Cutting operations in the natural stone industry are very laborious and the accurate assessment of cutting performance helps in the selection of equipment and machinery during quarry and plant planning stages. Natural stone quarry operators need to predict the wear amount of diamond beads on wire and energy consumption of mono-wire stone cutting machines for estimation of the facility costs before the block or slab production. The unit wear (UW) and the unit energy (UE) values in the sawing of natural stone samples with mono-wire block cutting machine which can be used for slab cutting or squaring operations in the natural stone industry are two quantitative parameters to be used for the accurate assessment of cutting performance. In the current study, Artificial Neural Network (ANN) and Regression Models were investigated for predicting of UW and UE as monowire cutting machine parameters. The uniaxial compressive strength (UCS), cutting speed (CS) and peripheral speed (PS) of diamond wire parameters were selected as inputs, while UW and UE were considered as outputs of the prediction models. Prediction performances of the regression and the ANN based models were evaluated by correlation coefficient, root mean square error and cross correlations between the measured and the predicted values of UW and UE. The prediction performance evaluations of the models revealed that the best prediction performances were obtained from the ANN models, while performance indicators of the regression based relations followed those of the ANN models. Although the ANN is a powerful learning tool for prediction purposes, the requirements of a software to use of the developed ANN models may be evaluated as a practical limitation for practitioners. Therefore, the prediction charts for both UE and UW were produced to improve the practical value of the ANN models which consider PS, CS and UCS as inputs. It is concluded that the ANN based models can be used for prediction of reliable UE and UW parameters of mono-wire cutting operations.