CONCEPTUALIZATION

Previously the construction of a Business Intelligence project, it is necessary to clarify some concepts.

1- Business Intelligence (BI)

Since 1980s the term Business Intelligence was used and disseminated by Gartner Group. Companies exploit internal information (available at the data warehouse) and external information (available at the market) in order to evaluate themselves and their competitors, developing growth strategies and assuring profits, client’s satisfaction, decreasing losses and leading to the company’s success.

The self-evaluation process and the study of competitors are not new; the book “Art of war”, written by Sun Tzu (TZU, 2000) presents teachings of how obtain success in a war: you have to be aware about your weaknesses and strengths and about your enemy’s. Another classic example is “The Godfather” trilogy, from Mario Puzo. It deals with the necessity of studying the competitors, constantly (PUZO, 2000). About the study of competitor’s methodology, Kotler affirms that “A company must know about its competitors (…) its goals, strategies, weaknesses, strengths and response patterns”. (KOTLER, 1999)

The BI term is used as a reference for intelligent processes of data collection, organization, analysis, sharing and monitoring in Data warehouses. These processes created information able to support the decision-making procedure inside business environment, providing faster and safer decisions and ensuring the company’s success.

2- Key Performance Indicator (KPI)

Another important concept inside BI project context is KPI. Key performance indicators measure the performance of the company's processes. Based on it the company can work to achieve its goals.

An example of KPI is “Sales By Region”. Using this KPI the company can point the region that is meeting sales objectives and, for example, create strategies for the regions that are under performing.

3- Data warehouse (DW)

Considering that a company deals with loads of data, it is important to select the most relevant data to analyse. Data warehouse (DW) is where the most important company’s data are located. The data are organized and structured, at most times, using a multi-dimensional modelling. According to Inmon (1997), Data warehouse is “an integrated data collection, subject-oriented, time-variant and non-volatile, used to support the decision-making processes.”

4- OLAP Tools

Since data are selected, it is necessary to analyse it. Due to misunderstanding on data analysis tools and data analysis process, the majority of new Business Intelligence system users request reports. OLAP is On-line analytical processing and, according to Inmon (1999), it is a software technology that allows analysts, managers and executives to obtain data in a fast, consisted way, with interactive access to a great quantity of possible views.

OLAP tool is able to run MDX queries and work on what is called Cube. Inside the cubes it is possible to cross data, leading to different views about the same information. In the example below the possibility of crossing data of products, stores and period is shown.

Figure 1: OLAP Cube Source: IT4biz

An OLAP Cube has a fact table and dimension tables. Their definitions are presented above:

  • Fact table: Table where the dimensions and metrics are stored
  • Dimension table: Table where dimension data are stored

Examples of dimension tables are: products, stores, time, suppliers, etc. Examples of hierarchical dimensions: “Client Location”. The following grouping is required: Country -> State -> City.

There are two ways of modelling OLAP Cube: star schema and snow flake schema.

At this moment, focused on the monetary aspect of our project we have to understand a little bit more about the open source world.