Handling Domain Knowledge in Formal Design Models: An Ontology Based Approach

Excerpt

During the development of complex systems, several models corresponding to different analyses are built. These analyses refer to different engineering domains. In general, the knowledge carried out by an engineering domain is not explicitly taken into account during system development process. The system development process leads to the production of several heterogeneous models corresponding to different views or analyses of the same system. In this context, the most important heterogeneity factor, in addition to the one due to the use of different modelling languages, is related to information, knowledge and assumptions of the domain (the environment and context where designed systems evolve). In general, this knowledge is not explicitly formalised and therefore not used in the models associated to the design of these systems. One of the reasons is the absence, in the modelling languages, of resources to model such domain knowledge. As a consequence, very few domain knowledge information is handled by the designed models. The developer handles this information by himself in the development process. Indeed, although systems are developed in accordance with development standards and good practices, a large part of the knowledge required for the interpretation and validation of these models of systems remains implicit. This situation may lead to inconsistent model verification and/or validation and models may loose some of their properties if the information related to their domain, context and environment are integrated to the models. …