Analyzing several million data in just a few minutes

Need for a powerful data analysis tool

Efficiently managing and exploiting a huge database of information is a challenge to any system. Processing times can be long, which hampers working efficiency and service quality. SAS offers a solution that enables rapid processing, efficient data analysis and data mining, even with very large databases. That is what the Belgian Federal Public Service (FPS) Finance experienced in 2010. Several specialized services (e.g. computation of the exact value of real estate, carousel fraud) within FPS Finance recently switched to a SAS environment for datamining, risk analysis, and – in a few cases – performance management.

The FPS Finance developed a business analytics (BA) environment to meet its analysis needs. Involving all fiscal departments, they implemented a data analysis system.

The ICT Department created a data warehouse coupled to a risk analysis environment in an integrated BA platform. 'We initially chose another data analysis and data mining solution, and built a team of business analysts to use that system,' recalls Michel Vessière, SAS Tool Project Manager at the FPS Finance. 'These analysts combine tax knowledge, ICT competence, and data analysis capabilities, and required a powerful tool to efficiently support their enquiries. Because of the vast amount of data involved, however, our previous system was neither fast nor flexible enough. Up to half a day was sometimes necessary to obtain the required information.'

The FPS Finance therefore wrote out a set of specifications for a new data mining tool. SAS was selected following the call for tender, which was based on a large number of (confidential) criteria.

Increased efficiency in analyzing risks

Today, the FPS Finance has implemented a SAS platform for data mining and analysis. The tool is used by business analysts within several departments for a wide variety of operations, including data mining, risk analyses, and ad hoc inquiries. Risk analyses are carried out to identify which individuals or companies present a risk of not meeting their fiscal obligations, and which ones are likely to become insolvent. The data mining functions are used, for instance, to identify various forms of fraud or to calculate scores that help predict bankruptcies earlier.

'Our business analysts already appreciate the greater power of the tool,' stresses Vessière. 'Whereas it previously took them up to half a day to produce a scoring list, for instance, they can now obtain this in a few minutes. Overall, the SAS tool is better able to handle the huge volume of data involved, compared with the previous system.'

Managing change through staff training

A complex part of the project was to migrate all data from the previous data analysis package into the SAS environment. SAS experts helped FPS Finance personnel to carry out this migration. In addition, the switch to SAS required a number of adaptations in order to fully exploit the vast potential of the tool. 'Implementing SAS was like switching from a hand-crafted data analysis environment to an industrial one,' illustrates Vessière. 'Many staff members had to be retrained as a result.'

SAS helped the FPS Finance provide the necessary training for staff, considering their specific activities. The training courses were customized and involved numerous practical exercises from day-to-day operations. 'This included, among other things, identifying fraud cases as well as the typical characteristics of high risk companies.'

Access rights clearly defined

The SAS implementation occurred within a context of additional changes resulting from the increasing need for ICT security demanded by federal authorities. Databases, for instance, needed to be more centralized and access to certain files increasingly restricted. Stricter privacy rules were also taken into account. All of these evolutions required a complete review of working methods. Tasks and access rights had to be allocated differently and the organization of services had to be adapted. In this respect, SAS allowed user access rules and profiles to be clearly and easily defined.

'We should now consider implementing additional reporting features,' concludes Vessière. 'These are already included in SAS and will enable us to provide members of Parliament with a regular and detailed overview of specific figures and trends.'

Example: the Collection and Recovery Department follows up and recovers due taxes

The Collection and Recovery Department of the FPS Finance collects money from taxpayers and follows up on late payers. It also recovers amounts due from nonpaying individuals and companies, using if necessary all legal means at its disposal. Every year, the department handles information related to over three million income tax returns, which results in one of the largest databases in Belgium.

'In addition to the regular follow-up of taxes, our department must often answer ad hoc questions from members of Parliament. It is essential for us to be both reactive and flexible, and to provide precise information when answering these questions,' explains Vessière. 'SAS also enables us to efficiently follow up on the performance of the FPS Finance's 45 local collection offices.' This is done using a series of key performance indicators (KPIs). Using these KPIs, benchmarks can also be established to compare the performances of the local collection offices.

Copyright © SAS Institute Inc. All Rights Reserved.