The most important performance indicator of a TBM is the penetration rate, which is defined as the ratio of the drilling distance to the operating time during tunneling. Geological and geotechnical factors, machine specifications, and operating parameters can affect the penetration rate of the machine. Predicting the penetration rate of the drilling machine can significantly reduce the costs of mechanized drilling. Over the past few decades, penetration rate prediction models have emerged successively, which can be broadly classified into three categories: theoretical models and laboratory experiments, and empirical models based on the historical field performance of TBMs. In this study, after component analysis, removing outliers, and considering geotechnical factors and various machine parameters, the penetration rate of the EPB machine in 5 projects was investigated and predicted. For this purpose, linear regression and neuro-fuzzy methods were used. To validate each model, the statistical index of the coefficient of determination (R²) was used. The results of the studies showed that the neuro-fuzzy method has a better accuracy (R²=0.94) in predicting the penetration rate than other methods. Also, the results of the sensitivity analysis showed that the torque of the device has the greatest effect on the penetration rate of the EPB device.