Business Intelligence Indicators: Types, Models and Implementation

Business Intelligence Indicators: Types, Models and Implementation

Sandro Bimonte (IRSTEA, Aubière, France), Michel Schneider (University of Clermont-Ferrand, Aubière, France) and Omar Boussaid (ERIC Laboratory, University of Lyon 2, Lyon, France)
Copyright: © 2016 |Pages: 24
DOI: 10.4018/IJDWM.2016100104


Nowadays, more and more data are available for decisional analysis and decision-making based on different indicators. Although different decision-making technologies have been developed, the authors note the lack of a conceptual framework for the definition and implementation of these indicators. In this paper, they propose a first classification of these indicators. Furthermore, motivated by the need for formalism for the definition of these indicators at a conceptual level, they present the Business Intelligence Indicators (BI2) UML profile to represent indicators for OLAP, OLTP and streaming technologies. They also present their implementation in existing industrial tools. In addition, they show how these indicators can coexist in the same environment to exchange data through a chaining model and its implementation.
Article Preview


In recent years, new data acquisition technologies have been developed allowing collecting huge amounts of information (Big Data) (Chen, 2012). It concerns varied data (spatial data, multimedia data, etc.) coming from social networks, sensor networks (mobiles), satellites, etc (Shekar, 2012). In this overwhelming quantity of data, several methods for storing and querying data have been proposed (Cuzzocrea, 2013). New analysis capabilities are available to decision-makers who have a wide variety of data and decision-making tools (Chen, 2012). Decision support systems (DSS), or Business Intelligence (BI) tools, including Data Warehouses (DWs) and OLAP, reporting, data mining, data streaming, etc. Therefore, the establishment of a DSS is no longer confined to a single category of decision-making indicators today. Decision makers may rely on several methods, and therefore several BI technologies.

On the other hand, conceptual modeling is recognized as an essential element for the success of BI projects (Torlone, 2003). Conceptual design allows designers to define models easily understandable by decision-makers. In this context, UML is considered as the object-oriented standard for modeling various aspects of software systems. Indeed, UML provides a powerful tool for formalization to designers and decision makers, during the development and the implementation phases. It can be also interpreted by commercial CASE tools.

In this paper we do not investigate the way the conceptual (multidimensional) model is obtained. In the context of OLAP and database systems, there exist different methodologies for designing a multidimensional model: user-driven methodology based on decision-makers’ requirements, data-driven methodologies based on data, mixed that combine user and data driven methodologies (Romero, 2009). Our purpose is to propose a UML profile which helps to model and implement indicators once the multidimensional model is obtained. We define eight classes of indicators. Existing works permits to deal only with one class: the one of OLAP indicators.

The decisional indicators in which we are interested are under the form of aggregate queries (e.g. the average sales per month and region). This type of query is essentially managed by DWs and OLAP systems, because data is organized according to the multidimensional model that defines the concepts of dimensions (analysis axes) (e.g . time and location) and facts (subjects of analysis) (e.g. sales) (Kimball, 1996). Warehoused data are aggregated along the dimension hierarchies (e.g. day, month, year) with aggregate functions (e.g. min, max, sum, etc.). Different conceptual formalisms have been proposed for DWs (Boulil, 2015).

Complete Article List

Search this Journal:
Open Access Articles: Forthcoming
Volume 15: 4 Issues (2019): Forthcoming, Available for Pre-Order
Volume 14: 4 Issues (2018): 2 Released, 2 Forthcoming
Volume 13: 4 Issues (2017)
Volume 12: 4 Issues (2016)
Volume 11: 4 Issues (2015)
Volume 10: 4 Issues (2014)
Volume 9: 4 Issues (2013)
Volume 8: 4 Issues (2012)
Volume 7: 4 Issues (2011)
Volume 6: 4 Issues (2010)
Volume 5: 4 Issues (2009)
Volume 4: 4 Issues (2008)
Volume 3: 4 Issues (2007)
Volume 2: 4 Issues (2006)
Volume 1: 4 Issues (2005)
View Complete Journal Contents Listing