Masonry arch bridges and viaducts comprise a significant portion of European rail bridge infrastructure. Routine monitoring and assessment of the existing masonry bridge stock are regularly conducted to ensure the safe operation of railway networks. However, current damage assessment is mainly based on simplistic visual inspections. Furthermore, standard structural assessment methods primarily focus on bridge load capacity, failing to accurately predict the initiation and progression of damage, which is crucial for addressing maintenance and repair operations. This paper applies advanced structural health monitoring and numerical modelling techniques to study the response of an existing viaduct in London, UK. The monitoring of the viaduct has been conducted via digital image correlation (DIC) to capture the response of the analysed structure under train loading. High-fidelity viaduct models have been developed and validated against the results from DIC monitoring and then used to explore the development of damage under serviceability loading and the bridge response to increased traffic loading up to collapse.