Post by Harish Panghal

🎉 Excited to share our latest publication! I am delighted to announce that our research article has been published in Rock Mechanics and Rock Engineering (Springer Nature). 📄 Title: Explainable Gradient-Boosting Framework for Robust Prediction and Geomechanical Interpretation of Drilling Rate Index (DRI) in Granitic Rocks Authors: Ebrahim Sharifi Teshnizi, Suhaib Rasool Wani, Harish Panghal, Mohammad Ghafoori, and Gholam Reza Lashkaripour In this study, we developed an explainable machine learning framework that integrates XGBoost, LightGBM, CatBoost, AdaBoost, Bayesian optimization, SHAP, and Partial Dependence Plots (PDP) to accurately predict the Drilling Rate Index (DRI) of granitic rocks while providing meaningful geomechanical interpretations. Key highlights: ✅ Bayesian-optimized gradient-boosting models with exceptional predictive accuracy (R² > 0.994). ✅ CatBoost achieved the best performance (R² = 0.9987, RMSE = 0.43). ✅ SHAP and PDP identified S20, Sievers' J-value (SJ), Brazilian Tensile Strength (BTS), and Cerchar Abrasivity Index (CAI) as the dominant factors controlling DRI. ✅ External validation confirmed excellent generalization capability. ✅ A user-friendly GUI was developed for real-time engineering applications. This work demonstrates how Explainable Artificial Intelligence (XAI) can bridge the gap between predictive accuracy and engineering interpretability, supporting more informed decision-making in tunnel excavation and rock engineering. Many thanks to my co-authors and collaborators for their valuable contributions throughout this research. 📖 Read the article here: https://lnkd.in/dBiSzgzn #RockMechanics #RockEngineering #MachineLearning #ExplainableAI #XAI #CatBoost #XGBoost #GeotechnicalEngineering #TunnelEngineering #TBM #DataScience #EngineeringResearch #SpringerNature