Adaptive Multi-Agent Systems for Multidisciplinary Design Optimisation

International audienceThis article presents a multi-agent method to tackle the difficult task of multidisciplinary optimisation, based on the notions of cooperation and self-regulation. It is focused on the preliminary aircraft design process which demands complex compromises between many actors. In our approach several cooperative agents collectively act to achieve a common goal, i.e. optimising a multi-objective function, even if the environment of the system (the user's requirements) changes during the solving process. In MASCODE, one agent encapsulates one discipline and is designed individually without considering the dependencies with the others. So the computation is conceptually distributed without central control. Experimental results are provided and efficient in comparison to the classical FSQP method. But designing an aircraft is a complex process, and thus we show that adaptive behaviour of MASCODE provides new capabilities to understand and to manage the preliminary aircraft design