Replicating Bus Trajectories in SUMO based on Tracking Experiments with Implications for Energy Consumption

Abstract We present a methodology for creating a bus route microsimulation model from open-source data to assess electric bus (e-bus) energy consumption under varying traffic, vehicle, and driving behavior conditions. The approach aims to provide transport planners and researchers with a testbed for evaluating the impact of e-bus adoption. While datasets such as OpenStreetMap (OSM) and GTFS enable large-scale modeling, they often require corrections, particularly to speed limits, to ensure realistic outcomes. For e-buses, accurate energy estimation depends on calibrating driver behavior using observed speed trajectories, making parameter fitting and network refinement essential. Our case study focuses on Kyoto, Japan, replicating operations of the city’s only current e-bus line. Combining corrected map data, limited tracking experiments, and traffic data, we generate realistic simulation scenarios with the opensource microsimulation SUMO. Speed trajectories from tracking are used to calibrate vehicle and driver parameters to explore vehicle- and route-specific energy use. The proposed method is transferable to other contexts. The results demonstrate that realistic e-bus simulations can be achieved with modest data inputs, but that OSM network data need to be carefully revisited by the analyst and that calibration of driver behavior is important to obtain realistic energy consumption estimates.