The aim of this paper is to optimize the performances of an Interior Permanent Magnet Synchronous Machine (IPMSM) for an electric vehicle application. The originality of this work concerns the use of a Reluctance Network (RN) method which furnishes the harmonics of output parameters (i.e. torque and back-emf) and automatically generates the derivative of the model. Then, precise and very fast optimizations are performed own to gradient method as SQP. Results are also compared to Finite Element Analysis (FEA) and several mono-objective optimizations are presented. Finally, the proposed method helps engineers to have a better sense of machines behavior and to quickly optimize the geometry of their machines during preliminary design process