Title: Towards a new method for designing multidimensional models

Authors: Nawfal El Moukhi; Ikram El Azami; Aziz Mouloudi

Addresses: MISC Laboratory, Department of Computer Science, Faculty of Sciences Kenitra, Ibn Tofaïl University, P.O. Box 242, Kenitra, Morocco ' MISC Laboratory, Department of Computer Science, Faculty of Sciences Kenitra, Ibn Tofaïl University, P.O. Box 242, Kenitra, Morocco ' MISC Laboratory, Department of Computer Science, Faculty of Sciences Kenitra, Ibn Tofaïl University, P.O. Box 242, Kenitra, Morocco

Abstract: Designing a data warehouse as a support of the decision-making process remains a complex task and a major challenge for companies and organisations. This complexity is principally due to data sources that are heterogeneous and to the absence of a conventional method for modelling the data warehouse. This paper analyses the most important data warehouse design methods; it enumerates key strengths, limitations for each method, and presents a new design approach that allows to construct multidimensional schemas from relational models by using the model driven architecture (MDA) techniques. This new approach will be able to produce good quality multidimensional schemas in an automated and standardised way.

Keywords: decision support system; DSS; data warehouse; data warehousing; relational models; multidimensional models; design methods; designing data warehouses; model driven architecture; MDA.

DOI: 10.1504/IJBIS.2018.091161

International Journal of Business Information Systems, 2018 Vol.28 No.1, pp.18 - 41

Received: 16 May 2016
Accepted: 29 Sep 2016
Published online: 13 Apr 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article