This paper describes a modeling study that was conducted to estimate the two output responses represented by the resultant force and the surface roughness obtained during the dry turning of GG 25 grey cast iron using a coated CBN (Cubic Boron Nitride) cutting tool. The modeling is based on the experiments that were planned using a Taguchi L27 design. The analysis of variance (ANOVA) approach is used to determine and classify the cutting parameters, which are expressed as cutting speed, feed and depth of cut that influence the output responses. To establish the prediction models, the response surface methodology (RSM) is empolyed. The results demonstrate that the CBN cutting tool is very efficient for the machining of grey cast iron, with minimal surface roughness (Ra = 0.7µm). The mathematical models derived by the RSM approach have been shown to be quite useful for the optimization process. Furthermore, there was a strong correlation between predicted and measured values, with (R²) varying between 0.95 and 0.98. Finally, the Genetic Algorithm (GA) method was used to successfully accomplish multi-objective optimization task.