Integrating Machine Learning and Digital Twin to Improve Plant Equipment Efficiency in the Mining Sector: A Systematic Review (#775)

Authors

Mamani López, Heidy Yadira

Neyra Enciso, Astrid Nohelí

Canahua Apaza, Nohemy Miriam

Abstract

Mining is an environment that demands increased operational efficiency, sustainability, and cost reduction, where digitalization is emerging as a crucial and strategic solution. The purpose of this systematic review is to evaluate the use of machine learning and digital twins in relation to the operational efficiency of mining plant equipment. Methodologies such as PRISMA ensure the effectiveness of the selection process for the articles considered, allowing the analysis to be structured using the PICO strategy, considering equipment characteristics, ML-DT integration mechanisms, and the results obtained after its implementation. The search was conducted using search engines such as Scopus and SciencieDirect, filtering publications that were not published between the years 2019 and 2025, obtaining a total of 74 articles that met the inclusion criteria. The findings revealed a growing trend in the use of these technologies to optimize processes such as flotation, conveying, predictive monitoring, and mill energy control. Overall, significant improvements were identified in reducing energy consumption, lowering maintenance costs, and increasing equipment availability and reliability. Furthermore, ML-based predictive models demonstrated high accuracy in early fault detection and real-time operational decision-making. In short, the integration of Machine Learning and Digital Twin in mining plants represents a key advance toward more efficient, safe, and sustainable operations. This technological synergy not only optimizes equipment performance but also paves the way for a more competitive and resilient mining industry in the face of future challenges