Geopersia

Geopersia

Assessment and Prediction of Rock Drillability in Hard Granitic Rocks Using Experimental Testing and Machine Learning Models

Articles in Press

Document Type : Research Paper

Authors
Ebrahim Sharifi Teshnizi 1
Mohammad Ghafoori 1
Gholam Reza Lashkaripour 1
Jafar Khademi Hamidi 2
Amir Hossein Yousefi 3
1 Department of Geology, Faculty of Science, Ferdowsi University of Mashhad, Mashhad, P.O. Box 91775-1436, Iran
2 Department of Mining Engineering, Faculty of Engineering, Tarbiat Modares University, Tehran, Iran
3 Department of Civil Engineering, Shahi.C., Islamic Azad University, Shahinshahr, Iran
10.22059/geope.2026.401662.648840
Abstract
Accurate prediction of rock drillability is critical for optimizing tunneling, mining, and excavation operations in hard rock environments. This study investigates the drillability of granitic rocks by integrating petrographic, physical, mechanical, and abrasivity parameters with both conventional regression and machine learning (ML) approaches. Laboratory tests were conducted on samples from six granitic rock groups, measuring properties such as brittleness, hardness, abrasivity indices, and the Drilling Rate Index (DRI). Statistical analyses, including linear and nonlinear regression, revealed
strong nonlinear relationships between DRI and engineering parameters, with R² values up to 0.95. Machine learning models, particularly Random Forest (RF) and Artificial Neural Networks (ANN), were applied independently, with RF achieving superior predictive performance (R² > 0.99) and lower error indices compared to ANN and regression models. The study also highlights the influence of different rock groups on model performance and discusses limitations related to dataset scale, in situ conditions, and model generalization. These results confirm that integrating laboratory measurements with ML techniques provides a reliable framework for predicting rock drillability, offering practical
guidance for excavation planning, equipment selection, and operational efficiency.
Keywords
Granitic rocks
Drilling Rate Index (DRI)
Machine Learning
Random Forest
Artificial Neural Networks
Rock drillability
Abrasivity
Rock mechanics
Subjects

Article Title Persian

_

Geopersia

Articles in Press, Accepted Manuscript
Available Online from 24 January 2026

  • Receive Date 01 September 2025
  • Revise Date 15 January 2026
  • Accept Date 24 January 2026