ABSTRACT
This chapter demonstrates the use of smart card data to study crowding on platforms and its impact on customer experience through a case study using Sydney Opal data. A two-stage modelling approach is developed to develop a crowding profile at the platform level from smart card data. The first stage involves the use of an application program interface (API) and general transit feed specification (GTFS) real-time data to identify a set of possible routes for each passenger’s transaction using tap on/tap off locations and time information. The second stage uses a utility-based assignment rule that accounts for access, egress, in-vehicle times and number of required transfers to assign passengers to trains. The chapter shows how it is possible to estimate crowding at the platform level by combining smart card data with open data (i.e., GTFS and journey planner) for a very complex train network such as the Sydney Trains network where different trains (limited stops, express and all-stop services) share the same track. These results could be useful in designing station layout but also for train timetable planning and economic appraisal of initiatives aiming to improve the customer experience by increasing station and/or platform capacity.
