Joint design of timetabling and capacity for mixed transit system with autonomous and human-driven buses considering passengers’ acceptance

Abstract

Human-driven buses (HBs) with fixed-capacity are difficult to adapt to time-varying demand, resulting in overcrowding and excessive operating cost. With the introduce of self-driving technology in public transit, autonomous buses (ABs) with flexible capacity are intended to alleviate this problem. In this paper, based on a mixed transit system (MTS) with ABs and HBs, a mixed integer nonlinear programming (MINLP) model is formulated for jointly optimizing timetabling and capacity, by taking into account time-varying demand and passenger acceptance of HBs and ABs. The objective function is to minimize the operating cost and passenger cost. Genetic algorithm is employed to solve the model. The results show that MTS, compared to HBs system, can reducing total cost by 10.57% during peak periods, increase the frequency by 29.73%, and improve full load factor by 7.45% during off-peak periods. The sensitivity analysis of passenger acceptance reveals that the total cost of MTS tends to decrease when the proportion of passengers who accept ABs increases.