Answers are grounded in this paper and cite their sources.
Background The growing digitalization of the mining sector has accelerated the adoption of predictive analytics to improve operational efficiency and reduce uncertainty in mineral production systems. In this context, predictive modeling has established itself as a strategic tool for addressing the dynamic complexity and geological heterogeneity of mining operations. This study presents a systematic review aimed at identifying, analyzing, and classifying the methodological approaches used in predictive models applied to mineral production. Methods The review was conducted in accordance with the PRISMA 2020 protocol, using searches in Scopus, Web of Science, and ScienceDirect for the period 2010–2025. After applying predefined eligibility criteria, 76 studies were included in the qualitative synthesis. The analysis considered modeling techniques, validation strategies, levels of analytical maturity, and operational domains. Results The results show a shift from traditional statistical models toward machine learning approaches, highlighting the use of Random Forests, Support Vector Machines, and artificial neural networks to model nonlinear relationships in high-dimensional operational data. Furthermore, there is a growing integration of hybrid frameworks and optimization-based approaches, indicating a gradual shift toward decision-support systems. A structured taxonomy of predictive techniques and an analytical maturity framework are proposed to classify models according to their methodological sophistication. Conclusions Although significant advances are identified, challenges related to external validation, reproducibility, and methodological standardization persist. These findings provide a conceptual basis for the development of hybrid prediction–optimization architectures in smart mining ecosystems based on real-time analytics.
No takes yet. Share an insight, caveat, or question.
Seclén et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: