The wear rate of diamond wire saw plays a vital role in the performance of sawing process. Predicting the sawing performance is very important in the production's cost estimation and planning of the dimension stone quarries. In this research, an adaptive neuro-fuzzy inference system (ANFIS) is applied to estimate the wear rate of diamond wire saw under uncertain processes; hence, indirect prediction in ANFIS is carried out using subtractive clustering method (SCM) and fuzzy c-means clustering method based on four effective rock properties, such as Shore hardness, Schimazek's F-abrasivity, uniaxial compressive strength and Young modulus. For this purpose, 38 rock samples were selected to test the proposed model from Turkey quarries. The results of indirect prediction indicated that the best performed model was related to ANFIS-SCM with highly acceptable degrees of accuracy 0.998 and 0.59 for R-2 of the train and test data sets, respectively. In addition, group method of data handling type of neural network is used to assess the factors influencing the wear rate of the diamond wire saw. A sensitivity analysis was performed on the laboratory test results of studied rocks using three methods. In comparison to the existing models, the estimated results showed that a satisfactory performance could be obtained using the proposed ANFIS-subtractive clustering method.