Home / Current Issue / Paper 1714840
Real-Time Prediction of Drilling Torque and Drag Using XGBoost on MWD Data from Directional Wells
Subject area: Science,Engineering and Technology · Area of research: Engineering
Abstract
The exact prediction of the torque and drag forces is necessary to improve directional drilling operations and reduce non-productive time in the complicated wellbore environments. Nonlinear relationships between downhole forces and parameters of drilling are however complex and not properly modeled using conventional empirical frameworks. This reduces the effectiveness in drilling process. The study discusses the design of a machine learning model, which uses Measurement While Drilling (MWD) data of directional wells in determining the real-time torque and drag. It deployed and tested three superior algorithms, namely, XGBoost, random forest (RF), and support vector regression (SVR). Our well data points were used in this assessment and the input features used were inclination, azimuth, hookload, Weight on Bit (WOB) and Revolutions per Minute (RPM). Our outlier removal was implemented based on Z-score cut-offs, we normalised the features with MinMax scaling, and we down-sampled with 30s temporal intervals. An 80 by 20 train-test split in conjunction with 5-fold cross-validation was used to validate the model. XGBoost demonstrated good performance in predicting torque where it attained a R2 of 0.9235 and an RMSE of 2.059 Nm. It had a R2 of 0.9762 and RMSE of 5.294 kN in terms of prediction of drag. The performance of the Random Forest model was also comparable and the model attained R2 of 0.9253 in torque and 0.9749 in drag. Conversely, Support Vector Regression (SVR) technique had R2 values of 0.9283 and 0.9781 in predicting torque and drag respectively. Further, the feature importance analysis indicated WOB, inclination, and RPM to be important predictors of both targets. The SHAP analysis conducted (SHAPley Additive Explanations) provided the contribution of each feature and thus increased the level of transparency of the model. This technology makes real-time drilling optimisation possible through the use of field-deployable computational efficiency that is used to minimise the effect of non-productive time and enhance operational safety.
Keywords
Torque and Drag, XGBoost, Random Forest, Support Vector Regression, Machine Learning, MWD, Directional Drilling, Real-Time Prediction, SHAP
References
[1] A Note on Predicting Rate of Penetration Using Machine Learning Models. (2025). Journal of Engineering Sciences, 54.
[2] A stacked generalization ensemble model for optimization and prediction of the gas well rate of penetration: A case study in Xinjiang. (2021). Journal of Petroleum Exploration and Production Technology, 11.
[3] Automated real-time prediction of geological formation tops during drilling operations: An applied machine learning solution for the Norwegian Continental Shelf. (2024). Journal of Petroleum Exploration and Production Technology, 14.
[4] Chen, Y., Zhang, D., & Dongxiao. (2020). Ensemble long short-term memory (EnLSTM) network.
[5] Challenges and Troubleshooting in Directional Drilling. (2021). International Journal of Multidisciplinary Research and Growth Evaluation, 3, 1--6.
[6] Etaje, D. (2022, March 21). Implementing soft string and stiff string models for torque and drag calculations. Open-Source Drilling Conference.
[7] Johancsik, E., & al. (1983). A new approach to torque and drag calculations in directional wells. Journal of Petroleum Technology, 35.
[8] Khushaba, N. R., Melkumyan, A., & Hill, J. A. (2022). A machine learning approach for material-type logging and chemical assaying from autonomous measure-while-drilling (MWD) data.
[9] Lundberg, S. M., & Lee, S.-I. (2017). Consistent feature attribution for tree ensembles.
[10] Mirhaj, S., Kaarstad, S., Aadnoy, E., & B. S. (2016). Torque and drag modeling: Soft-string versus stiff-string models.
[11] Orion II Data Compression Platform. (n.d.). SLB. https://www.slb.com/products-and-services/innovating-in-oil-and-gas/well-construction/measurements/measurements-while-drilling-services/orion-data-compression-mwd
[12] SVM Regression. (n.d.). Oracle. https://docs.oracle.com/en/database/oracle/machine-learning/oml4sql/23/mlsql/svm-regression.html
[13] Sterge, N., Sriperumbudur, B., Rosasco, L., Rudi, A., & Alessandro. (2019). Gain with no pain: Efficient kernel-PCA by Nyström sampling.
[14] Support vector machine. (n.d.). Wikipedia. https://en.wikipedia.org/wiki/Support_vector_machine
[15] TPE-XGBoost breakdown pressure prediction study based on physical constraints. (2024). In H. Zhang, M. Wang, L. Xie, & B. Yang (Eds.), Abstract Book of ICCEE2024 & ICEER2024.
[16] Wu, A., Hareland, G., Fazaelizadeh, M., & Mohammad. (2025). Torque and drag analysis using finite element method. Modern Applied Science, 19.
[17] XGBoost. (n.d.). https://xgboost.ai
[18] Yang, T., Liang, Y., Wang, Z., & Ji, Q. (2024). Dynamic prediction of shale gas drilling costs based on machine learning. In M. C. de Hoces (Ed.), Applied Sciences (Vol. 14, p. 10984).
How to cite this paper
@article{1714840,
author = {Dr. Ichenwo John Lander , Marvellous Amos },
title = {Real-Time Prediction of Drilling Torque and Drag Using XGBoost on MWD Data from Directional Wells},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {9},
pages = {451-460},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1714840.pdf},
abstract = {The exact prediction of the torque and drag forces is necessary to improve directional drilling operations and reduce non-productive time in the complicated wellbore environments. Nonlinear relationships between downhole forces and parameters of drilling are however complex and not properly modeled using conventional empirical frameworks. This reduces the effectiveness in drilling process. The study discusses the design of a machine learning model, which uses Measurement While Drilling (MWD) data of directional wells in determining the real-time torque and drag. It deployed and tested three superior algorithms, namely, XGBoost, random forest (RF), and support vector regression (SVR). Our well data points were used in this assessment and the input features used were inclination, azimuth, hookload, Weight on Bit (WOB) and Revolutions per Minute (RPM). Our outlier removal was implemented based on Z-score cut-offs, we normalised the features with MinMax scaling, and we down-sampled with 30s temporal intervals. An 80 by 20 train-test split in conjunction with 5-fold cross-validation was used to validate the model. XGBoost demonstrated good performance in predicting torque where it attained a R2 of 0.9235 and an RMSE of 2.059 Nm. It had a R2 of 0.9762 and RMSE of 5.294 kN in terms of prediction of drag. The performance of the Random Forest model was also comparable and the model attained R2 of 0.9253 in torque and 0.9749 in drag. Conversely, Support Vector Regression (SVR) technique had R2 values of 0.9283 and 0.9781 in predicting torque and drag respectively. Further, the feature importance analysis indicated WOB, inclination, and RPM to be important predictors of both targets. The SHAP analysis conducted (SHAPley Additive Explanations) provided the contribution of each feature and thus increased the level of transparency of the model. This technology makes real-time drilling optimisation possible through the use of field-deployable computational efficiency that is used to minimise the effect of non-productive time and enhance operational safety.},
keywords = {Torque and Drag, XGBoost, Random Forest, Support Vector Regression, Machine Learning, MWD, Directional Drilling, Real-Time Prediction, SHAP},
month = {March},
doi = {https://doi.org/10.64388/IREV9I9-1714840}
}