Geomechanics and Geoenergy Journal

Geomechanics and Geoenergy Journal

Estimating Oil-Water Relative Permeability Using Machine Learning: A Case Study from Southwest Iran

Document Type : Original Article

Authors
Ali Ranjbar 1
Mohammad Rasul Dehghani 2
Mahdi Maleki 3
1 Faculty of Petroleum, Gas and Petrochemical Engineering, Petroleum Engineering Department,, Persian Gulf University, Bushehr, Iran
2 Shahid Bahonar University
3 Persian Gulf University
10.22107/ggj.2025.559197.1260
Abstract
Relative permeability is a critical petrophysical parameter that controls multiphase flow behavior in porous media and significantly impacts reservoir simulation accuracy, recovery forecasting, and EOR planning. Conventional laboratory-based methods for determining relative permeability, while accurate, are time-intensive, costly, and spatially limited. This study focuses on applying machine learning techniques to estimate oil and water relative permeability using core data from a reservoir in southwest Iran. Sixteen core samples were analyzed, and seven input features—absolute permeability, porosity, irreducible water saturation, oil permeability at Swi, viscosities of oil and water, and pressure differential—were used to train four machine learning models: Extra Trees, K-Nearest Neighbors, Categorical Boosting, and Extreme Gradient Boosting.Models were optimized using Bayesian hyperparameter tuning and evaluated using R², RMSE, and MAE. For water relative permeability, the Extra Trees model delivered the best performance, achieving an R² of 0.9974 on the overall dataset, with the lowest RMSE (0.0045) and MAE (0.0007), indicating high accuracy and generalizability. KNN also performed well, especially in the 0.1–0.2 permeability range. In contrast, for oil relative permeability, the KNN model achieved the highest accuracy with an R² of 0.9973 and the lowest error metrics (RMSE = 0.0113, MAE = 0.0024), outperforming the other methods in both training and testing sets. Extra Trees performed poorly for oil permeability, especially in capturing higher permeability ranges. SHAP sensitivity analysis revealed that water saturation is the most influential factor for both oil and water models. For water, oil permeability at Swi also had a major impact; for oil, oil viscosity played the second most critical role.Overall, this study demonstrates that machine learning offers a robust, efficient alternative to laboratory experiments for estimating relative permeability, with model selection varying depending on the target fluid. These data-driven models can enhance reservoir simulation, reduce costs, and support improved reservoir management in heterogeneous systems.
Keywords
Relative permeability
Machine learning
Core analysis
Reservoir simulation
artificial intelligence
1.         Muccino, J.C., W.G. Gray, and L.A.J.R.o.G. Ferrand, Toward an improved understanding of multiphase flow in porous media. 1998. 36(3): p. 401-422.
2.         Chapter 7 Relative permeability, in Developments in Petroleum Science, J.R. Fanchi, Editor. 2000, Elsevier. p. 85-101.
3.         Viera, M.A.D., et al., Mathematical and numerical modeling in porous media: Applications in geosciences. 2012: CRC Press.
4.         Honarpour, M.M., G.V. Chilingarian, and S.J. Mazzullo, Chapter 8 Permeability and Relative Permeability of Carbonate Reservoirs, in Developments in Petroleum Science, G.V. Chilingarian, S.J. Mazzullo, and H.H. Rieke, Editors. 1992, Elsevier. p. 399-416.
5.         Porges, F., Chapter 5. Relative Permeability Concepts. 2006.
6.         Honarpour, M. and S.J.J.o.p.t. Mahmood, Relative-permeability measurements: An overview. 1988. 40(08): p. 963-966.
7.         Siginer, D. and S.J.J.A.M. Bakhtiyarov, Flow in porous media of variable permeability and novel effects. 2001. 68(2): p. 312-319.
8.         Malkovsky, V., A. Zharikov, and V.J.I. Shmonov, Physics of the Solid Earth, New methods for measuring the permeability of rock samples for a single-phase fluid. 2009. 45: p. 89-100.
9.         Bryant, S. and M.J.P.r.A. Blunt, Prediction of relative permeability in simple porous media. 1992. 46(4): p. 2004.
10.       Schembre, J., A.J.J.o.P.S. Kovscek, and Engineering, A technique for measuring two-phase relative permeability in porous media via X-ray CT measurements. 2003. 39(1-2): p. 159-174.
11.       Alizadeh, A. and M.J.R.o.G. Piri, Three‐phase flow in porous media: A review of experimental studies on relative permeability. 2014. 52(3): p. 468-521.
12.       Sigmund, P. and F.J.S.o.P.E.J. McCaffery, An improved unsteady-state procedure for determining the relative-permeability characteristics of heterogeneous porous media (includes associated papers 8028 and 8777). 1979. 19(01): p. 15-28.
13.       Hao, L., P.J.I.J.o.H. Cheng, and M. Transfer, Pore-scale simulations on relative permeabilities of porous media by lattice Boltzmann method. 2010. 53(9-10): p. 1908-1913.
14.       Blom, S., J. Hagoort, and D.J.S.J. Soetekouw, Relative permeability at near-critical conditions. 2000. 5(02): p. 172-181.
15.       Hossain, Z. Relative permeability prediction from Image Analysis of thin sections. in SPE Europec featured at EAGE Conference and Exhibition? 2011. SPE.
16.       Ecay, L., et al., On the prediction of permeability and relative permeability from pore size distributions. 2020. 133: p. 106074.
17.       Kadet, V., A.J.J.o.P.S. Galechyan, and Engineering, Percolation modeling of relative permeability hysteresis. 2014. 119: p. 139-148.
18.       Wang, Y., et al. The application of machine learning algorithm in relative permeability upscaling for oil-water system. in International Petroleum Technology Conference. 2021. IPTC.
19.       Alfonso, C.E., F. Fournier, and V. Alcobia. A machine learning methodology for rock-typing using relative permeability curves. in SPE Annual Technical Conference and Exhibition? 2021. SPE.
20.       Zhao, B., et al., A hybrid approach for the prediction of relative permeability using machine learning of experimental and numerical proxy SCAL data. 2020. 25(05): p. 2749-2764.
21.       Zhao, J., et al., A Permeability Prediction Model of Single-Peak NMR T 2 Distribution in Tight Sandstones: A Case Study on the Huangliu Formation, Yinggehai Basin, China. 2024. 56(6): p. 1303-1333.
22.       Okon, E.I. and D. Appah. Application of machine learning techniques in reservoir characterization. in SPE Nigeria Annual International Conference and Exhibition. 2021. SPE.
23.       Dehghani, M.R., et al., Estimation the pH of CO2-saturated NaCl solutions using gene expression programming: Implications for CO2 sequestration. 2025. 25: p. 104047.
24.       Okoro, E.E., et al., Application of artificial intelligence in predicting the dynamics of bottom hole pressure for under-balanced drilling: Extra tree compared with feed forward neural network model. 2022. 8(2): p. 227-236.
25.       Salih, A.K. and H.A.A.J.T.I.G.J. Hussein, Lost circulation prediction using decision tree, random forest, and extra trees algorithms for an Iraqi oil field. 2022: p. 111-127.
26.       Dehghani, M.R., et al., Estimation of hydrogen solubility in aqueous solutions using machine learning techniques for hydrogen storage in deep saline aquifers. 2024. 14(1): p. 25890.
27.       Omisakin, S. Ensemble Machine Learning Methods to Predict Oil Production. in Innovations and Interdisciplinary Solutions for Underserved Areas: 7th International Conference, InterSol 2024, Dakar, Senegal, July 3–4, 2024, Proceedings. 2025. Springer Nature.
28.       Cao, L., et al., Interpretable Soft Sensors using Extremely Randomized Trees and SHAP. 2023. 56(2): p. 8000-8005.
29.       Hashemizadeh, A., et al., Experimental measurement and modeling of water-based drilling mud density using adaptive boosting decision tree, support vector machine, and K-nearest neighbors: A case study from the South Pars gas field. 2021. 207: p. 109132.
30.       Khamis, Y.E., et al., Rate of penetration prediction in drilling operation in oil and gas wells by k-nearest neighbors and multi-layer perceptron algorithms. 2023. 14(3): p. 755-770.
31.       Martín-Martín, M., et al., Using python libraries and k-Nearest neighbors algorithms to delineate syn-sedimentary faults in sedimentary porous media. 2023. 153: p. 106283.
32.       Saddiqi, H.A., et al., Modelling and predicting lift force and trans-membrane pressure using linear, KNN, ANN and response surface models during the separation of oil drops from produced water. 2024. 66: p. 106014.
33.       Mohammadi, M.-R., et al., Modeling hydrogen solubility in hydrocarbons using extreme gradient boosting and equations of state. 2021. 11(1): p. 17911.
34.       Yakoot, M.S.e., A.M.S. Ragab, and O. Mahmoud. Multi-class taxonomy of well integrity anomalies applying inductive learning algorithms: analytical approach for artificial-lift wells. in SPE Annual Technical Conference and Exhibition? 2021. SPE.
35. Shakouri, S. and M.J.E. Mohammadzadeh-Shirazi, Modeling of asphaltic sludge formation during a
Geomechanics and Geoenergy Journal
Volume 8, Issue 1
Spring 2025
Pages 67-81