4. Business Intelligence (BI), Technology and Information Architecture

Business Intelligence (BI) can be conceptualized as a set of information systems that support decision making, based on other data storage, analysis, and data mining technologies. However, there are theoretical gaps in some aspects of the concept that best defines Business Intelligence as a technological tool. Suppliers of IT solutions hold a variety of views in relation to the operational scope of these systems and their emergence. BI is geared towards obtaining information using data stored in transactional databases. In this respect, the aim of a BI system as a technology is to create a structure that transforms data registered within the organization into information useful in decisionmaking, thereby creating value for the company.

Information is structured via fact and dimension tables into a data storage component. Dimension tables hold descriptive data reflecting an aspect of the business, that is, they help to define a component via information. Examples of dimension tables are: products, time, and brand. These tables are described by attributes, that is, characteristics that define them. For example, the dimensionproduct can be identified by the attributes color, weight, size, and unit price, while the concept of fact is related to the storage of numerical measurements for the business, that is, a factual measurement associated with a dimension. For example, the dimension product may be associated to a fact,sales, which is also related to the dimensiontime. In this hypothetical relationship, an analysis could be conducted to obtain the monthly sales for a product.

Business Intelligence consists of three basic components: a) data storage, in data warehouses (DW) and data marts (DM); b)analysis tools, called On Line Analytical Processing (OLAP), that allow navigation among information; c) data mining(data mining), which enables the extraction and discovery of patterns of information in data and specific cases.

The implementation of BI technology provides better results, however, when aligned with two elements. First, it should be based on the information needs of executives, enabling them to centralize information regarding the critical success factors of an organization. According to Petrini, Pozzebon, and Freitas, another important issue is that implementation should be based on supporting executive decisions, that is, on information relevant to the business and indicators created in accordance with strategic objectives. Barbieri suggests that BI also incorporate balanced scorecard indicators, vital for extracting strategic information from transaction processing systems.

In order to achieve the desired results, it is essential that technological tools be used to construct information architecture that supports decisions. As per Tupper, information architecture supports executive tasks and it supplies specific numbers, reports, access to data and forms, allowing learning to occur regarding the products and services of a company, thereby improving business performance based on facts and dimensions. Figure 2 depicts a frameworkwith the Technologies and components of an information architecture system.

Figure 2 Framework of Information Architecture.

According to Barbieri, strategic business plans (SBP), the IT plan (ITP), and balanced scorecard (BSC) must be aligned. Business Intelligence (BI) technology should provide the data required for the plans and indicators of the BSC as the core technology, since it defines the structures of information needs that can be stored in the system. The SBP has been considered an administrative technique that creates awareness of certain elements among the entire organization: strengths, weaknesses, goals, targets, threats, and opportunities, among others. The ITP, on the other hand, refers to the need for strategic, tactical, and operational structuring of "IT and its resources: hardware, software, telecommunications systems, data and information management".

Construction of an information architecture system should follow a gradual process of analysis and development, based on what Inmon calls the "data warehousing process". This allows the use of a database stage for extracting, transforming, and loading data (ETL process). Data warehousing can be performed using source systems, either the organization's own integrated system or external systems, such as: electronic spreadsheets, workflow systems, and information obtained online.