Abstract: In order to solve the problem of performance fluctuation caused by manufacturing error of double wishbone suspension, this paper constructs a suspension multi-body dynamics model based on Adams/car, and carries out robustness analysis and optimization based on the ISIGHT platform. Taking the suspension hard-point coordinates as the variable, the 6 Sigma method was used to quantify the sigma level of the key performance indicators of the suspension under the initial design, and the approximate model was constructed by the response surface method to replace the physical simulation, which significantly improved the optimization efficiency, and the hard-point coordinates were optimized with the goal of maximizing the sigma level. The results show that robustness optimization can significantly reduce the influence of manufacturing tolerances on suspension performance, and provide an efficient multidisciplinary collaborative design framework for improving vehicle handling stability.