From Rocks to Intelligence: AI and Data-Driven Technologies for Sustainable Mining

Abstract

Mining is fundamental to modern society and is becoming increasingly important as the global transition towards renewable energy, electrification and advanced technologies drives demand for minerals and critical resources. At the same time, the industry faces increasing pressure to improve safety, productivity, resource efficiency and environmental performance. Artificial intelligence (AI), machine learning (ML), automation and data-driven technologies offer significant opportunities to address these challenges.
This keynote explores the emerging role of AI and ML in transforming conventional mining into a smarter and more sustainable industry. Modern mining operations generate large volumes of geological, geotechnical, operational and environmental data. When combined with engineering knowledge, these data can support improved prediction, optimisation and decision-making throughout the mining value chain.
Particular attention is given to applications in rock mechanics, drilling and blasting, rock fragmentation, blast-induced vibration and geotechnical risk assessment. Machine-learning models can identify complex relationships between geological conditions, rock properties, blast-design parameters and operational outcomes. Such approaches can assist in optimising fragmentation, reducing unwanted vibration and overbreak, improving explosive utilisation and enhancing overall mining efficiency.
However, predictive accuracy alone is not sufficient for engineering applications. AI models used in mining must also be reliable, interpretable and physically meaningful. Explainable AI, physics-informed machine learning and hybrid models that combine engineering principles with data-driven approaches provide promising pathways towards trustworthy decision-support systems.
AI can also make an important contribution to sustainable mining. Improved blast fragmentation can reduce downstream crushing and grinding requirements and associated energy consumption. Intelligent monitoring can assist in managing vibration, dust, emissions and geotechnical hazards, while predictive systems can improve equipment utilisation and resource efficiency.
The future mine will therefore be characterised by the integration of AI, advanced sensing, automation, digital technologies and engineering expertise. Nevertheless, challenges related to data quality, geological uncertainty, model transferability, workforce capability and responsible implementation must be addressed.
AI should not replace engineering judgement; rather, it should enhance it. By combining computational intelligence with physical understanding and human expertise, the mining industry can progress towards operations that are safer, smarter, more productive and environmentally responsible.