Quantitative analysis of data warehouse design quality

Abstract

Information systems allow companies to collect a large number of operational and transactional data. Data warehouses are increasingly used by organisations to extract concise information supporting decision processes. However, data warehouse design strategies as well as the structure and content of the original database largely influence the effectiveness of such tools. Our research is focused on data-driven approaches, and in this paper, we present a solution that, based on a set of metrics measuring different characteristics of the original data sources, effectively supports the creation of data warehouses. Moreover, combining the indexes computed by our metrics on selected dimensions and measures, it is also possible to derive quantitative information on the design quality of the final data warehouse. To demonstrate the effectiveness of our solution, we briefly present the results obtained from the analysis of a selling data warehouse derived from a real-word ERP system.