure with base planning level of Technical Weeks.

We would like to have another key figure which calculates the average demand of the next 3 Months.

How is it possible to calculate this average when the input key figure is not at the same time period level?

Thanks

Adam

Accepted Solutions (1)

Accepted Solutions (1)

revanchatraban
Participant

Hi Adam,

You can use rolling aggregation with aggregation type AVG. Example for Average of Statistical forecast from weekly to monthly buckets is explained below:

1. Define key figure to aggregate the data from weekly to monthly.

2. Define key figure to calculate the rolling average for 3 months

Output looks as follows:

Regards,

Revan C

adamcohen7
Explorer
0 Kudos

Hi Revan

Thanks for your detailed and very useful answer. I was missing the fact the we could set a calculation for the AGGFCST key figure with Request level at planning level root (Month) different to the planning level root (Technical Week) of the key figure we want to sum.

Do you have a recommendation for what Base Planning Level to put for the two key figures AGGFCST and TMPROLLINGAVG? I tried both with Base Planning Level set to Technical Weeks and Months and it seemed to make no difference. In both cases the key figures get disabled when I display with Time Period lower than Months.

One other question: when using IBP_RAGGR, the 4th parameter is the duration of the rolling aggregation. Do you know if there is a way to refer to a master data attribute for this value instead of defining a fixed integer value in the calculation?

Thanks

Adam

revanchatraban
Participant

Hi Adam,

1. The reason for selecting the AGGFCST is to ease the calculation and understand the results. We need to mention the rolling periods in the function IBP_RAGGR. In case of technical weeks, it is difficult to provide the exact number of weeks and makes it difficult to analyze the results. You can achieve the same without AGGFCST key figure as well with directly using the STATISTICAL FORECAST. The details are provided as follows.

TMPROLLINGAVG@REQUEST = TMPROLLINGAVG@MTHPRODLOCCUST
TMPROLLINGAVG@MTHPRODLOCCUST = SUM(TMPROLLINGAVG@WKPRODLOCCUST)
TMPROLLINGAVG@WKPRODLOCCUST = IBP_RAGGR( STATISTICALFORECASTQTY@WKPRODLOCUCST, AVG, 0, 13, CURRENTFUTURE ) Here the 4th parameter should contain technical periods to be considered for average.

Yes, the key figure gets disabled as the lowest time period is Month.

2. 4th parameter: duration of rolling aggregation (mandatory). SAP says "It has to be a positive integer". So i guess there is no possibility to bring the attribute here.

adamcohen7
Explorer
0 Kudos

Thanks Revan, very clear!

Answers (1)

Answers (1)

piyush_parekh
Active Contributor

Hi Adam,

You can easily aggregate key figure from TW to Month level using SUM and then calculate average by using Period Shift, rolling or cumulative aggregation functions in IBP. Refer Simplified Key Figure Calculations section for details with examples in Model Configuration Guide.

https://help.sap.com/viewer/bf99e931b8d44aafb4e306ec3602cbdd/2008/en-US/27dce8dc9cbc47dab382478ffbca...

Hope this helps!!

Regards,

Piyush

adamcohen7
Explorer
0 Kudos

Thanks for your response Piyush. The link is useful.