Skip to Content
avatar image
Former Member

Monthly FC consumption in SNP

Hello experts,

i'm new to this forum and pretty new to APO.

I need help to find a solution for a requirement expressed by our client.

The requirement is: (in SNP) S.O. must consume only forecast that lies in the s.o. request month. Even if the S.O. quantity is bigger than the monthly FC for that product/location(market), FC from the previous or successive month must not be consumed by s.o. that have a request date for "central" month.

As you know this is not possible to obtain using standard functionalities, because is not possible to specify "monthly" consumption horizons in the product master. And another problem should be that orders which are requested in past month but still not confirmed would appear on the "initial" column, therefore consuming FC from current month (this would be a big issue for the client).

The only solution that came to my mind is to modify the "Stock balance" macro, using a custom function copied by standard function:




Product master settings: backward and forward consumption = 32 days

Custom function must be able to understand that the consumption should stop at the 1st and the last day of the "S.O. request" month.

Do you think this is an applicable/feasible solution?

I was not able to find any other solution.

Opinions /suggestion will be greatly appreciated.



Add comment
10|10000 characters needed characters exceeded

  • Follow
  • Get RSS Feed

4 Answers

  • Best Answer
    avatar image
    Former Member
    May 12, 2008 at 11:23 PM


    We have a similar set-up where-in sales orders created consumes forecast in monthly buckets. The demand released to SNP is middle of the period and the master data settings for Consumption under the Demand tab (Requirement Strategy) are:

    Consumption mode is 2 - enables forward and backward consumption.

    Bwd Consumption Per. 15

    Fwd Consumption Per. 13

    This ensures that Sales orders raised in R/3 consumes AD in the same month.

    Revert in case you need further information.



    Add comment
    10|10000 characters needed characters exceeded

    • Former Member

      Hi Vinod,

      thanks for your answer. Unfortunately. the solution you apply is not feasible for our client, as forecast will be released as a daily quantity (the first working day of each week) and distributed according to a distribution function.(we will use standard "period split" functionality).



  • avatar image
    Former Member
    May 14, 2008 at 01:55 PM

    Hi Mauro,

    I think we have a similar situation. I have argued that artificially dropping off the forecast when the month changes isn't really a good idea, since causes the MFG plant to see strange behavior at the beginning of a month, but this is what our old system did so the users want that comfort.

    Our requirement is that an order in this month can't consume forecast from the previous month. An order that is entered May 15th can consume forecast all the way back to May 1st, but can't consume anything from April 30th.

    Forecast is released every day from DP to SNP, using a profile which is 60 days, then 46 weeks and then 6 months (18 month horizon).

    Forecast consumption mode is 2 and most plants are set to 30 days forward and backward consumption period.

    Our solution is this:

    Release the forecast every day via an ABAP job from within a process chain. The ABAP job is a variant of the program used in transaction /SAPAPO/MC90. The ABAP job uses a dynamic date calculation to release a range of (Beginning of month - 31 days, Beginning of month + 550 days).

    When the forecast is released, we then run a Forecast Reorganization task every day from within a subsequent process chain that reorganizes the forecast in SNP with a date range (Beginning of month - 360 days , Beginning of Month - 1 day).

    This method will allow sales orders from last month to consume forecast from last month. It then reorg's the forecast up until the last day of the previous month so that the forecast from last month is not available for any new orders that come across the CIF. The gap in this method is that when there is a situation where a sales order from last month is much higher than the forecast from last month, the sales order will spill over into the current month. We are still working this out, but it is not really a problem for us at the moment. You site this as one of your requirements, so you could solve this by creating a process that adds 1,000,000 to the forecast for last period and releasing it to SNP. The warning here is that you might not want to do that to the forecast key figure that you send to your BW system (it could possibly mess up your reporting).

    For the record, I don't think this is a great method, but it's what our users are comfortable with. I think a better method is to set the forecast reorg to end just when the backward forecast consumption period begins (on a plant-by-plant basis, since different plants can have different consumption periods).

    Add comment
    10|10000 characters needed characters exceeded

    • Former Member Former Member

      Hi Dan,

      thank you very much for your answer.

      That forecast re-organization could be interesting for me, because, as you say, it is necessary to "clean out any unconsumed forecast from last month so that there isn't a slug of forecast in the initial column on the supply planners' data view".

      In the initial column, s.o. row, we'll have two kinds of s.o. with "open quantities":

      1. s.o. with avail/req date in the past month

      2. s.o. with avail/req date in the past days of the current month,

      In our scenario, s.o. with open quantities with a delivery date in the past month, must continue to generate a demand in the initial column, but do not have to consume fc released for the current month.

      I believe this can be obtained through customizing, specifying in the planning strategy a category group that contains category BM (s.o.) - Qty type: 1 (requested quantity).

      SPRO/ Advanced Planning and Optimization/ Supply Chain Planning/ Demand Planning (DP)/ Basic Settings/ Consumption/ Specify Requirements Strategies

      SPRO/ Advanced Planning and Optimization/ Supply Chain Planning/ Supply Network Planning (SNP)/ Basic Settings/ Maintain Category Groups

      Before reading your post, i thought we should have deleted FC from previous months before releasing new forecast, to avoid having FC from the previous month in the initial column.Now you say it is possible to obtain the same result using the reorganization function /SAPAPO/MD74. I don't know how this transaction works, neither what it does; i'll try to make some test and look for some documentation.

      As for the "monthly" consumption, we are now trying using the "period" consumption (planning strategy 20, backward and forward consumption set to blank) and a BAPI at FC release.



  • avatar image
    Former Member
    May 19, 2008 at 05:09 PM

    Hello, I made this using a macro (I know that it's not a good idea by the performance side). I did the release creating a Z category and created the forecast using a simple macro that make the consumption and only stop when the day is 1. You have create a time series KF to store the last consumption situation and use this as a initial consumption (for the sales order already served).

    It's not easy but it's possible. I don't remember more details because I did it more than 2 years ago.

    Add comment
    10|10000 characters needed characters exceeded

  • avatar image
    Former Member
    May 13, 2008 at 03:47 AM


    I think you should write a custom program which will keep the bwd and fwd consumption period rollin depending on the day of the month such that it does not go beyond the monthly limits, we had similar issue and we tried this soln.

    Is your days of month on which you receive sales order fixed, then you can release the demand using infocube on that day and keep bwd and fwd period 0, you can use this provided your saled order day is fixed for every month.

    What is you planning bucket days or weeks?



    Add comment
    10|10000 characters needed characters exceeded