Three dimensional, steady state computational fluid dynamics (CFD) simulations of flow around a generic pickup truck are performed to optimize the aerodynamic performance of a pickup truck model. Detailed comparison between the data of the CFD model and the experiment are made. By using deformation techniques, surrogate models and optimization methods, the drag is reduced. Four design variables are used for deformation: the cabin height, bed height, ground clearance and bed length. The optimization is single objective: minimizing the drag coefficient. A response surface model is built to reduce the sampling points for optimization, and the simulation time is reduced accordingly. Results show that the design variables are not fully independent with each other, and by proper combinations of the variable change, the drag coefficient of the pickup truck model can be reduced effectively. In this study, the drag coefficient reduced about 9.7% through optimization algorithm. The results also show that the single tailgate itself is not always profitable for drag reduction. This paper is limited to an existing generic pickup truck model where abundant experimental data can be achieved in publications. The lift coefficient is not considered. Although optimization methods have been used in CFD analysis more frequently in recent years, few publications can be found in aerodynamic optimization of the pickup truck model. The methodology in this paper is both effective and efficient, and the results show a considerable reduction of the drag of the pickup truck model.