AI-Driven Digital Twins in Mining Operations: A Comprehensive Review

Цифровые двойники на основе искусственного интеллекта в горнодобывающей промышленности: всесторонний обзор
Shouki A. Ebad, Aws I. Abueid, Marwa Amara, Rabie Ahmed, Shouki A. Ebad, Aws I. AbuEid
2026-04-29

AI-driven digital twinsPRISMA methodmachine learning and deep learningmining operationssystematic literature review
The mining industry is going through a big digital change because of the use of new technologies that are meant to make work safer, more productive, and more environmentally friendly. AI-driven digital twins (AI-DTs) are one of these new ideas. They combine real-time data collection with smart analytics to make it possible for decisions to be made in a predictive, adaptive, and autonomous way. This paper provides a thorough systematic literature review (SLR) of AI-DT applications in mining operations, encompassing studies published from 2015 to 2025. According to the PRISMA method, 68 primary studies were chosen and looked at from many angles, such as publication trends, demographic analysis, research methods, data sources, mining domains, and the AI techniques that were used. The findings reveal an increasing scholarly interest in AI-DTs, characterized by a significant prevalence of machine learning and deep learning methodologies, alongside a preference for real-world sensory data to augment model accuracy. Most applications deal with physical assets, processing plants, and operational systems. Subsurface environments, on the other hand, are still not well understood. The review also points out some major problems with data integration, scalability, interoperability, and the fact that there has not been much large-scale industrial validation. Based on these findings, the paper points out important areas of research that need more work and suggests ways to move forward with the development and use of AI-DTs in mining. In conclusion, this study gives researchers and practitioners a clear plan for how to use AI-DTs to make mining operations more efficient, resilient, and long-lasting.
1
A systematic PRISMA-based review identified and analyzed 68 primary studies on AI-driven digital twins in mining published between 2015 and 2025.
2
Applications commonly use real-world sensor data to improve model accuracy and primarily target physical assets, processing plants, and operational systems.
3
Major barriers include data integration, scalability, interoperability, and limited large-scale industrial validation, highlighting priorities for future research.
4
Research interest in mining AI-driven digital twins is increasing, with machine learning and deep learning representing the predominant analytical approaches.
5
Subsurface mining environments remain comparatively underexplored in existing AI-driven digital twin research.

AI-driven digital twins in mining operations

Applications, AI techniques, data sources, and implementation challenges of AI-driven digital twins for predictive, adaptive, and autonomous mining operations

Publication Details
Publication Date
2026-04-29
Journal
Publisher
ISSN
Cited by
10
Access Type
Author Information
Authors
Shouki A. Ebad
Aws I. Abueid
Marwa Amara
Rabie Ahmed
Shouki A. Ebad
Aws I. AbuEid
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%