APO DP- Heirarchy Defination

[Former Member]

Dear All,

Can anybody educate me as to wher do we define the Product, Sales Org,Geographical Heirchies for data data consolidation in Data Views.

Product</u> Product Group

Product Variant

Sales Organization Zonal Manager

Regional Manager

Customer

Geographical Country

Zone

State

Thank you in advance

Ashok

[kenneth_snyder]

Somnath,

Can you elaborate on what happens when you define dimensions?

However after defining the characteristics in the Master Planning Object Strcuture you can assign them to different dimensions (like location, product, geography) but that has nothing to do with data consolidation in data views. It creates different DIM tables in the APO-BW component for managing CVC values in the internally generated infocube.

I know how to create them, but really don't understand how they are used.

Ken

[somnath_manna]

Hi Ken,

Dimension is a logical grouping of characteristics. It helps in building dimension tables which in turn has the key for that group of characteristics. This DIM ID is then connected to the central fact table which contains the keyfigure data. So when you pull up a set of products for a particular product category, system looks into the Product dimension table - determine the matching keys and use that to access data from the fact table. This is the star schema that enables better (read faster) access to the fact table.

Every characteristic added to a new dimension creates a sub-table of the relevant characteristics called the dimension table.

Please see <a href="http://help.sap.com/saphelp_nw70/helpdata/en/4c/89dc37c7f2d67ae10000009b38f889/content.htm">this link</a> for star schema in help.sap.com

Thanks,

Somnath

[Former Member]

Hi Somnath,

it is me again on this Hierarchy topic. BW-wise I understand the use of dimensions, as they allow different and independent drill-downs.

And also for DP, the dimensions can be used to raise the limitations of certain drill-down-paths. However, what does that mean system-wise?

I mean, the number of CVCs is not influenced, as it is calculated by the product of the values of the most detailed characteristics in each dimension.

But when it comes to performance, does it make a difference here???

Thanks for your valuable insights,

regards,

Klaus

[somnath_manna]

Hi Claus,

I am not sure if Interactive Planning drill-down has to do something with dimensions especially in terms of performance. However given the fact CVCs are called from the Internally Generated Infocube (unique one for each planning object structure) I am sure having multiple dimensions will help performance-wise. You are correct about the CVC calculation but I guess the selection condition in the selection profile has an impact of how the CVCs are collated - or at least what amount of records in the DIM tables are searched before the selection condition is satisfied.

Its kind of hazy esp. for a functional person but I would not surprised if there is something technical in this regard which makes a difference. I guess you can post a new thread in the MaxDB forum to demystify this : )

Thanks,

Somnath

[somnath_manna]

Hi Ashok,

What is the exact requirement. In APO DP planning is carried out at a characteristic value combination (CVC) level. For example Product, Product Group, Customer Location, Sales Office etc. are different characteristics. This is defined in Master Planning Object Structure. Then you create CVCs by populating unique value combinations for these characteristics.

In the data view when you access data you select these CVCs (lowest level or at an aggregate level) in the shuffler and the according data in keyfigures is displayed.

For example if you want to see data for a entire Product Group - you load the data for required product group from Shuffler. Internally the system will determine all the CVCs having that Product Group as a value and aggregate the keyfigure data to display in the data view.

So there is no separate hierarchy that you need to create in APO DP.

However after defining the characteristics in the Master Planning Object Strcuture you can assign them to different dimensions (like location, product, geography) but that has nothing to do with data consolidation in data views. It creates different DIM tables in the APO-BW component for managing CVC values in the internally generated infocube.

Hope this helps.

Thanks,

somnath