In the United Kingdom, and throughout the world, tunnelling in urban environments is essential to provide necessary infrastructure. Further, major infrastructure projects, many of which will require tunnelling, will be likely to increase with continued urbanization. However, tunnelling can place buildings at risk. Therefore, reliable assessment tools to predict tunnelling-induced damage are essential to determine whether protective measures are necessary. Useful models exist for predicting tunnelling-induced damage, but these methods need refinement, making use of new and existing data and increasing computational resources. The primary aim of this proposal is to obtain the necessary experimental data to improve analytical and computational methods of predicting tunnelling-induced settlement damage. Specifically, this will be achieved through three primary steps. First, centrifuge testing, in which small scale models can be used to investigate full scale structures, will be used to evaluate the interaction of the soil and the structure during the tunnelling process. In particular, realistic scale models will be 3D printed and the tunnelling process directly simulated. Second, these physical tests will be simulated computationally, to evaluate and improve computational methods. Third, additional simulations will allow extension of experimental results to a wide range of tunnelling scenarios. The experimental and computational results will be directly used to create improved analytical and computational guidelines for predicting tunnelling-induced settlement damage. These guidelines will be evaluated with respect to existing field data. It is envisioned that these results will have an immediate impact on the tunnelling industry in the UK and around the world.