Mitigation of global warming is one of the greatest challenges society is facing in the 21st century. As a consequence, different global initiatives to decarbonise the energy system have emerged and have led to major technological innovation and integration of renewable energy sources (RES) into the energy systems. This, as well as increasing energy efficiency, has been identified in the literature as the most promising options for a sustainable energy system. Both are major challenges for current energy systems and their future planning and operation. Modelling can support the necessary transformation process by providing insights into the complex relationships of possible future energy systems with high shares of renewa-ble energy sources. This Ph.D. thesis deals with the modelling of exergy-efficient multi-energy systems (MES). In MES, various energy sources and sectors are linked by appropriate coupling technologies. This holistic approach allows cross-sectoral synergies to be exploited for implementing effi-ciency measures and the integration of renewable energy sources. In such a case, where different forms of energy are considered in one model, exergy is a good criterion for as-sessing resource efficiency because it also considers the second law efficiency. In a first step a comprehensive literature review on the state of the research and the funda-mentals of exergetic optimisation of MES is carried out. Based on the results, the require-ments for a model for exergy optimisation are defined. The cumulative exergy consumption (CExC) concept fits them best. It considers all exergetic expenditures from the raw material to the final product or service. This means that both the exergy expenditures for energy im-port and the expenditures for installing the infrastructure are considered. The existing CExC concept was adapted to create a methodology for exergy optimisation of municipal MES. It is applied to three different case studies. The open source modelling framework oemof is used for modelling. The results have shown that it is important to opti-mise design and operation of the energy system together as an exergy efficient operation is only possible if the design allows it. In addition, the modelling of the boundary conditions is of particular importance. In open systems, such as municipal MES, incorrectly chosen ones may lead to biased results. If the spatial resolution is modelled, different network coverages and limited line transfer capacities can be considered. Depending on the modelling method of the load flow equations, the results and computing times can differ significantly.