When Did the Train Arrive? A Bayesian Approach to Enrich Timetable Information Using Smart Card Data

Smart card data from the Automatic Fare Collecting systems (AFC) and timetable information, such as Vehicle Location (AVL), are used in combination by practitioners researchers to gain a deeper understanding of public transit network. In some cases, AVL not available due records being missing system. people resort schedule General Transit Feed Specification (GTFS) match smart Since delays or changes contained scheduled timetable, it can result wrong matches between This paper shows how uncertainty arrival departure times affects passengers train assignments proposes method for estimating time trains when recorded information is available. The uses knowledge tap-outs distributed hierarchical, latent Bayesian model predict trains. Evaluated on 15,136 arrivals, infer 70% arrivals with an average error 28 32 seconds depending station.