A model-driven approach for enforcing summarizability in multidimensional modeling

Abstract

The development of a data warehouse system is based on a conceptual multidimensional model, which provides a high level of abstraction in the accurate and expressive description of real-world situations. Once this model has been designed, the corresponding logical representation must be obtained as the basis of the implementation of the data warehouse according to one specific technology. However, there is a semantic gap between the dimension hierarchies modeled in a conceptual multidimensional model and its implementation. This gap particularly complicates a suitable treatment of summarizability issues, which may in turn lead to erroneous results from business intelligence tools. Therefore, it is crucial not only to capture adequate dimension hierarchies in the conceptual multidimensional model of the data warehouse, but also to correctly transform these multidimensional structures in a summarizability-compliant representation. A model-driven normalization process is therefore defined in this paper to address this summarizability-aware transformation of the dimension hierarchies in rich conceptual models.