Abstract: Supersonic civil aircraft have become one of the important directions for the future development of civil aviation. For the rapid conceptual design and performance optimization of supersonic civil aircraft layouts, an aerodynamic optimization framework for transonic and supersonic dual-cruise regimes was developed using a medium-order solver based on the Euler equations with boundary layer correction to ensure computational efficiency while capturing key flow features. Using this framework, the baseline configuration was optimized, and the optimization results were selected from the Pareto optimal solution set in the dual-speed domain. Building on this, a high-precision low-drag adjoint optimization based on RANS equation for the wing-body combination was carried out under supersonic conditions, resulting drag reductions of 38.3 counts at Ma = 0.9 and 54.1 counts at Ma = 1.8. To further address the multidisciplinary design challenge of simultaneously minimizing both drag and sonic boom intensity, a prediction method combining ray tracing technique and augmented Burgers equation was employed to simulate the propagation of near-field sonic boom signals. The ground waveforms were analyzed using the Mark VII criteria developed by Stevens to calculate the perceived loudness in deciBel (PLdB) of the sonic boom. A multidisciplinary optimization framework considering both aerodynamic performance and sonic boom signals at different azimuthal angles was established to achieve a low-drag and low-sonic boom design. The optimization results showed that, compared to the baseline configuration, the final shape achieved comprehensive improvements in both drag and sonic boom performance throughout the entire sonic boom carpet region: the drag coefficient was reduced by 30.6 counts at Ma = 0.9 and by 50.3 counts at Ma = 1.8, while the average ground-level perceived loudness in deciBel was reduced by 4.84 PLdB.