Andriy Solovyov - Fotolia

Enterprises deploy AI in mining projects to improve analytics

By implementing AI in mining processes, enterprises are able to utilize data algorithms and automated machines to preserve the sensitive environment and their human workforce.

Oil, gold, copper, coal and other natural elements power our growing economies. As resources become scarcer, however, industries must go into increasingly more hostile environments for extraction. AI and machine learning technologies are aiming to keep resources flowing without risking human lives. The use of autonomous vehicles and equipment powered with computer vision and predictive analytics are improving mining performance, reliability, safety and operations.

Focused mining energy

Mining is often a highly disruptive activity requiring significant investment before any return is realized. For this reason, precise drilling or excavating is necessary for ROI. AI in mining is helping miners better understand the terrains of new developments. Drones capture aerial images of mining operations, and computer vision better understands data on the environment where operations will begin. By using satellite imagery, aerial photography and computer vision, resource extraction companies are better able to predict the best locations for mining potential resources, thus reducing wasted searching and low ROI. These technologies help make the mine more efficient by reducing the overall footprint of the mining operation, generating higher yields and minimizing the environmental impact.

Mining excavation displaces a huge amount of ground material. Separating the valuable material from invaluable dirt or rocks is streamlined using AI in mining technologies to save time and money. Goldcorp, one of the largest gold mining companies in the world, partnered with IBM Watson to improve data targeting new deposits of gold. By analyzing previously collected data and utilizing various cognitive technologies, IBM Watson was able to improve geologists' planning skills. AI, in this case, had the ability to glean new information from existing data, better determine specific areas to explore next and reach high-value exploration targets faster.

Assisted monitoring

Today's mining customers are eager to leverage data and machine insights to become more in tune with their day-to-day operations, allowing them to optimize mine performance and enable autonomous operations.
Jason KnuthSenior manager of data solutions, Komatsu

Once mines are operational, daily monitoring of mining operations maintains safety, while ensuring that the right resources are being extracted. Daily monitoring of people, equipment, materials and the environment produces millions of data points. The concept of a digital mine has been adopted by companies as a way of applying big data technology and concepts to the deluge of information that mining operations generate.

"Today's mining customers are eager to leverage data and machine insights to become more in tune with their day-to-day operations, allowing them to optimize mine performance and enable autonomous operations," said Jason Knuth, senior manager of data solutions at Komatsu, a mining industry provider.

"We're providing near-real-time insights on important aspects that affect total cost of operation, including operator performance data, global benchmarking and machine downtime," he added.

Autonomous mining vehicles

Autonomous mining vehicles are becoming more commonplace in mines as they are able to operate continuously and protect humans against harsh mining conditions. These mining-specific autonomous vehicles improve safety, equipment availability and overall productivity, as well as decrease the number of employees needed to efficiently run operations. Rio Tinto has made extensive use of autonomous mining vehicles by utilizing self-driving trucks that have already logged over 1 million miles of operation.

Autonomous loaders are being deployed in mines to scoop up and haul various invaluable and valuable materials. Mines are moving toward autonomous blast-hole drill systems and increasing overall safety by limiting human operators. Autonomous drilling systems need only one remote operator to control and manage multiple drilling rigs -- far more than they could operate without this system.

Additionally, mines are automating geotechnical inspections by performing automated open pit inspections and assessments using drones powered with computer vision. This AI in mining enables more frequent inspections and saves lots of time in the process.

Limited environmental impact

Mining has a large impact on the environment and ecosystem. From impacting wildlife habitats, contaminating and redistributing topsoil and underground water and generating erosion in the environment to polluting the environment with leakage of chemicals and related materials, mining operators have the ability to cause irreversible damage. Mining operations want to reduce this impact and have implemented edge devices with AI capabilities to monitor mining activities and output in all locations -- even remote and underground -- with real-time detection of leaks or potential hazards. Sensors monitor a variety of environmental factors, such as temperature, groundwater or underground changes, to help assess the impact that mining activities are having on the surrounding terrain. Computer vision and sensing technologies monitoring environmental changes are also using predictive analytics to determine the potential changes in erosion, wildlife habitats, topsoil redistribution and vegetation that mining could cause.

Though less glamorous, AI in mining enterprises is surprisingly advanced. They deploy intelligent sensors, autonomous vehicles and predictive analytics to a greater extent than most other industries. As cognitive technologies continue to improve, mining companies will likely increase their use of robotics and AI technologies to make mines safer and more economical, as well as make better use of their data and real-time monitoring to prevent environmental harm.

Dig Deeper on Enterprise applications of AI

Business Analytics
Data Management