You are in the accessibility menu

Please use this identifier to cite or link to this item: http://acervodigital.unesp.br/handle/10400.2/1290
Title: 
Business intelligence no suporte a decisões sobre comunicações : descrição de um caso
Author(s): 
ISSN: 
1646-6330
Abstract: 
O projecto descrito tem o objectivo de apoiar decisões de investimento em infraestruturas de comunicação na Electricidade Dos Açores (EDA), a empresa responsável pela geração, transporte e venda de corrente eléctrica na Região Autónoma dos Açores. A decisão imediata a apoiar consistia em saber se as comunicações entre ilhas deveriam passar para tecnologias Voice over IP (VoIP), um serviço actualmente contratado em regime de outsourcing. Foi estabelecido um projecto de business intelligence, usando tecnologias OLAP do Microsoft SQL Server, para ler e pré-processar os ficheiros CSV de grande dimensão, combinar esses dados com bases de dados existentes e apresentar os resultados sobre a forma de cubos multidimensionais. Posteriormente, implementaram-se igualmente algoritmos de data mining, integrando na metodologia CRISP-DM as duas técnicas utilizadas. Construindo vários modelos foi possível, além de apoiar a decisão pretendida, identificar situações ineficientes e mesmo fraudulentas. Os modelos construídos foram ainda disponibilizados aos decisores estratégicos e de controlo, assim como toda a estrutura de reutilização, manutenção e realimentação que suporta o OLAP e os modelos de data mining. Abstract:This project addresses decisions of investment on communication infrastructures in Electricidade dos Açores (EDA), the local Electric Company in the Azores Islands. The main decision was that EDA communications should be moving to Voice over IP (VoIP) from present telephone lines, outsourced to an external communications company. At the beginning, a business intelligence project was set, with the objective of getting data from the communications company and analyzing it in order to offer useful information to decision makers. The system uses Microsoft SQL server technologies to establish an OnLine Analytical Processing (OLAP) application. It translates big CSV flat files in a ROLAP infrastructure and presents the results as multidimensional data cubes. Latter some data mining models were implemented and both techniques were incorporated in the CRISP-DM process model. Different models identified several inefficient procedures and even fraud situations as long as supporting the investment decision. These models as long as all the technology developed for gathering data, maintain an manage the OLAP cubes and data mining models were made available to control and strategic decision makers.
Issue Date: 
2008
Citation: 
Mendes, Armando B.; Alfaro, Paulo Jorge; Ferreira, Aires - Business intelligence no suporte a decisões sobre comunicações : descrição de um caso. "Revista de Ciências da Computação" [Em linha]. ISSN 1646-6330. Ano 3, vol. 3, nº 3 (2008), p. 23-37
Publisher: 
Universidade Aberta
Keywords: 
  • Decision support systems
  • Data mining
  • OLAP
  • Business intelligence
  • Telecomunicações
  • REA
URI: 
Rights: 
openAccess
Type: 
outro
Source:
http://repositorioaberto.uab.pt/handle/10400.2/1290
Appears in Collections:Recursos Educacionais Abertos (REA) - Universidade Aberta de Portugal

There are no files associated with this item.
 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.