Wave-plate mist eliminators are widely employed as gas-liquid separation devices to prevent the liquid escaping from thermal power plants or other cooling towers. In this study, the wave-plate mist eliminator with drainage plates was numerically analyzed and the effects between geometrical variables on two objectives, namely, pressure drop (ΔP) and separation efficiency (η), were revealed. Plate spacing, width, and length, as well as the relative position of the drainage plate, were thoroughly investigated. A combined strategy was developed for multi-objective optimization of the wave-plate mist eliminator by integrating computational fluid dynamics (CFD) simulation, response surface methodology (RSM), non-dominated sorting genetic algorithm-II (NSGA-II), and a technique for order of preference by similarity to ideal solution (TOPSIS) method. The results demonstrated that the relative position of drainage plates has a greater impact on the overall performance, whereas the width of drainage plates has the minimum effect. With the implementation of NSGA-II and the TOPSIS method, an optimal solution for the design of the mist eliminator was obtained. After comparing with the baseline case, the optimized case presents promising characteristics with high separation efficiency (enhanced by 3.6% ~9.06%) and a low energy consumption coefficient (reduced by 72.30% at η = 45%).