Effects of system configurations of automated fare collection on transit trip origin–destination estimation in Greater Toronto and Hamilton Area

Document Type

Journal Article

Publication Date


Subject Area

place - north america, place - urban, mode - bus, mode - rail, technology - ticketing systems, technology - passenger information, ridership - modelling


Automatic fare collection (AFC), System configuration, Rail service, Bus service, Origin-destination estimation


Automated fare collection (AFC) systems such as smart cards are becoming increasingly popular among transit agencies worldwide. Two main configurations of AFC systems can be found, characterized by whether users are required to scan their cards at the beginning of their trips (i.e., entry-only system) or both at the beginning and end of their trips (i.e., entry–exit system). Recently, there has been growing interest in implementing the latter configuration in order to provide more equitable fare structures that charge users based on distance, while arguably providing more accurate data for an origin-destination analysis of users. Therefore, this study explores the spatial and temporal differences in transit users’ origin-destination estimations that are based on the two AFC system configurations. To achieve this goal, it uses AFC system data collected from GO Transit, the operator of the regional commuter rail and bus systems in the Greater Toronto and Hamilton Area (GTHA). A comprehensive model was developed which suggests that the entry–exit AFC system configuration helps in estimating about 14% more trips than an entry-only AFC system configuration; the latter uses a set of common assumptions that are normally employed by researchers in estimating users’ destinations. Spatial, temporal and mode related differences between the two system estimations were also found. This paper offers policy makers and planners a better understanding of the possible benefits/impacts of implementing entry–exit systems vs. entry-only systems on transit users’ origin-destination estimations.


Permission to publish the abstract has been given by SpringerLink, copyright remains with them.