Chapter (14): Industrial AI in Mining Towards Mine 4.0

CHAPTER DESCRIPTION

The new mining equipment is supplied with a broad variety of sensors allowing them to read various parameters. This information plus the contextual parameters might be fused to perform prognostics and predict the need for maintenance and repair.

Therefore, data collection, structuring, analysis, and communication between OEM and mines operators is crucial, but advanced analytics becomes essential for RUL estimations of subsystems and systems and further delivery of services such as:

Inventory management

  • Total production quality control, based on online monitoring of the ore flow
  • Automated production and haulage solutions,
  • Calibration of mine plan compliance with actual asset/fleet performance
  • Improvement using KPIs such as technical availability, asset utilization,
  • Maintenance metrics (MTTR, MTBF) and process-related criteria
  • Root cause analysis, trends & event correlation

To deliver these services, the data collected from the assets, the contextual data from the mine, ore characteristics, and production planning need to be analyzed through:

  • Anomaly detector classifiers and clustering
  • Failure forecasting and prognosis both data and model-based
  • Prescriptive analytics of mining assets to balance productivity and asset health
  • Deep learning of all data sources, both asset data and mine data in the pit and process plant
Not Enrolled

Chapter Includes

  • 6 Parts