Balancing aesthetic design and aerodynamic performance is an important task in automotive body development. In our previous study [1], we analyzed the proportional characteristics of production vehicles using principal component analysis (PCA). The resulting principal component scores were then used as design variables for morphing-based shape transformations. This approach enabled aerodynamic optimization under styling constraints. However, real-world vehicle development requires additional physical performance considerations. To satisfy this requirement, we propose a two-stage optimization framework and demonstrate its effectiveness through a case study. The first stage performs multi-objective aerodynamic optimization while maintaining design characteristics, using a genetic algorithm. The second stage introduces local shape modifications to further improve another performance criterion, employing Efficient Global Optimization (EGO) to minimize computational cost while refining the design. In the case study, we used a simplified model of a production SUV as the target vehicle. The first stage aims to minimize drag, lift, and drag increase under a crosswind condition in order to enhance fuel efficiency and drivability. Pareto-optimal solutions obtained from this stage represent aerodynamically efficient yet stylistically constrained shapes. In the second stage, we demonstrate additional refinement by minimizing aerodynamic noise, using the intensity of pressure fluctuations on the side windows as an approximate indicator. This sequential approach demonstrates the feasibility of integrating design aesthetics with multiple physical performance targets in a practical vehicle development context. The case study confirms that the combination of global morphing-based optimization with localized refinement can effectively address aerodynamic efficiency, noise reduction, and styling constraints.