The challenges of urbanization and climate change necessitate more sustainable and efficient transport systems, which can be evaluated and optimized in a robust simulation environment before real-world implementation. In this context, a city-scale digital twin can map physical entities and their attributes, structure, state, performance, function, and behavior to the virtual world. It creates a high-fidelity, dynamic, multi-dimensional, multi-scale, and multi-physical model that effectively translates between the real and virtual worlds. Gaming engine-based digital twins can offer visually appealing simulation environments integrating agent-based models widely researched in transportation planning. A digital twin of King Abdullah University of Science and Technology (KAUST) has been developed to assess transportation emissions on campus. Origin/destination data extracted from the simulation were compared to Google Distance Matrix API information. This paper investigates the viability of using the KAUST campus digital twin as a tool for fleet emissions management. The model resulted in a distance deviation of ±0.4 kilometers and a time deviation of ± 10 minutes for the majority of randomly selected trips across the campus for 24 hours. The deviation in distances between the KAUST Digital Twin (KDT) and Google Maps trips results in an error of only 116 g of CO 2 per trip. These results suggest that the model provides a potentially accurate simulation environment and hence a credible approach to fleet emissions management. Thus, the KDT can manage fleet emissions, improve transportation efficiency, and improve city-scale performance. Finally, the KDT can contribute towards quantifying the impact of deployment policies and urban planning strategies on a sector basis and their interactions with the UN SDGs 9,11, and 13, particularly in areas currently underserved by existing applications.