Introduction to Business Intelligence

Follow FullPress on Google

Add us to your preferred sources.

Business Intelligence (&Decision) Systemsare the systems dedicated to B.I. (&D.) and consist of various solutions of: • application software (analytical applications, query tools, navigation and data presentation, etc.) • environment software (RDBMS, MDBMS, warehouse admin, modelling tool, ETL tool, data catalog, …) • hardware and basic software (DB server, Application Server, Web server, operating system and network, …) BI(&D)S integrate and complement Datawarehousing systems. From a technological architecture perspective, three basic levels can be identified. The first architectural level consists of “feeding” operational information systems which may be integrated ERP systems (Enterprise Resource Planning), current or historical, systems dedicated to Customer Support or various applications of Web technologies (eCommerce, Portals, eSupply Chain, Help Desk, etc.). In companies that have implemented ERP systems or have deeply renewed their operational S.I. on the Supply Chain or customer relationship fronts (CRM, Customer Relationship Management), one way to leverage the informational potential of ERPs (the only broad and integrated data base) is precisely to build the second and third architectural levels, i.e., Datawarehousing and Business Intelligence (B.I.) systems. Between the first and second level, there are specialized software tools dedicated to mapping, cleaning, and transferring elementary data into the physical Databases of the second level. The latter represents the level of directional Databases, created with logical approaches of Datawarehousing or Datamarting and also fed by external environmental data to the company (increasingly collected from the Internet world with Web Farming systems): these directional databases “decouple” the operational and transactional environment from the control, analysis, and managerial decision environments and integrate multiple data sources. At this level, directional data can be stored both using relational database technologies and multidimensional database technologies that allow specific OLAP type analyses, as well as achieving consistent technical performance in data analysis and interactive navigation aimed at decision support. These Databases in turn feed the third architectural level, that of Business Intelligence systems, made up of various types of software tools (application packages + software tools). Among the application packages are: • analytic applications, whose application software also contains managerial control or decision support logic (e.g., applications for sales data analysis, management consolidation, sales budgeting, and so forth), to be parameterized based on the specific needs of the company • DSS Tools, divisible into two parts along the dashed diagonal dividing the B.I. systems (figure 1): A) the portion closer to Analytic Applications, which contains various decision support applications based on quantitative methods of statistical analysis or forecasting (e.g., correlation, clustering, scenarios, risk analysis, etc.); B) the portion of DSS Tools more oriented towards business simulation (e.g., what-if analysis, goal seeking analysis, etc.) closer to software tools The Software Tools, instead, consist of: • a wide range of “pre-packaged” (“push”) access and visualization tools, graphical or tabular for directional data (EIS toolkit, aimed at constructing company dashboards), OLAP tools (On-Line Analytical Processing), such as the multidimensional analysis of a business phenomenon/fact, or tools for official, standard, and predefined reporting (enterprise reporting), or tools for georeferencing and cartographic visualization of commercial and marketing data. Sometimes these software tools provide “templates” or “content” (pre-packaged structures and models of indicators or analyses), for specific business functions (e.g., sales, marketing, personnel, etc.) or for economic sectors (e.g., banking, chemical, etc.), to be parameterized according to company needs, which can serve as guidance in the logical setup of control and analysis information systems • query, reporting, analysis, and free “navigation” (“pull”) tools within directional data, aimed at producing ad hoc, highly personalized and targeted information (Query, reporting & analysis tools); generally the software tools available on the market have both working environments (Push and Pull), complementary and well integrated, the first more structured and guided by specialists, the second freer and manipulable by the user directly To produce the necessary managerial information, heterogeneous data from various internal and external company sources must be transformed into consistent information. The technique that deals with this transformation is called “Data Warehousing” (data storage), which involves mapping the sources of necessary elementary data, cleaning and transforming them, and a special modeling of the data itself performed based on the marketing phenomena and facts that one wishes to observe, analyze, and on which decisions must be made. Datawarehousing is the link between data, applications, and operational and transactional information systems and between managerial information systems supporting control and decision-making activities in marketing, also known as “Mktg Knowledge System”.