Abstract
The development of Digital Twins has become a central topic in digital transformation, offering new possibilities for the prediction, control, and optimization of physical systems. In a general sense, a Digital Twin features real-time data exchange between a precise replica of a physical object in the virtual world, and vice versa. The virtual replica of the physical entity is usually referred to as Digital Model. A faithful representation can however affect the computational effort required by the model and consequently the data exchange in real-time. This is why, in recent years, the evolution of kinematic and dynamic models of mechanical systems in Digital Twins is essential but challenging, particularly due to the trade-off between model fidelity and computational feasibility required for real-time integration. This paper presents a systematic literature review of the methods and practices employed in modeling such systems for Digital Twin applications, focusing on tools and simulation environments used, as well as communication and computational challenges. As a final contribution, a theoretical framework is proposed as a guidance in the development of fully integrated Digital Twins for mechanical systems.