Laser assisted hybrid machining being researched in past decade on various difficult to machine materials to improve the machinability. Predictive modeling approaches such as response surface method (RSM) and artificial neural network (ANN) are widely applied for model development. However, no reported work using RSM and ANN approaches to predict the relationship between the experimental variables (speed, feed, laser power and beam apporach angle) on surface roughness Ra (μm). Furthermore, coefficient of correlation (R2), root mean square error (RMSE) and model predictive error (MPE) are considered as a performance measures for their effectiveness. The results show that the ANN model estimates the machinability indices with high accuracy with a limited number of experiments compared to the response surface model. From the comparative study, ANN model is found to be capable for better prediction of response than the RSM model. ANN model provides a maximum precision benefit of 10% for surface roughness Ra (μm) compared with RSM model. Also the calculated Pearson correlation coefficient showed a robust relationship between the laser beam angle and Ra, surface roughness followed by the speed. © 2017 Elsevier Ltd.