Multi-physics simulations in the urban environment make different key-performance indices accessible which help to understand the performance of cities. Many different tools are available that enable different kinds of simulations in the built environment. These are often very specific to single physics, require expert knowledge, are based on different input data and produce software-specific result files. Digital twins of cities facilitate multi-domain or multi-physics simulations by providing the common input data for the different kinds of simulation and gather the simulation results for combined data accumulation, storage, post-processing, and visualization. However, there is still a lack of standardization of the input and output data models and data formats which makes interoperability and data exchange difficult. Within the Digital Twin Cities Centre (DTCC), funded by the Swedish agency for innovation systems Vinnova and hosted by Chalmers university, the digital twin concept applied to cities is in focus. Utilizing the core technology of the DTCC platform for generating triangulation surfaces for buildings and the terrain from Lidar data and building footprints from the Swedish Mapping, Cadastral and Land Registration Authority Lantmäteriet, we develop and utilize software for multi-physics simulations. In this work, we present our approach to enable automized workflows for three different simulations tools, namely CFD-based urban comfort simulations, agent-based pedestrian movement simulations and SPH simulations of urban flooding. We also show how we use the synergies of keeping similar data formats for those three simulation tools facilitating simpler data exchange and interoperability between simulation experts, visualization experts and stakeholders. Further, limitations and the need for further research is discussed.