Taking the turning parameters of 1Cr18Ni9Ti stainless steel as the research object, the turning test matrix was established based on the central composite method, the influence of the turning parameters on the surface roughness was analyzed by using the extreme value method, then the optimal combination of turning parameters that minimized the surface roughness was obtained. By using the response surface method, a surface roughness prediction model was built, the variable components in the model were adjusted based on significance analysis to improve the accuracy of the model. JMP software was used to optimize the machining parameters for surface roughness targets commonly used in engineering. The results show that the roughness prediction model based on the response surface method can be used to accurately predict the optimal turning parameters of the set target, and it has more practical application value than the extreme value method.