Worn tool geometry when reaches a critical state, has a significant effect on machined surface quality. Identification of the optimal tool life so that the surface quality is kept at a desirable level is an essential task especially in machining of hard materials. Unfortunately, this approach has not been developed enough in literature. In this paper, an experimental study and intelligent methods were used to identify the optimal tool life and surface roughness in turning process of the Inconel 718 alloy. At first, the effect of the machining time at the different cutting parameters (including depth of cut, feed rate and cutting speed) was extensively investigated on the surface roughness using the Artificial Neural Network (ANN) model trained by the optimization algorithm. Then, the modified Non-Dominated Sorting Genetic Algorithm (NSGA-II) was developed to simultaneous optimization of tool life and surface roughness. For this purpose, a new approach was implemented and the machining time was taken into account as both input and output parameter during the optimization. Finally, the results of optimization were classified and the optimal states of the tool life and surface roughness were found. The results indicate that implemented strategy in this paper provides an efficient approach to determine the desirable criterion for tool life estimation in machining processes.