Introduction
Accurate prediction of mineral grades is a fundamental step in mineral exploration and resource estimation, which plays a significant role in the economic evaluation of mining projects. Currently available methods are based either on geometrical approaches or geostatistical techniques that often considers the grade as a regionalised variable (Kaplan & Topal, 2020). Geostatistics has been widely used for qualitative estimation of ore deposits for many decades. However, ore quality does not vary uniformly in three dimensions which results in a poor quality estimation with the conventional geostatistical methods (Jain et al., 2022). Consequently, these limitations and complexities inspired researchers to investigate alternative approaches that can be utilised to overcome such obstacles. Over the past few decades, several researchers focused on various computational learning techniques that can predict grades more accurately without having to rely on an underlying assumption (Jang & Topal, 2014). The main aim of this article is to show the importance of using artificial intelligence to estimate the grades of a given deposit.
Key AI techniques in ore grade estimation
Artificial Neural Network (ANN)
Artificial neural networks (ANNs) are computer models that are designed to emulate human information processing capabilities such as knowledge processing, speech, prediction, and control. The ability of ANN systems to handle a large number of variables with complex relationships, spontaneously learn from examples, reason over inexact and fuzzy data, and to provide adequate and quick responses to new information has generated increasing acceptance of this technology in different engineering fields (Al-Alawi & Tawo, 1998). In order to limit estimation errors, systems based on neural networks, with their ability to process billions of data items in record time, make it possible to obtain results with a higher level of confidence than those that could be obtained using conventional methods such as ordinary kriging.
Support Vector Regression (SVR)
SVR is a regression problem extension of the support vector machine (SVM) classification model. To generalize the SVM to SVR, an insensitive zone known as the ε-tube is added around the function (Cortes & Vapnik, 1995). A non-linear kernel function is used to convert input features from the original space into a higher dimension. The problem becomes the construction of an optimal linear surface that matches the data in the feature space (Awad & Khanna, 2015). SVR has been applied to model the relationship between geological variables and ore grades, providing robust predictions even with limited data samples. Research comparing various machine learning models, including SVR, found them effective for ore grade estimation in different mineral deposits.
The artificial intelligence techniques described above are not the only ones used in the mining industry to estimate grades using artificial intelligence. There are several in the literature yet to be discovered.
Application and benefits
The use of artificial intelligence-based approaches for estimating content offers significant advantages.
Improved prediction accuracy
AI models can capture complex relationships between geological variables and ore grades, leading to more precise estimations. According to (Ismael et al., 2024), a study was carried out at El-Gezera region in El- Baharya Oasis, Western Desert of Egypt for forecasting the iron ore grade. a novel Artificial Neural Network (ANN) model, geo-statistical methods (Variograms and Ordinary kriging), and Triangulation Irregular Network (TIN) were employed and the presented ANN model estimates the iron ore as a function of the grades of Cl%, SiO2%, and MnO% with a correlation factor of 0.94. This final result demonstrates that, ANN is an excellent tool for grade estimation.
Cost reduction
By providing accurate grade estimations, AI helps in optimizing drilling and exploration activities, thereby reducing operational costs. The integration of AI in predicting ore grades significantly enhances the precision of assessments in mineral exploration.
Handling complex deposits
The classical method of estimation is less used in ore grade estimation than geostatistics (kriging) which proved to provide more accurate estimates by its ability to account for the geology of the deposit and assess error. AI techniques have been employed in diverse ore deposit types and have proven to provide comparable or better results than traditional methods, especially in complex structurally controlled vein deposits (Abuntori et al., 2021).
Conclusion
The application of artificial intelligence in predicting ore grades represents a significant advancement in the mining and minerals sector. By leveraging advanced algorithms and data analytics, AI technologies enhance the accuracy and efficiency of ore grade estimation, enabling more informed decision-making and optimized resource extraction.
Reference
Abuntori, C. A., Al-Hassan, S., & Mireku-Gyimah, D. (2021). Assessment of Ore Grade Estimation Methods for Structurally Controlled Vein Deposits—A Review. Ghana Mining Journal, 21(1), Article 1. https://doi.org/10.4314/gm.v21i1.4
Al-Alawi, S. M., & Tawo, E. E. (1998). Application of Artificial Neural Networks in Mineral Resource Evaluation. Journal of King Saud University – Engineering Sciences, 10(1), 127–138. https://doi.org/10.1016/S1018-3639(18)30692-5
Awad, M., & Khanna, R. (2015). Support Vector Regression. In M. Awad & R. Khanna (Eds.), Efficient Learning Machines: Theories, Concepts, and Applications for Engineers and System Designers (pp. 67–80). Apress. https://doi.org/10.1007/978-1-4302-5990-9_4
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273–297. https://doi.org/10.1007/BF00994018
Ismael, A., Embaby, A. K., Ali, F., Farag, H., Gomaa, S., Elwageeh, M., & Mousa, B. (2024). Prediction of Iron Ore Grade using Artificial Neural Network, Computational Method, and Geo-statistical Technique at El-Gezera Area, Western Desert, Egypt. Journal of Mining and Environment, 15(3), 889–905. https://doi.org/10.22044/jme.2024.13879.2581
Jain, G., Pathak, P., Bhatawdekar, R. M., Kainthola, A., & Srivastav, A. (2022). Evaluation of Machine Learning Models for Ore Grade Estimation. In A. K. Verma, E. T. Mohamad, R. M. Bhatawdekar, A. K. Raina, M. Khandelwal, D. Armaghani, & K. Sarkar (Eds.), Proceedings of Geotechnical Challenges in Mining, Tunneling and Underground Infrastructures (pp. 613–624). Springer Nature. https://doi.org/10.1007/978-981-16-9770-8_40
Jang, H., & Topal, E. (2014). A review of soft computing technology applications in several mining problems. Applied Soft Computing, 22, 638–651. https://doi.org/10.1016/j.asoc.2014.05.019
Kaplan, U. E., & Topal, E. (2020). A New Ore Grade Estimation Using Combine Machine Learning Algorithms. Minerals, 10(10), Article 10. https://doi.org/10.3390/min10100847
