Quantitative Assessment of Penetration Rates of CCAM Applications on GHG Emissions in EU27

Due to significant climate change in the past decades, the reduction of Greenhouse Gas (GHG) has been set as the top priority. In recent years, many Connected, Cooperative and Automated Mobility (CCAM) applications have been proposed and developed to improve the network and increase traffic efficiency. By following the net-zero emission reduction by 2050 in European Union (EU), a slowly increased percentage of connected vehicles (penetration rate) should be considered, and yet, most studies only show benefits of CCAM at 100% penetration rate, i.e., under assumption that all vehicles are connected. Therefore, we study the effect of different penetration rates for CCAM applications in the urban scenario. In this work, some representative German cities are chosen and classified into three Tiers based on the population. Then the urban scenarios is built in Simulation of Urban MObility (SUMO), and the CCAM applications control the vehicles via Traffic Control Interface (TraCI). Simulation results show a significant reduction in CO2 emissions and an increase in vehicle speed. Then, these results are fed into a Machine Learning (ML)-based estimator, called Random Forest Regressor (RFR), to predict further improvements of CCAM at the 27 European Union Countries (EU27)-level. The predicted results can act as references or guidelines for developing or deploying connected vehicles to reduce GHG and improve traffic efficiency.