In order to improve the dry turning quality of titanium alloy, the response surface methodology was used to optimize the main turning parameters. A BoxBehnken experimental model was developed the cutting speed, cutting depth and feed rate were regarded as the processing parameters,Ra of workpiece surface roughness and VC of tool wear were regarded as evaluation indexes. Variance and fitting residual probability distribution were used to analyze the significance and interaction of three factors. Furthermore, the validity of the secondorder response prediction model of surface roughness and tool wear was verified by experiments. The result shows that the optimum cutting speed, cutting depth and feed rate is 20 m/min, 0.178 8 mm, 0.1 mm/r, respectively. The surface roughness and tool wear obtained by cutting with optimized three parameters, is 1.031 μm, 155.6 μm, The errors are 9.93% and 1.58% respectively compared with predicted value. It is proved that the prediction model of surface roughness and tool wear based on response surface methodology(RSM) is accurate and effective.