A Decision-Support Model for Sustainable Management of Raw Material Flows in Hard Coal Preparation Plants

This paper presents an integrated decision-support model for the sustainable management of raw material flows in hard coal preparation plants, aimed at optimizing technological processes and supporting circular economy principles. Developed within the DynGOSP project, the framework combines a stationary process model with an automated UAV–GIS workflow for stockpile inventory. The computational component, designated as the “Forecast” module, incorporates empirical classification and partition-curve models to evaluate separation performance, product recovery, and quality under varying feed characteristics. The predictive accuracy of the beneficiation module was verified using laboratory and industrial reference data, yielding absolute relative errors ranging from 0.89% to 7.51% for concentrate yield and 0.73% to 3.37% for concentrate ash content. To extend the management concept beyond plant operations, the system integrates photogrammetry, QGIS spatial data processing, and GRASS GIS tools to generate Digital Surface Models (DSMs) for continuous volumetric monitoring of external coal stockpiles and waste dumps. The proposed solution constitutes a comprehensive Industry 4.0 framework that bridges process evaluation with spatial inventory, providing a scalable tool for resource efficiency, waste reduction, and supply chain optimization in the mining sector.