Use of Mobile Ticketing Data to Estimate an Origin–Destination Matrix for New York City Ferry Service

One of the fundamental components of transit planning is understanding passenger demand, which is commonly represented with origin–destination (O-D) matrices. However, manual collection of detailed O-D information through surveys can be expensive and time-consuming. Moreover, data from automated fare collection systems, such as smart cards, often include only entry information without tracking where passengers exit the transit network. New mobile ticketing systems offer the opportunity to prompt riders about their specific trips when they purchase a ticket, and this information can be used to track O-D patterns during the ticket activation phase. Therefore, the objective of this research is to use back-end mobile ticketing data to generate passenger O-D matrices and compare the outcome with O-D matrices generated with traditional onboard surveys. Iterative proportional fitting was used to create O-D matrices with both mobile ticketing and onboard survey data. These matrices were compared using Euclidean distance calculations. This work was done for the East River Ferry in New York City, and the results show that during peak periods, mobile ticketing data closely match survey data. However, in the off-peak period and during weekends, when travelers are more likely to be noncommuters and tourists, matrices developed from mobile ticketing and survey data have greater differences. The impact of occasional riders making noncommute trips is the likely cause for these differences, because commuters are familiar with using the mobile ticketing product and occasional riders are more likely to use paper tickets on the ferry service.