Research Article
Ensemble Machine Learning Methods to Predict Oil Production
@INPROCEEDINGS{10.1007/978-3-031-86493-3_27, author={M. D. Adewale and I. A. Adeyanju and J. Oju and O. C. Ubadike and U. I. Muhammed and S. T. Omisakin}, title={Ensemble Machine Learning Methods to Predict Oil Production}, proceedings={Innovations and Interdisciplinary Solutions for Underserved Areas. 7th International Conference, InterSol 2024, Dakar, Senegal, July 3--4, 2024, Proceedings}, proceedings_a={INTERSOL}, year={2025}, month={4}, keywords={Oil Price Machine Learning Oil Consumption Oil Production Oil Reserve}, doi={10.1007/978-3-031-86493-3_27} }- M. D. Adewale
I. A. Adeyanju
J. Oju
O. C. Ubadike
U. I. Muhammed
S. T. Omisakin
Year: 2025
Ensemble Machine Learning Methods to Predict Oil Production
INTERSOL
Springer
DOI: 10.1007/978-3-031-86493-3_27
Abstract
This research unveils an ensemble prediction model tailored for Nigeria’s oil production, addressing the need for accurate forecasting in a critical economic sector. Nigeria, a leading oil producer in Africa, relies heavily on its oil sector, necessitating robust prediction models for economic planning and stability. This study aims to create a predictive model integrating pivotal factors such as oil reserves, oil consumption, oil prices, and political stability. These factors were chosen due to their significant impact on oil production dynamics, encompassing economic, political, and consumption-related influences that critically determine production outcomes. Using data from 1980 to 2016, we employed advanced machine learning algorithms—including Extra Trees Regressor, XGBoost Regressor, and Random Forest Regressor—to enhance prediction accuracy. The Extra Trees Regressor emerged as the superior algorithm, demonstrated by a correlation coefficient of 0.8155, a mean absolute error of 0.2812, and a root mean squared error of 0.3929. Our findings confirm the model’s predictive power, highlighting the significant influence of critical variables on Nigeria’s oil production trends. This study offers indispensable insights to stakeholders, aiding in informed decision-making and showcasing the significant capabilities of ensemble machine learning in improving oil production forecasts. Ultimately, this work enhances understanding of the factors affecting oil production and supports strategic planning within Nigeria’s oil sector, promoting economic stability and growth.
