Abstarct: Determination of the relationship between rock properties and cutting rate is the aim of research due to the heavy using of circular diamond saws and the importance of predicting the cutting rate for the cost estimation and the best planning of the plants. For this purpose, 15 samples of travertine were tested and uniaxial compressive strength, tensile strength, Schmidt hammer value, Los Angeles abrasion, density, porosity were determined. Moreover, chemical analysis for check the percent of hard elements and texture studies for check the effect of grain size were performed. Also, cutting rate of samples was measured in some stone plants of Mahalat, too. After that principal component analysis was used for reducing the number of variables and predictive equations of cutting rate were presented by using linear and nonlinear regression. This equations were created between uncorrelated variables and some of the correlated variables because it is better than stone described baxsed on both physical and mechanical properties. After that the coefficient determination, root mean square error, variance account for and mean absolute percentage error used for select the better model. Since the values of these four criteria were suitable for two equations, better model selected by using test data and compare the measured values with the predicted values by this two equations. Finally, after select the best model by using the sensitivity analysis, porosity selected as the most effective of variables for prediction the cutting rate of travertine stones.