20CrMnTi is a kind of material widely used in gear manufacturing. In order to improve the surface quality and processing efficiency of 20CrMnTi finishing, the surface roughness of 20CrMnTi was studied. The orthogonal test was designed to carry out the hard alloy tool turning test on the CNC lathe GENOSL250E. The effect of cutting parameters (cutting speed, feed rate, cutting depth) on surface roughness was explored. The relationship model between cutting parameters and surface roughness was established by using multiple regression method, and a multiobjective optimization model with processing efficiency and surface roughness as objects was constructed. The optimal solution was obtained by using particle swarm optimization.The test results show that the use of optimized cutting parameters can reduce surface roughness and improve processing efficiency.