Dynamic Vehicle Dispatching at a Transfer Station in Public Transportation System

Dynamic vehicle dispatching at the transfer station can improve the transit service quality by optimizing the transfer coordination of routes. In this paper, a dynamic vehicle dispatching model is proposed that aims to minimize the total waiting time of passengers at the transfer station and the downstream stops. A prediction model based on support vector machines (SVM) is also developed to forecast the arrival time of the next vehicles at the transfer station. To reduce the disconnected cases between the transfer routes, an SVM-based model is introduced to forecast the elastic time for the estimated arrival time at the transfer station. According to the estimated arrival time and elastic time, vehicles are dispatched in a dynamic way to reduce the total waiting time of passengers. The dynamic vehicle dispatching approach at the transfer station is examined with the data of three transfer routes in the city of Dalian, China. Results show that the approach proposed in this paper can reduce the total waiting time of passengers at the transfer station and the downstream stops.