Predicting the cuttability of rocks using artificial neural networks and regression trees

This paper is concerned with the applications of artificial neural networks (ANN) and regression trees along with the multivariate statistical tools for predicting specific cutting energy (SE) of rocks from their intact properties. For that purpose, data obtained from three rock cutting projects have been subjected to statistical analyses using MATLAB software. Principal components and factor analyses have shown that uniaxial compressive strength (UCS), Brazilian tensile strength (BTS), static modulus of elasticity (Elasticity), and cone indenter hardness (CI) seemed to be the most influential independent variables in the data set. Hierarchical cluster tree analysis has divided the variables into three different natural clusters. Three predictive models for SE were developed using multiple nonlinear regression, ANN, and regression trees methods. Regression tree model has been understood to fit the data better than the other two models. ANN model also produced a high correlation coefficient, indicating its significance in predicting rock cuttability.