Abstarct: Due to the costs of drilling tool wear in tunnel construction projects, the growing need to provide a comprehensive and practical model for estimating and predicting wear rates has been felt more than ever. Many researchers, following the importance of abrasion and its effect on drilling efficiency and time, have tried to estimate and predict the wear rate using experimental and semi-theoretical models. Due to the fact that these models have been developed in specific geological conditions and also the impact of all parameters affecting the wear rate in these models has not been considered, they can not be generalized to all geological conditions. The shortcomings of these methods on the one hand and the growth of science in the fields of artificial intelligence and neural network on the other hand, have led researchers to use and study the application of this method compared to previous methods. Neural networks are a good idea for modeling when the relationship between the input and output parameters of a quantity is not clear. In this study, the effect of rock and machine parameters on disk cutter life prediction models (CSM, NTNTU and Gehering) has been investigated. The results show a different effect of each of these parameters on the models of disc cutter wear, which shows the importance of choosing a suitable model according to the geological conditions of the region. In the following, using the neural network method, a model for estimating and predicting the life of the disc cutter and comparing it with other common models for predicting the life of the disc cutter is presented. After determining the parameters affecting the wear rate and creating a suitable structure of the neural network, the wear rate in the Uma Oya and Kerman water transfer tunnels has been predicted. Considering the high coefficient of determination of 0.6 and the average error obtained for the neural network model and its comparison with CSM, NTNTU and Gehering methods, it shows the proper efficiency of the neural network in predicting the rate of disc cutter wear in tunnels with high abrasivity class. But in tunnels with low abrasivity class, more investigation is needed.