Abstract: The mass of a machining center’s moving parts is critical to the machine's dynamic characteristics, machining accuracy and energy loss. Propose to take the moving parts of machining centers as the research object, explore the relationship between their mass and the dynamic characteristics of the entire machine, and obtain the optimal configuration of the moving parts’ mass based on multi-objective particle swarm optimization algorithm (MOPSO). Firstly, a parametric three-dimensional model of machining center is established; secondly, a high-precision response surface model between the bed mass m₁, column mass m₂ and table mass m₃ and the first three orders of the machine's intrinsic frequencies f₁, f₂, f₃ is established by combining the design of experiments and the method of least squares; and then, a multi-objective particle swarm optimization algorithm is written based on the Matlab platform to obtain the best Pareto solution for the mass of each moving part by taking the first three orders of the machine's intrinsic frequencies, f₁, f₂, f₃, as the optimization objectives. Finally, the mass of the optimized moving parts is used as the basis for the secondary design, which provides guidance for the design of the whole machine scheme. The results show that after mass matching optimization, the first and second natural frequencies of the whole machine were increased by 45.58% and 21.43% respectively, while the total mass of the moving parts of the machine was reduced by 20.34%. The maximum response amplitude was reduced by 69.91%.