Conference Agenda

The objective of this systematic review was to analyze whether the implementation of the Ore Sorting system with sensors and artificial intelligence, compared to conventional sorting methods, improves operational profitability in mining operations with polymetallic deposits. To this end, the PRISMA methodology and the PICO approach were applied, formulating specific questions and establishing inclusion and exclusion criteria. Fifty-six scientific articles published between 2014 and 2024 were selected from academic databases such as Scopus and SciELO. The results were organized into five thematic areas: sensors used, classification techniques applied, comparison with conventional methods, financial indicators, and operational applications. XRT, NIR, and HSI sensors were identified as the most widely used, with classification efficiencies exceeding 85% in appropriate contexts. The most frequent techniques were Particle Sorting, Bulk Sorting, and On-belt Sorting, each with specific operational advantages. Compared to traditional methods such as flotation or gravimetric separation, Ore Sorting showed greater energy efficiency and classification accuracy. In addition, the studies reviewed reported positive financial indicators, with NPVs exceeding USD 1.8 million and IRRs between 38% and 52%. It is concluded that Ore Sorting represents a viable technological alternative for improving operational efficiency, reducing OPEX, and increasing metal recovery in polymetallic mining. Its implementation also contributes to environmental sustainability by reducing liabilities and utilizing waste rock, establishing itself as a key tool for smart mining.