EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access · CC BY 4.0

Predicting rate of penetration (ROP) with machine learning models: A physics-validated comparative case study of Gradient Boosting Regressor (GBR), Random Forest (RF), and K-Nearest Neighbours (KNN)

Musaab Magzoub, Noora Al Mansoori, Mohammed Abdalla Ayoub Mohammed iD, Almostafa Alhadi, Kaushik Sivaramakrishnan, Munawar Shaik iD, Abu Bakar Adni iD

PLOS One · 06.11.2026 · 21 (11) · e0360046

DOI: 10.1371/journal.pone.0360046 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift PLOS One, online erschienen am 06.11.2026. Verfasst von Musaab Magzoub und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemical and Petroleum Engineering, United Arab Emirates University, Al Ain, United Arab Emirates. Hauptthemen laut MEDLINE: Machine Learning. Die Arbeit nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: College of Graduate Studies (CGS), United Arab Emirates University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A successful drilling operation plays a vital role in the profitability of exploration and production of hydrocarbon reserves. However, these operations can come with high drilling rig rental costs. Reliable predictions of Rate of Penetration (ROP) can avoid unnecessary drilling equipment and rig rental expenses. In this study, Gradient Boosting Regressor, Random Forest, and K-Nearest Neighbours were developed to predict rate of penetration based on drilling operational data. The machine learning models were benchmarked with established empirical models, such as the Bingham and Bourgoyne and Young models, which are typically used to predict the rate of penetration. All machine learning models achieved, coefficient of determination (R 2 ) > 0.93 and Average Absolute Percentage Error (AAPE) < 3% on the test dataset. Random Forest under Scenario 1 was identified as the most physically consistent model, attaining strong predictive performance with R 2 = 0.9494 & AAPE = 1.96%, satisfying both the Weight on Bit (WOB) flounder point, and rotary speed optimum criteria used within the industry. These results support the practical use of machine learning-based ROP prediction as a reliable tool for real-time optimization of drilling parameters, offering a pathway to reduce non-productive time and rig costs while preserving the physical interpretability that drilling engineers require for field decision making.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningRandom ForestBoosting Machine Learning AlgorithmsPredictive Learning ModelsPrediction Algorithms

Autor:innen und Institutionen

  • Department of Chemical and Petroleum Engineering, United Arab Emirates University, Al Ain, United Arab Emirates.
  • Hildebrand Austin Department of Petroleum and Geosystems Engineering, The University of Texas at Austin, Austin, Texas, United States of America.
  • Chemical and Process Engineering, Faculty of Engineering, University of Strathclyde Bahrain, Kingdom of Bahrain.

ORCID-Profile: Mohammed Abdalla Ayoub Mohammed, Munawar Shaik, Abu Bakar Adni

Förderung

  • College of Graduate Studies (CGS), United Arab Emirates University

Zitieren

APA

Magzoub, M., Al Mansoori, N., Mohammed, M. A. A., Alhadi, A., Sivaramakrishnan, K., Shaik, M., & Adni, A. B. (2026). Predicting rate of penetration (ROP) with machine learning models: A physics-validated comparative case study of Gradient Boosting Regressor (GBR), Random Forest (RF), and K-Nearest Neighbours (KNN). PLOS One, 21(11), e0360046. https://doi.org/10.1371/journal.pone.0360046

Vancouver

Magzoub M, Al Mansoori N, Mohammed MAA, Alhadi A, Sivaramakrishnan K, Shaik M, et al. Predicting rate of penetration (ROP) with machine learning models: A physics-validated comparative case study of Gradient Boosting Regressor (GBR), Random Forest (RF), and K-Nearest Neighbours (KNN). PLOS One. 2026;21(11):e0360046. doi:10.1371/journal.pone.0360046

BibTeX
@article{Magzoub2026,
  author = {Magzoub, Musaab and Al Mansoori, Noora and Mohammed, Mohammed Abdalla Ayoub and Alhadi, Almostafa and Sivaramakrishnan, Kaushik and Shaik, Munawar and Adni, Abu Bakar},
  title = {Predicting rate of penetration (ROP) with machine learning models: A physics-validated comparative case study of Gradient Boosting Regressor (GBR), Random Forest (RF), and K-Nearest Neighbours (KNN)},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {11},
  pages = {e0360046},
  doi = {10.1371/journal.pone.0360046}
}
Kostenlos starten →