Modèle d’archivage d’entrepôts de données multidimensionnelles

Data warehouses are commonly used for decision making. Data integrated by these systems is, on the one hand, periodically updated and permanently stored on the other hand. When data is historized, decision-makers are usually less interested in old information, but still integrating it into their analysis. In this paper, we present a new mechanism for data archiving and summarizing across the time. This would satisfy evolving needs expressed by decisionmakers. In particular, we propose a conceptual model that represents multidimensional data archiving. Moreover, we propose a logical model and a set of transformation rules to automatically generate the logical model from the conceptual one. MOTS-CLÉS : Modélisation multidimensionnelle des entrepôts , Archivage de données, Transformation QVT