Artificial intelligence and machine learning are revolutionizing the mining industry. Machine Learning is a growing and diverse field of Artificial Intelligence which studies algorithms that are capable of automatically learning from data and making predictions based on data. Machine learning is one of the most exciting technological areas of study today. Each week there are new advancements, new technologies, new applications, and new opportunities. It’s inspiring, but also overwhelming. That’s why we created this guide to help you keep pace with all these exciting developments. Whether you’re currently employed in the mining industry, or just pursuing an interest in the subject, or working with us at Produvia, there will always be something here to inspire you!

Here are a few ways artificial intelligence and machine learning can be used for mineral prospecting, mineral discovery, mine development, mine production, and mine reclamation.

1. Prospecting And Exploration

Machine learning can be used to answer the question: “where to explore?”

Classifying Rock Faces

  • Automatically identify rock faces using well-logging data (spontaneous potential logging, and resistivity logging)

Classifying Lithology

  • Automatically identify lithology (rock and soil classes) using remote sensing data (multispectral satellite data) (Yu et al., 2013)

Predicting Mineral Prospectivity

  • At Produvia, we predict the locations of potential ores using satellite imagery and known mineral locations (Mapo)

2. Discovery And Advanced Exploration

Machine learning can be used to answer the question: “what’s in the ground?”

Predicting Core Drilling Targets

  • Predict targets for drilling using previous core drill data, soil samples, mine site surveying, or high impact data (geophysics, channel sampling, trenching and diamond drilling)

Classifying Subsurface Conditions

  • Automatically identify subsurface minerals, folds and fractures using acoustic signals

3. Development/Construction

Machine learning can be used to answer the question: “how to build the mine?”

Predicting Construction Phases

  • Predict development and construction phases using mineral prospectivity, aerial imagery, and previous mine site construction designs

4. Operation And Production

Machine learning can be used to answer the question: “how to mine, mill, and process the discovered ores?”

Classifying Froth Imagery

  • Perform segmentation and classification of froth images based on bubble size (small, middle and large) and texture (roughness, smoothness, regularity) (Wang et al., 2016)

Improving Operation Efficiency

  • Compute optimal process control actions without violating the operating constraints by combining Artificial Neural Networks (ANNs), statistics and multivariable modeling (NeuralWave)

Predicting Ore Reserves

Predicting Mineral Output

  • Predict the impurities output of an iron ore processing plant using mineral production data (Willingham, 2016)

Automating Mining Vehicles

  • Automate vehicles using sensors (laser range- finders, radar, electromagnetic antennae, ultrasound, cameras)

Managing Assets

  • Manage mining assets using operational data and equipment down-time

Analyzing Water Usage

  • Assess water sources and water usage patterns using on-site data

Predicting Downtime, Mill Loads, and Plant Performance

  • Predict downtime using pump pressure, SAG (semi-autogenous grinding) mill overloads, and processing plant performance and geological data (Mining Magazine, 2016)

Predicting Machine Failure

Assessing Ore Fragmentation

  • Automatically assess ore fragmentation in underground and open-pit operations using 3d mapping point cloud data (UGPSRapidMapper, 3D Laser Mapping or MVS) (MQWorld, 2016)

Automating Geotechnical Inspections

  • Perform automated underground wall inspections and assessments of spalling, cracked shotcrete, and plate deformation, missing plates and mesh bagging using 3d mapping data (digital mine surveys) (MQWorld, 2016)
  • Perform automated open-pit inspections and assessments using aerial photography

Detecting Machinery Missing Teeth

  • Prevent crusher downtime or damage to conveyor belts resulting from dis-engaged teeth/adapters using thermal camera imaging (Motion Metrics)

Reducing Machinery Blind Spots

  • Prevent equipment collisions with real-time surveillance views using a series of cameras (Motion Metrics)

Creating Fault Detection and Classification (FDC)

  • Create maintenance prevention policies using time-series data of alarms (Wang et. al, 2015)

5. Reclamation

Machine learning can be used to answer the question: “how do we rehabilitate the land and protect the environment, people, and animals?”

Monitoring Environmental Changes

  • Monitor the environment (fire, vegetation, and water) using remote-sensing technologies (satellite imagery and/or aerial photography)

Predicting Mine Rehabilitation

  • Predict restoration progress of mined or abandoned lands and waters using satellite imagery
  • Predict changes in erosion and acid mine drainage, wildlife habitats, topsoil redistribution and vegetation using satellite imagery

Monitoring Animal Migration

  • Monitor changes in animal migration paths using satellite imagery and aerial photography

Predicting Environmental Risks

Automatically Assessing and Managing Risks

  • Automatic risk assessment and risk management using sensor network technologies, data on pollution, levels of radiation, meteorological parameters (Kanevski et al., 2011)

Here are five additional AI projects to explore:

Underground Automation

  • Track people and schedules tasks
  • Increase production through monitoring operator and equipment location
  • Track real-time production
  • Improve safety and increase utilization of mining equipment

Digital Maintenance Work Management

  • Increase availability
  • Reduce parts spend
  • Reduce unplanned work and work overruns

Processing Automation

  • Automate mining operations to increase throughput and recovery
  • Improve the quality of decisions in processing to maximize production

Intelligent Data Platform

  • Improve data quality and transparency to optimize operations

Predictive Maintenance

  • Predict mining exhaust failures one week in advance

Next Step

Do you work in the mining industry? Are you interested in developing artificial intelligence technologies?

Schedule a call with Slava Kurilyak, Founder/CEO at Produvia.

"Most businesses struggle to innovate. We help companies innovate faster by developing artificial intelligence technologies so they can crush their competition."

— Slava Kurilyak, Founder/CEO at Produvia