Air pollution is one of the emerging problems in our society. (Plug-in) hybrid electric propulsion concepts offer the possibility to reduce emissions as well as total cost of ownership (TCO) for public transport vehicles (e.g. bus application). In this context, typically diesel engines are used to form the hybrid propulsion system of the vehicle. Gaseous fuels represent an attractive alternative to diesel for both ecologic and economic reasons. In the framework of the research project 'Optimised Real-world Cost-Competitive Modular Hybrid Architecture' (ORCA), which was funded by the EU, two different hybrid electric propulsion concepts have been investigated: one using a diesel engine as internal combustion engine and one using CNG engine. The aim of the current study is to analyse specific benefits for the aforementioned hybrid propulsion systems for representative driving scenarios with regard to emissions (e.g. NOx and CO2) and total cost of ownership in bus application. Engine models based on experimental data for diesel and CNG were developed to estimate the emissions from engine torque and rotational speed. For designing optimal energy management strategies for each propulsion system, maps-driven quasi-static models for specific engine types are used in this study. This paper compares the two aforementioned propulsion concepts regarding their CO2 and NOxemissions. This comparison is performed for relevant bus missions (urban, suburban, with and without zero-emission zone) and with different energy management strategies. In addition to the emissions, also the downsizing potential of the combustion engine has been analysed to minimize the powertrain TCO (pTCO) for plug-in hybrid electric buses. The results of the performed analyses show that the hybrid propulsion concept using the CNG engine shows advantages both with respect to emissions as well as to pTCO. The pTCO is 10% lower, CO2 emissions are 13% lower and the NOx emissions are more than 50% lower than with the diesel combustion engine. These results are consistent across all usage profiles under investigation.