The classification system in R/3 provides BW with characteristics that are similar to InfoObject attributes. Classification attributes are not delivered with the system, however, and must be created and maintained by end users. No predefined DataSources are available in BW for this highly customized data. The author explains the classification system and how to extract the data out of R/3 into BW.
Key Concept
The classification system in R/3 groups similar objects into classes as defined by criteria established by end users. Because the classification system is not application specific, classes of objects can be extracted, transferred, and loaded in BW and other applications with the appropriate extraction structures. The classification system in R/3 often is a mystery to many BW folks, especially those of us who have never worked on an R/3 configuration before. The cross-application component in R/3, the classification system, is application agnostic and describes objects according to their characteristics. Consequently, similar objects can be grouped into classes based on criteria that the R/3 team defines.
The decision to use the classification system lies with the R/3 group and not all companies employ it. It is set up as part of the R/3 variant configuration process, and its use varies widely among companies. For some firms, use of the classification system is optional. However, companies that produce configurable products such as automobiles, which can have variants like paint, trim, or engine size, usually tap this functionality.
It’s up to the R/3 group to configure the classification system, but BW professionals need to have a basic understanding of the classification system so they can access the information it holds. They should know at the very least if their company uses it and how to extract data out of the classification system and into BW.
I will provide you with an overview of the R/3 classification system so you can identify its classes. Then I will show you how to build extractors designed to bring data from classes in R/3 into BW. I’ll also share a few tips and tricks. If you need comprehensive information on the subject, SAP offers courses such as WNAP45: Variant Configuration Part 1 & 2 for SAP R/3 4.7 and PLM 130: Classification.
The R/3 Classification System
Classification is the act of allocating objects to classes. Individual characteristic values are mapped using object values. Table 1 presents a glossary of classification terms.
Configurable product | A product that can have variants — for example, a car can have different paint, trim, and engines | Class | A group of objects with the same or similar properties | Object | A classifiable unit such as vendor, material, or customer | Characteristic | A property of an object — for example, length is a characteristic of material. Note: In BW terms, this is the same concept as an InfoCube attribute. | Characteristic value | The value of a characteristic — for example, four inches for length | |
Table 1 | Classification terms |
Use the IMG to see how the classification system has been configured for the R/3 system. Follow the menu path SPRO>SAP Customizing Implementation Guide>Cross-Application Component>Classification System. This allows you, for example, to identify objects defined in the classification system and better design an extractor to transfer this data into BW.
In the R/3 IMG, click on the Maintain Object Keys and Class Types in the Classes structure to see a list of all classifiable objects along with their tables, transaction codes, and keys (Figure 1). Select an individual table and click on the display icon to access more information about the object’s key fields (Figure 2). This table information is required to configure classification system extractors, which I will discuss later.

Figure 1
A list of classifiable objects

Figure 2
Details about the object’s key fields
To view a class, go to transaction code CL03. Enter the Class and Class type into the appropriate fields, then click on the Char. tab to see the available characteristics for that class (Figure 3).

Figure 3
Display a class
You can display details about a characteristic by selecting the appropriate characteristic and clicking on the glasses icon (Figure 4). This brings you to another window that presents detailed characteristic information (Figure 5) to assist in your extractor-design efforts.

Figure 4
Choose a characteristic to display

Figure 5
Characteristic details
Another way to display data in the classification system is to go through the actual tables. Table 2 lists the most commonly used tables in this system.
CABN | Characteristics | CABNT | Characteristics description | AUSP | Characteristics values | TCLA | Class types | TCLAT | Class type text | TCLT | Classifiable objects | TCLC | Classification status | INOB | Link between internal number and object | KSSK | Allocation table: object to class | KSML | Characteristics for a class | |
Table 2 | List of tables for classification system |
The Classification System and SAP BW
Now that you have a better understanding of what makes up a classification system in R/3 and how to find its components, let’s apply this information to BW and develop a classification extractor. Characteristics of objects in the classification system are similar to InfoObject attributes in BW, so you can use the classification system to supply your BW system with InfoObject attributes.
The only difference between regular BW InfoObject attributes and classification attributes is that classification attributes are not delivered with the system. If your users decide to employ the classification system, then they must create and maintain their classification attributes. Because characteristic classifications are created and maintained individually, no predefined DataSources are available in BW business content. The exception is BW’s hierarchy extractors, which I will discuss later.
To retrieve data from the highly customized reporting structures in R/3, the BW team must create extractors and DataSources for classification characteristics using the same functionality used to extract data from the CO-PA module. A specialized tool set is provided in BW to help create these extractors. The benefit of creating these extractors is that they do not require the complex ABAP coding needed to create a generic extractor.
Generating DataSources
To create a DataSource for classification data, first use transaction SBIW and follow menu path Data Transfer to the SAP Business Information Warehouse>Settings for Application Specific DataSources (PI)>Cross-Application Components>Classification System>Maintain Data Sources for Classification Data. Select Maintain Data Sources for Classification Data. You can also use transaction CTBW to get to this screen.
In the overview screen, select Basis DataSource to extend along with the Client, Class Type, and Obj table that you want to copy (Figure 6), then click on Characteristics. Next, click on the New Entries button and enter the characteristics you want to assign in the next screen to appear (Figure 7). If you want to define a sequence for the characteristics, you must also enter position numbers.

Figure 6
Select the items to copy

Figure 7
Enter the characteristics to be assigned.
Note that the status values in Figure 7 are assigned automatically: The N value indicates that a DataSource has been created. The R status indicates the DataSource is ready and is achieved after generation. If you change a characteristic, then you must reset the status to N before generation.
Note
As of R/3 Release 4.0B, you can extract configuration data from the OLTP system. Before this version, only valuations can be extracted from R/3. For those using Plug-In 2002.2 and R/3 Release 4.5, it is possible to extract configuration data with IBASE using the function module VC_I_GET_CONFIGURATION_IBASE
Now double-click on Data Sources to generate the new DataSource. This also generates the characteristic DataSource for characteristics of data type CHAR, which I noted earlier in Figure 3 . An internal key with 1CL prefix is assigned as a DataSource name. As a result of this process, the following DataSources are generated:
- 1CL_OMAT001 for master data attributes
- CL_AMFG002, a text DataSource for extracting value descriptions
A replication of the DataSources now must be carried out in the BW system so that those created in the OLTP source are available in OLAP system. You access this function via the Source system for InfoObject menu and right-clicking on Replicate DataSources option. Table CLBW_SOURCES stores this data in R/3.
Tips and Tricks
Objects in a classification system can be arranged into a hierarchy that does not require a custom extractor. DataSources delivered with BW business content that extracts this information include: 0CONSUMER_LKLS_HIER, 0CUSTOMER_LKLS_HIER, 0MATERIAL_LKLS_HIER, 0PLANT_LKLS_HIER, and 0VENDOR_LKLS_HIER. If the object has no delivered extractors, you can create a custom DataSource for it and assign the object type the DataSource in table CLBW_SOURCES.
Metadata for the classification characteristics also can be generated by the system. This saves BW developers a lot of time because they do not have to develop each InfoObject manually. InfoObjects are created automatically using metadata and links to the DataSources for these InfoObjects are also established automatically. To do this, log on to the BW system and go to transaction CTBW_META. Check the Source system and the DataSource boxes, set the Create InfoObjects/Source, and press enter (Figure 8).

Figure 8
Transaction CTBW_META
Several SAP notes are available that cover the relationship between BW and the R/3 classification system including SAP notes 596196 (InfoSource existence check in transaction CTBW_META), 587087 (Extraction in the incorrect client), 588184 (Role of client in data extraction), and 535370 (Interval-value and multiple-value characteristics for BW).
July Hartono
July Hartono has more than six years of experience as an SAP BW consultant. Currently with Alpha Net Consulting, LLC, she has helped several large clients implement BW. Data conversion and extraction, development of custom DataSources, and data modeling are among her best skills. July received mySAP.com Delta Certification for Business Information Warehouse in 2003.
You may contact the author at julyhartono@sbcglobal.net.
If you have comments about this article or publication, or would like to submit an article idea, please contact the editor.