A Model-Driven Approach to align Heterogeneous Models of a Complex System.
Résumé
To understand and manipulate a complex system, it is necessary to apply the separation of concerns and produce separate models, called viewpoints models. These models represent views on the system that correspond to distinct business domains. They are generally heterogeneous, i.e. conform to different meta-models. The management of the system's global model (a complete view of the system) requires the identification of the existing correspondences among the viewpoints models. However, in practice these correspondences are either incompletely identified or not sufficiently formalized to be maintained when models evolve. This restricts their use and does not allow their full exploitation for managing the global model. To fix this issue, we propose AHM (Alignment of Heterogeneous Models), an approach to organize the source models as a network of models through a virtual global model called M1C (Model of correspondences between models) that conforms to a Meta-Model of Correspondences (MMC). AHM proposes also a process, allowing for both the creation of the global model, and its consistency control. Partial automation of this process is done through a refining mechanism supported by a semantics expression described in a Domain Specific Language (DSL). The application of AHM is illustrated by the example of a conference management system. A prototype of a tool called Heterogeneous Matching and Consistency management Suite (HMCS) has been developed to support this approach.
Origine : Fichiers produits par l'(les) auteur(s)
