Transit operation optimization in the context of connected and automated vehicles: A case study of Beijing

Connected and automated vehicles (CAV) initiatives are revolutionizing the entire transportation society. Not limited in the area of passenger cars, public transit agencies also seek for the possibility of replacing tremendous manpower with the state-of-the-art "self-driving" technologies. In this study, we developed an optimization model to quantify how the CAV benefits the route-level transit operations if all current non-automated buses are converted into autonomous buses. A bi-objective mixed integer programming model is established to minimize passenger waiting times and maximize passenger flows considering all buses are running on a bus dedicated lane. By taking one actual bus route in Beijing as a case study, we computed the operational performance indicators of non-automated bus based on smart card and GPS data, and compared with the optimized bus routes with CAV technologies. The findings reveal that autonomous buses can significantly reduce vehicle usages, reduce bus bunching, increase passenger loads, and lower average passenger waiting time.