Shared autonomous vehicles implementation for the first and last-mile services

Abstract

Many studies have been conducted on the feasibility of shared autonomous vehicles to solve the first-mile and last-mile (FMLM) connectivity problem of public transport (PT). Numerous implementation strategies mainly focused on the operational strategies, with little attention being paid to formulate the optimum strategy of SAV coupled with PT. Thus, this research develops a multimodal SAV implementation that couples with PT by considering the travelers' mode choice preferences and evaluating the integration of SAVs with PT in improving overall transport network performance and PT ridership. The simulation of SAVs was conducted on the road network of Kuala Lumpur during the morning peak hour. Results indicated that SAVs' implementation that coupled with PT increased the PT usage by 3% and reduced personal vehicle kilometers traveled (VKT) by 6%, which may potentially solve the FMLM connectivity and reduce traffic congestion. Simulation through modification of SAV's waiting time, the operation cost of a personal vehicle, and riding cost for SAV were also conducted for a more comprehensive analysis. Results showed that when SAV’s passenger waiting time decreases, the passenger trips for SAVs increase, together with passenger trips for PT. However, when an extreme case was considered, that was a 20% decrease in the SAV's waiting time; a massive increase in passenger trips for SAVs with reduced PT ridership was observed. This potentially contributed to traffic congestion. As such, the results signified without proper planning of SAV-PT integration could worsen existing traffic network's performance.