Modularization of multi-physics complex systems
Résumé
This paper addresses the modularization of multi-physics complex systems. The aim of this research focuses on optimization methods useful for complex systems integration in the context of aircraft industry. The type of product concerned introduces many criteria to be achieved and many constraints to be respected. The principle of modularisation systems can be approached via two different ways. In this paper we will study the fundaments of the approach. The use of these solutions is strongly influenced by the state or maturity level of the product design to modularize, and by the level of abstraction desired. In the case of an aircraft whose definition is successfully completed, or well advanced maturity, namely that the components are defined and implemented and the system architecture is already established, we will address methods of modular decomposition. In the case of designing new products, innovative or not, so with a definition not started or less advanced, we strive to speak of modular design upstream. The objectives that motivate the industry to move to a modular design or modular decomposition of their products are a strong ramp-up, elements interchangeability (standardization, reparability) and maintenance as well as time savings in the assembly process. The common methods are more oriented in modular decomposition and not in modular design. They are mainly based on tools decomposition and organization of data such as QFD, DFM, Functional Flow Analysis, or Design Structure Matrix. Further upstream, methods using the modular design are not yet well defined, but methods coupling DSM tools and Axiomatic Design principles or methods based only on DSM tools are already interesting solutions. The paper proposes a study of these two approaches based on a literature analysis, highlights the possibility to apply them in aeronautics