"Weld modelling" is a powerful tool in understanding the structural performance of welded structures. Conventional continuum-mechanics-based predictions of the stresses generated by welding have achieved considerable success in understanding the in-service performance and degradation mechanisms of welds in the UK's nuclear reactor fleet. However their practical use is currently limited to materials that do not undergo so-called solid state phase transformation (SSPT) during welding, since the presence of SSPT makes it necessary to predict changes in the material microstructure in order to predict the stresses. In addition, the microstructural changes imposed by welding have a profound influence on a weld's resistance to creep, thermal ageing, oxidation, stress corrosion and other in-service degradation mechanisms, and upon its sensitivity to the presence of cracking. The Fellowship research programme aims to extend conventional weld modelling into a multi-disciplinary tool that can predict both continuum parameters such as stress & distortion, and microstructural parameters such as grain size and shape, the occurrence of secondary phases, and precipitate distributions, and hence both directly predict long term structural performance and be used for "virtual prototyping " of weld processes and procedures for novel welding processes. Success offers the prospect of better understanding of in-service performance of welds in both the existing UK nuclear reactor fleet, and in any industrial sector where the long term structural performance of welds is important. It will also aid the choice of weldment materials, joint design and welding process for structural welds in new-build nuclear power plants, and in advanced Generation IV designs that may be built on a longer time frame.