ARCHIVES

This research paper presents a conceptual framework for an Intelligent Digital Geometallurgical System (IDGMS) integrating multi-agent algorithms and digital twins. The proposed architecture is organized into three hierarchical levels—operational, analytical, and coordination-control—connected through a unified information and semantic data bus, ensuring end-to-end data flow, semantic interoperability, and robust decision-making under uncertain and dynamic conditions. At the operational level, agents perform real-time data acquisition, filtering, calibration, and geospatial referencing from sensors, laboratory systems, IIoT devices, and equipment telemetry. The analytical level implements predictive analytics, geostatistical modeling, three-dimensional geomodeling, and digital twins, enabling scenario-based evaluation and adaptive process control. The coordination-control level aggregates analytical results, executes multi-criteria optimization, and generates strategic decisions, ensuring alignment of production objectives and resource allocation. The integration of digital twins provides a virtual environment for “what-if” scenario analysis, continuous model refinement, and predictive adjustment of technological regimes. Ontology-driven data unification enhances semantic consistency across heterogeneous data sources, reducing ambiguity in agent interactions. The proposed multi-agent IDGMS demonstrates high adaptability, self-regulation, and predictive capability, offering a scientific basis for improving geometallurgical modeling, optimizing metallurgical processes, and enhancing the digital maturity and operational efficiency of mining and metallurgical enterprises