Measures of Travel Reliability on an Urban Rail Transit Network

Abstract

To evaluate the travel reliability of an urban rail transit (URT) in normal operation, three travel reliability indicators are proposed from passengers’ travel. The connectivity reliability, travel time reliability, and capacity reliability indicators are computed from the weighted average of the number of tolerable travel paths, the ratios of 95–50 quantiles of travel times, and effective residual capacity coefficients between pairs of stations, respectively. The weight of an origin-destination (OD) station pair in a certain time period is the ratio of the passenger trips of this station pair to the passenger trips entering the origin station during that period. The tolerable coefficient, which is the ratio of passengers’ tolerable travel time to shortest possible travel time, is proposed to determine the tolerable travel paths. The effective residual capacity coefficient between an OD station pair is calculated with residual capacities and capacities of all tolerable paths between them. The travel reliability indicators are used to evaluate the travel reliability of Chengdu’s URT stations and network during different time periods and for different tolerable coefficients. The correlations among the travel reliability indicators of stations are analyzed. The results show the travel reliability of a URT in different time periods and indicate how that reliability is related to the tolerable coefficient. A station with better connectivity reliability often has higher capacity reliability also. Because passengers may choose different tolerable travel paths when multiple tolerable travel paths exist between station pairs, a station with high connectivity reliability may correspond to low travel time reliability.