One of the important challenges in the auto industry is to reduce the mass of the vehicle while meeting structural performance requirements for Crashworthiness, Noise, Vibration and Harshness (NVH) etc. In this paper, a multidisciplinary optimization (MDO) of a car back-bonnet is investigated by using the Response Surface Method (RSM). Firstly, a car body is fully surface modeled in CATIA and meshed in HYPERMESH software. Then, modal analysis of the finite element model is performed by NASTRAN software to find natural frequencies. Frequency map of that component is extracted and compared with a reference map to detect defects. Design of Experiments (DOE) methodologies is used for a screening of the design space and for the generation of approximation models using RSM techniques. Therefore, to optimize the model, improvement of the NVH behavior and minimization of the weight are imposed.