Modélisation logique des données dans les systèmes multidimensionnels d'aide à la décision: la méthode MAP

Abstract

ABSTRACT Among the tools available to the designers of Decision Support Systems (DSS) and Executive Information Systems (EIS), there are generators, ad—hoc systems, groupware and end—user products based on fourth—generation languages. Most of these tools are characterized by a logical structure based on the use of multidimensional tables. These DSS and EIS softwares permit to define the structure of the database and to manipulate it in a user—friendly fashion with an exceptionally short response time. They are based on a specific logical data model, consisting of dimension, variable and relationship concepts. Based on the OLAP technology (One Line Analytical Processing), these tools allow to express multidimensional views using the client—server architecture. Starting from a real situation, the designers of DSS and EIS often have difficulty in defining directly the dimensions, variables and relationships. To solve this problem, we suggest to proceed in two steps. The first step consists in modelling at the conceptual level, using the extended entity—relationship approach. The second step aims at moving to the logical level, by means of transformation rules. To this end, we present an algorithm which transforms a conceptual schema into a logical schema based on the objects of the multidimensional model. We illustrate the use of the transformation rules by means of an example and present the tool associated with the method.