Modeling Riders' Behavioral Responses to Real-Time Information at Light Rail Transit Stations

An advanced passenger information system (APIS) can play a significant role in improving the satisfaction of transit riders in the short term and increasing ridership in the long term. This research focuses on investigating riders' behavioral responses to en route real-time information on light rail transit (LRT). A survey was designed and conducted to collect LRT riders' behavioral responses by presenting hypothetical scenarios in Calgary, Alberta, Canada. Two scenarios were examined: an estimated arrival time of 10 min for the next LRT and an LRT service interruption attributable to an incident or weather with no information on expected recovery time. The survey collected 505 responses. Four multinomial logit models were developed and calibrated to explore the factors affecting trip decision making for the described scenarios for commuter and noncommuter trips. The results led to the conclusion that various socioeconomic attributes (e.g., age, gender, and number of autos per household), experience with an APIS (familiarity with APIS and perceived accuracy of APIS), and experience with transit and the LRT system (use of transit as the primary mode of transportation, frequency of LRT use, and familiarity with LRT) had strong influences on travelers' behavioral responses in the context of real-time LRT information. Analysis of the data also determined that travelers' actions varied by trip purposes, travel time, and weather conditions.