Supporting multidisciplinary vehicle modeling : towards an ontology-based knowledge sharing in collaborative model based systems engineering environment.

Summary Industrial systems (automotive, aerospace, etc.) are becoming more and more complex due to economic and ecological constraints. This growing complexity imposes new constraints on development. The question of mastering the analysis capacity of their architectures is then raised. To solve this issue, modeling and simulation tools have become a common practice in industrial environments afin order to compare multiple candidate architectures. These simulation tools have become essential to support decisions. Yet, the implementation of physical models is increasingly complex and requires a specific understanding of each simulated phenomenon as well as a thorough description of the system architecture, its components, and the connections between components. The objective of this thesis is twofold. The first concerns the development of a methodology and the tools needed to accurately build simulation models of the system architectures one wishes to study. The second is concerned with the introduction of an innovative approach for the design, production and integration of simulation models in a "plug and play" mode afin order to guarantee the conformity of the results to the expectations, especially at the quality and maturity levels. To accomplish these goals, model-based systems engineering (MBSE) methodologies and processes as well as information systems have been used. This thesis work proposes for the first time a detailed process and tool for the design of simulation models. A common repository named "Identity Card Model (ICM)" has been developed to standardize and strengthen the interfaces between the business and the suppliers on organizational and technical levels. MIC ensures the evolution and management of the consistency of the set of rules and the specifications of the knowledge of business domains with multiple semantics. MIC also reinforces model consistency and reduces anomalies that can interfere during the so-called IVVQ phase for Integration, Verification, Validation, Qualification. Finally, afin order to structure the simulation model design processes, the work drew on Enterprise Architecture frameworks by reflétant the integration and standardization requirements of the enterprise operating model. To validate the concepts introduced in this thesis, case studies from the automotive and aerospace domains were conducted. The objective of this validation is to observe the significative improvement of the current process in terms of efficiency, reduction of ambiguity and misunderstandings in the modeling and simulation of the system to be designed.