La conception de systèmes mécaniques, de par son aspect pluridisciplinaire et technologique, fait intervenir et interagir différentes personnes qui travaillent et prennent des décisions ensemble, et, participent ensemble à l’élaboration du produit. Elles travaillent de manière collaborative cependant elles ne se connaissent pas obligatoirement, ne se situent pas forcément géographiquement sur un site commun, n’ont peut-être pas la même culture et n’appartiennent pas systématiquement à la même entreprise. La conception préliminaire représente les premières phases du cycle de conception ou le produit est en cours de définition. Le nombre d’incertitudes sur les paramètres et les informations produit sont très importantes. Il y a un manque de connaissances important à cette étape du processus de conception qui doit être considéré afin d’améliorer et d’aider les prises de décisions dans les phases amonts. C’est ce manque de connaissances que je me propose de qualifier et caractériser en apportant une réponse à la question résultante: comment prendre en compte le manque de connaissances pour prendre des décisions durant la conception préliminaire collaborative ? Pour se faire, nous proposons un méta-modèle de connaissances permettant de structurer les informations du produit et les connaissances en intégrant la maturité du produit. Cette maturité est définie par une métrique et permet d’identifier le niveau de connaissances des concepteurs sur le produit et d’orienter la prise de décision grâce à l’utilisation d’une approche mixte, à la fois qualitative et quantitative. Enfin, nous évaluerons la capacité de ce méta-modèle à générer différent modèles produit, puis sa pertinence avec l’implémentation sur un cas industriel. / The design of mechanical systems, due to their multi-disciplinary and technological aspects, involves different people who, together, work and make decisions and jointly participate in the development of the product. They work in a collaborative manner; however, they may have different strategies, geographical positions, cultures and do not know the other members of the team. Preliminary design represents the early stages of the design cycle or product definition. A number of uncertainties regarding the parameters and product information are very important. There is an important lack of knowledge at this stage of the design process that must be managed or filled in order to improve and support the decision making in the early phases. It is this lack of knowledge that I propose to qualify and characterise, providing an answer to the question: how does one to take into account the lack of knowledge in decision making during the preliminary design collaboration? To do so, we propose a meta-model for structuring product information and knowledge by integrating product maturity. A metric allows this maturity to be defined, to identify the level of knowledge of the product designers and to guide the decision making, thanks to the use of a qualitative and quantitative approach. Finally, we evaluate the ability of the meta-model to generate the different models produced and its relevance to the implementation in an industrial case.

The design of mechanical systems, due to their multi-disciplinary and technological aspects, involves different people who, together, work and make decisions and jointly participate in the development of the product. They work in a collaborative manner; however, they may have different strategies, geographical positions, cultures and do not know the other members of the team. Preliminary design represents the early stages of the design cycle or product definition. A number of uncertainties regarding the parameters and product information are very important. There is an important lack of knowledge at this stage of the design process that must be managed or filled in order to improve and support the decision making in the early phases. It is this lack of knowledge that I propose to qualify and characterise, providing an answer to the question: how does one to take into account the lack of knowledge in decision making during the preliminary design collaboration? To do so, we propose a meta-model for structuring product information and knowledge by integrating product maturity. A metric allows this maturity to be defined, to identify the level of knowledge of the product designers and to guide the decision making, thanks to the use of a qualitative and quantitative approach. Finally, we evaluate the ability of the meta-model to generate the different models produced and its relevance to the implementation in an industrial case.