Modelling and accelerating the passenger evacuation process in a bi-modal system with bus and e-hailing modes after mass gathering events

Abstract

After mass gathering events, passenger trip demand surges immediately within a small area. While organized bus bridging services would be usually offered in such circumstances to speed up passenger evacuation, many people prefer the more comfortable e-hailing service to leave the area. However, the use of e-hailing services for speedy passenger evacuation is a double-edged sword. On the one hand, e-hailing vehicles supplement to buses by offering more transportation supplies; on the other hand, the large inflows of e-hailing vehicles and the time-consuming passenger-driver (offline) searching process would lead to a huge vehicle accumulation within the area, thus lower vehicle speed and reduce bus evacuation efficiency. This paper proposes a mathematical model to depict the dynamic passenger evacuation rates and traffic conditions in a bi-modal system with both bus and e-hailing services for passenger evacuation. Numerical examples show that e-hailing service could significantly deteriorate the evacuation process when the arrival rates of e-hailing vehicles are high. Thus we further propose two feedback based perimeter control strategies that control the total inflow rates of e-hailing vehicles and background traffic, based on the vehicle accumulation and the number of parked vehicles within the area respectively. As shown by simulation experiments, both perimeter control strategies can effectively improve evacuation efficiency in the presence of e-hailing vehicles, especially when bus frequency is low, passenger demand is high, and/or the number of parking spaces is limited.