The multiphase rotodynamic pump is crucial for the multiphase blending transportation process. However, desigining such pump with high hydraulic performance and good blending transportation ability is quite challenging because of its unique structure. This study constructed a multi-objective optimization design system, which combines the 3D inverse design theory, response surface methodology, genetic algorithm, computational fluid dynamics simulation, and multi-objective optimization strategy, to improve the multiphase pump performance. Five optimization variables, namely the impeller loading parameters ( m1,S, m1,H, kS, kH) and the high-pressure edge angle ( θ), were selected in this study. Furthermore, two optimization objectives were considered at inlet gas void fraction ( IGVF) = 10%, namely the pump efficiency and the gas uniformity at the impeller outlet. The optimization results demonstrated that the pump efficiency and blending transportation ability would improve if the impeller blades had for-loaded loading distribution and the negative high-pressure edge angle. After optimization, the pump efficiency and the gas uniformity for the optimized impeller Opt1 improved by 2.6% and 18.2% at IGVF = 10%, respectively. Moreover, the internal flow pattern in the impeller was improved significantly, resulting in the maximum amplitude of pressure fluctuation in impeller Opt1 being 0.42 times that in the original test impeller.