Business Intelligence

The term Business Intelligence (BI) was introduced by Gartner Group analyst Howard Dresner in the middle of 1990s and defined as a collective term for concepts and methods that support decision-making through information analysis, delivery, and processing. BI has become widespread in business practice and science and is widely used. However, there is still disagreement in understanding the term. This uncertainty leads to an undefined variety of definitions. A precise demarcation proves to be difficult since each selected definition remains vulnerable. In 1996, Business Intelligence was defined as follows: "Data analysis, reporting, and query tools can help business users wade through a sea of data to synthesize valuable information from it – today these tools collectively fall into a category called Business Intelligence".

Due to the different understanding of business intelligence, different architectures for business intelligence systems are presented in the literature. About the broad knowing of the term used in the present work, above all, various logical processes are given in the references, which forms the basis for a BI architecture. These processes are assigned to the individual concepts and techniques that are summarized in the term Business Intelligence.

In this paper, the following processes will be distinguished in BI architecture:
1. Data Collection
2. Data Integration
3. Data Storages
4. Data Processing
5. Data Presentation

The data collection includes the operational systems that provide the required data for the Business Intelligence system. In particular, a distinction must be made here between internal and external systems as sources. Through data integration, the required data is transferred from pre-systems, processed and condensed, which is referred to as ETL process. The purpose of the ETL process is to ensure that the processed data can be stored persistently in the data storage or maintenance. The data storage can be realized in different architectural variants. Here Data Warehouse and Data Marts are used. In data processing or data analysis, all concepts and tools that are primarily concerned with the evaluation and analysis of the data are assigned to this process. This level is therefore assigned to analytical applications, which evaluate the data stored in the data storage process according to predetermined criteria. This process also includes components that enable online analytical processing (OLAP) and data mining components that are used to detect data patterns. In the data presentation is the target group specific preparation and presentation of the analysis results for the user. For this purpose, different concepts are used, such as OLAP clients for the implementation of ad-hoc inquiries or prefabricated target-group-specific reports. This level can also be assigned dashboards or management cockpit, planning and balanced scorecards, which are becoming increasingly important.

The following figure gives an overview of the individual processes and shows which components belong to which process step.


Figure 1: Business Intelligence Process