top of page
Image by SJ Objio
Title: Mine-to-Mill Integrated Intelligent Optimisation
Organization

Vale - Julius Kruttschnitt Mineral Research Centre, The University of Queensland

Period

January 2023 - January 2025

Status

Completed

Est. Budget

USD 3.8M

Est. Business Benefit

USD 32.0M

Indicative Value-to-cost

8.4x

Keywords

Mine-to-mill, integrated optimisation, geometallurgy, energy efficiency, recovery, value chain analytics

Benefit basis

Estimated value from integrated mine-to-mill throughput, recovery, energy, and quality optimisation across a selected operating corridor or processing chain.

Business Challenges 
​Mining value chains consist of interdependent activities from drilling and blasting through loading, hauling, crushing, grinding, processing, and final product quality. Optimising each activity in isolation can create local gains but may reduce global value if impacts on downstream energy use, throughput, recovery, fragmentation, or product quality are not considered. The business challenge was to create an integrated optimisation approach that could connect previously separated operations and identify the best mine-to-mill operating strategy under ore variability, cost pressure, and declining average grades.
Expanded Technical Solution
The project completed an integrated AI-based mine-to-mill optimisation framework connecting mine, move, and mill activities. The approach treated ore as a value-carrying object through the value chain and linked drilling/blasting parameters, fragmentation, shovel loading, haulage performance, crusher behaviour, mill energy, recovery, throughput, and product quality. The technical framework included data integration, geometallurgical feature engineering, digital ore tracking, energy and mass-balance logic, multi-objective optimisation, and scenario evaluation.
Technical Work Packages/Methods
  • Integrated mine planning, drill-and-blast, fleet management, dispatch, crusher, mill, laboratory, geometallurgical, and production accounting datasets.
  • Developed ore-domain and material-tracking logic to connect upstream ore properties to downstream process behaviour.
  • Modelled fragmentation, throughput, specific energy, recovery, dilution, ore hardness, and product quality relationships.
  • Applied machine learning and multi-objective optimisation to balance tonnes, recovery, energy, cost, emissions, and product quality.
  • Designed a scenario engine for testing changes in blast design, mine sequencing, haulage strategy, crusher settings, and milling operating conditions.
Key Deliverables
  • Integrated mine-to-mill data model and analytical framework.
  • AI/ML models linking upstream mining variables to downstream processing outcomes.
  • Multi-objective optimisation methodology for whole-value-chain decision support.
  • Technical roadmap for operational deployment, data governance, and benefit tracking.
Results and Value Created
The project demonstrated how integrated optimisation can produce larger value than isolated departmental improvements. It created a technical basis for reducing energy intensity, improving throughput and recovery, stabilising product quality, and aligning mining and processing decisions around enterprise value. The estimated benefit reflects improved throughput, recovery, energy efficiency, product quality, and reduction of cross-functional value leakage.
bottom of page