cancel
Showing results for 
Search instead for 
Did you mean: 

statistical forecast key figure dissag issue for sales org level

Rchav
Discoverer
0 Kudos

I have a key figure statistical forecast  at TWPRODLOCCUST  which is setup as disaggregate- equal distribution with key figure for proportionality as  dissag factor and week weight , we run it at material id-customer id level in application job template aggregation mode and generate a forecast...the issue is  as below

Suppose customer 1234567 has 3 sales org as AA01, AA02, AA03 and only  AA01 has a sales history, statistical forecast still get disaggregated into all 3 Sales org , I want to modify it as if AA01 has a sales history when stat forecast is calculated only AA01 should have stat forecast and not other 2 sales org

I tried running at material-custid-sales org level, changed the disaggregation to key figure for proportionality to only sales history but it did not work,  can anyone suggest some solution for this

 

 

 

Accepted Solutions (0)

Answers (2)

Answers (2)

amandeepnagpal
Discoverer
0 Kudos

Hi,

You need to first ensure that you calculate proportions on the basis of Sales History. Then this key figure where you calculated proportions should be used for Proportional Disaggregation in the Stat Forecast key figure. Run the algorithm at high level, it should split per proportions.

Selecting Sales History key figure will not give results because Sales History key figure will be blank in future where you want to generate the Stat fact, thereby giving equal split.

Best,

Aman

Rchav
Discoverer
0 Kudos

Hi Aman,

i created a KF - Proportional factor  and used this as a target key figure in calculate dissag factor forecast model where input kf is sales history based on this proportional factor got calculated and it is zero on combinations  where i do not have sales history and has a values only on combinations where i have sales history which i thought is correct apprach.

Later I used this P- factor  KF  in stat forecast for key figure for proportionality and ran the stat fcst at higher level - PRD family/Month level in order to get this dissag as per proportion but  i see stat fcst is still disaggregated to combinations(sales org) which has no sales history  

am i missing something ?your inputs is really appreciated

#IBP

riyazahmed_ca2
Contributor
0 Kudos

Hi @Rchav,

The whole culprit is the proportional factor.

  • It should not be blank for all the combination, otherwise it will go equal split of the target data.
  • Also it should not be the same value for all the combinations, in this case too, it will go for an equal split.
  • Finally it's value should be blank on the combinations where we don't want the forecast to be disaggregated and it should have the value in the right combinations to make sure the forecast disaggregated on them.

1) While we maintained the 'Disagg Factor' as the keyfigure for proportionality, Disagg Factor has undergone one or all of the above 3 conditions because of which Forecast has been split across to the incorrect combinations too.

2) While we maintained the 'Sales History' as the keyfigure for Proportionality, We had the first situation where the Sales History is blank for all the combinations. Hence Forecast has split equally in all those combinations.

In order to make the forecast split only to the correct combinations, make sure to chose a keyfigure that satisfies the 3rd condition specified above thereby placing the proportional factors only in the right combinations.

Best Regards,

'Riyaz'

Rchav
Discoverer
0 Kudos
Hi Riyaz, I went through your 3 conditions and other comment and followed below approach i created a KF - Proportional factor and used this as a target key figure in calculate dissag factor forecast model where input kf is sales history based on this proportional factor got calculated and it is zero on combinations where i do not have sales history and has a values only on combinations where i have sales history which i thought is correct apprach. Later I used this P- factor KF in stat forecast for key figure for proportionality and ran the stat fcst at higher level - PRD family/Month level in order to get this dissag as per proportion but i see stat fcst is still disaggregated to combinations(sales org) which has no sales history am i missing something
riyazahmed_ca2
Contributor
0 Kudos

Hi Rchav,

Can you help me with the screenshot of the the keyfigures you have created.

Best Regards,

Riyaz