A two-level layout optimization method for metal stiffened thin walled cone is proposed. The sample points of layout variables are created by design of experiment (DOE). Then for each stiffened thinwalled structure corresponding to each sample point, the size optimization is performed under the constraints of strength and stability, therefore the corresponding target response value is obtained. Then the Kriging approximation model is built based on the layout variable sample points and the corresponding response surface. Finally, the approximation model is optimized by using the genetic algorithm and the optimal solution of the stiffened thinwalled cone is presented. The example results show that the method is simple and has high optimization efficiency.