Extending the dimensional templates approach to integrate complex multidimensional design concepts

Abstract

In the past, several approaches have been devised to semi- automate the multidimensional design (MDD) of Data Warehouse (DW) projects. Such approaches highly contribute to more expertise- independent and deterministic MDD results. Among them, only the Dimensional Templates Approach (DTA) focuses on solving the critical resource containment problems of DW prototypes. Originally, the DTA allows solely the generation of basic MDDs. In this paper, we depict an extension to address complex MDD issues. These include (i) date/time hierarchies, (ii) many-to-many relationships, (iii) hierarchically structured data and (iv) coverage facts. The proposed enhancements, including a rebuilt generation algorithm, allow more accurate and broadening results than the original DTA. Throughout the paper, references are made to a real case study to which the improved DTA has been applied using two developed prototype tools. © 2011 Springer-Verlag.