AI-methods to Link Mineralogy and Core Sawing to Comminution Efficiency phase 2 (eCoreX2)
Purpose and goal
The purpose is to develop and integrate a cost-effective method for predicting ore grindability using sawing energy, XRF data, and AI-based geometallurgical modelling. The goal is to operationalize the method directly within the end users’ drill-core and analysis workflows, thereby delivering scalable predictions that strengthen decision-making, energy efficiency, and process design from mine to concentrate.
Expected effects and result
The project is expected to deliver a validated method that links sawing energy, XRF data, and mechanical tests to predictions of ore competence. By integrating data collection into routine drill-core workflows, significantly more data points are obtained across larger portions of the ore body and throughout the mine’s lifetime. This enables more accurate models, improved process control, more energy-efficient comminution, and more robust decision-making.
Planned approach and implementation
The project is executed through laboratory testing of drill cores, followed by integration of sawing-energy and XRF measurements into the industrial partners’ routine workflows. These data are combined with extended grindability and strength tests. Predictive AI/ML models are trained and linked with DEM simulations to estimate process response based on ore competence. The work packages are carried out iteratively to ensure robust validation and operational adaptation.
The text has been written by the project team. The content is copied from the funding agency’s website and has not been reviewed by the Program Office.