Agent-Based Approach for Partitioning the Mechatronic Design

Complex mechatronic systems design is a hard task especially if performed by several different designers geographically distributes and using different modeling tools. Therefore, the design tasks should be divided into partitions easier to manage to reduce this complexity. However, the optimization of the overall design requires incorporating the relevant partitions in order to find the optimum mechatronic design. Efficient strategies of partitioning and coordination should be specified at the conceptual level to have a successful optimization process. In this paper we present an agent-based approach to define the design partitions. We use SysML language to describe the agents and we show how to associate optimization attributes to partitions before developing analysis models. The approach is applied to the case of preliminary design of an electric vehicle to illustrate how the use of SysML language with the multiagent approach helps designers in dividing efficiently the complex design of mechatronic systems.