Does this describe your current situation? You have implemented SAP wall-to-wall, you have followed all the SAP Best Practices; you have a solid Strategy Management system in place, such as the Balanced Scorecard, but so do your competitors? Are your financial results just so, so? Well, maybe it is time to roll out the big guns of the black art of Data Mining, and get all predictive modeling on your business processes.
SAP actually has a variety of data mining applications and analysis tools, including the Analysis Process Designer (APD) and the SAP Business Objects Predictive Analytics workbench. The output from the data mining process using the APD and workbench can both be stored in SAP BW or sent and used in the SAP Customer Relationship Management (CRM) system for further analysis. The APD out-of-the box data mining capabilities and data mining techniques which include:
Decision Trees
Clustering
Association Analysis
Scoring
Weighted Scoring Tables
ABC Classification
You can also create your own data mining models with the SAP BW APD tool as well, leveraging the data warehouse as a source of data. Of course, this tool allows your data miners to write ABAP programs and use them anywhere along the data flow you are setting up within APD to perform additional advanced processing functions, such as recognizing data patterns. The output from from any of these ABAP programs can be used as ‘seed’ data to be fed back into BW.
A little known, and perhaps underused capability of the APD is a data-cleansing workbench. The Predictive Analytics Workbench from SAP BOBJ provides predictive data mining capabilities including:
Trend and Pattern Analysis
Forecasting and Planning capabilities
Ability to feed data to both interactive dashboards and the BW system
The workbench is especially powerful in the Retail Industry when it comes to using data mining and predictive analytics to understand, model and respond to customer behavior.
What do I consider the secret to successfully using these tools?
Good quality data available from the data warehouse or other source system
An openness to going where the data takes you.
Having business processes that are amenable to mathematical analysis, i.e., customer affinity modeling.
One industry that I have seen some heavy duty statistical analysis used in is the transportation industry, particularly the trucking industry, where it gets pretty deep very quickly. Think enroute merges. Tends to bring a room full of servers to their knees. SAP APD is just one of the tools in the wider suite of SAP BW tools and data warehouse applications. That is one of the reasons the APD process can become complex, it's a powerful tool designed to deal with vast amounts of complex data.
Our data mining consultant engineers and data scientist have deep expertise in applied statistical analysis and would appreciate the opportunity to work with you to solve your toughest problems.
More reading:
What are 2 Primary Focus Areas for SAP Airline Customers?
3 Ways to Design KPIs Using SAP Solution Map Composer
Top 10 Benefits for Financial Analysis in the Latest SAP EP4
BusinessObjects Explorer Best Practices