Volume 48, No 2, 2026, Pages 274-293
Download full text in PDFA Review of Artificial Intelligence Enabled Digital-Twin Technologies in Tribology
Authors:
Raj Shah 
,
Mathew Stephen Roshan ,
Vikram Mittal
DOI: 10.24874/ti.2096.12.25.03
Received: 18 December 2025
Revised: 3 February 2026
Accepted: 3 March 2026
Published: 15 June 2026
Abstract:
Artificial intelligence and digital twin technologies have emerged as transformative tools for engineering systems that require accurate prediction, real-time monitoring, and adaptive decision-making. In tribology, conventional wear mitigation approaches rely heavily on empirical testing, simplified analytical models, and periodic condition-based maintenance strategies. While these methods have supported decades of engineering practice, they struggle to capture the multi-scale, multi-physics nature of friction, wear, and lubrication under dynamically changing operating conditions. Limitations such as delayed fault detection, limited adaptability to unseen regimes, high experimental cost, and an inability to integrate real-time system feedback motivate the need for more intelligent and responsive frameworks. Recent advances in artificial intelligence-enabled digital twins address these challenges by combining physics-based models, real-time sensor data, and data-driven learning algorithms within a continuously updated virtual representation of the physical system. In tribological applications, such frameworks enable near real-time prediction of frictional behavior, wear evolution, lubrication regime transitions, and remaining useful life, while supporting adaptive control and optimized maintenance decisions. This review consolidates the current research literature on artificial intelligence-enabled digital twin applications in tribology and lubricated mechanical systems, with emphasis on sensing technologies, hybrid modeling approaches, data management strategies, and system architectures. Key challenges, including data quality, computational burden, model interpretability, and scalability, are discussed alongside emerging solutions. Through an in-depth synthesis of peer-reviewed studies, this paper highlights how the integration of artificial intelligence and digital twins offers a robust pathway toward more reliable, efficient, and sustainable tribological systems.
Keywords:
Artificial intelligence, Digital twins, Tribology, Lubricated mechanical systems, Multi physics modeling, Condition monitoring, Wear prediction, Data driven optimization
