Digital twins are increasingly used to create dynamic virtual counterparts of engineering assets, enabling continuous performance assessment, predictive maintenance, and informed decision‑making. They combine computational models with measurement data to represent the evolving state of materials, components, and systems within a single, consistent framework. This contribution outlines a physics‑based digital twin concept for viscoelastic materials [1], subjected to cyclic loading. Building on earlier work on reduced‑order, data‑assimilative digital twins for linear elastic materials [2], in a material‑testing context, the present study extends the framework to time‑dependent constitutive behaviour. A finite element model with an appropriate viscoelastic description is employed to capture creep, relaxation, and degradation indicators relevant for fatigue and durability assessments. To address computational cost, especially for long load histories and parametric studies, model‑reduction techniques are introduced so that the essential viscoelastic response can be evaluated efficiently while retaining the link to the underlying governing equations. Measurement data from material testing are used to calibrate and update the digital twin over time. Within a data‑assimilation or inverse‑modelling setting, sensor readings are incorporated to adjust material parameters and correct model predictions, thereby improving robustness under uncertainty and variability. The proposed framework illustrates how combining physics‑based viscoelastic models, reduced‑order representations, and data integration can support condition monitoring and predictive maintenance of polymer‑based components in a broad range of engineering applications. REFERENCES [1] J.C. Simo, T.J.R. Hughes, Computational Inelasticity, Springer, 1998. [2] R. Aylwin, N. Beranek, G.B. Budihala, M. Mettenleiter, G. Oruc, K. Urban, “Building a digital twin for material testing: Model Reduction and Data Assimilation”, Proc. Appl. Math. Mech., 2025 (in production), DOI: 10.1002/pamm.70063.