TransitVision: Approximating Vehicle Locations Using SIRI-SM Real-Time Data

Abstract

Many public transportation agencies around the world track their vehicles using GPS. However, the GPS data is usually not directly accessible by passengers, but is used to provide real-time arrival estimates. This thesis attempts to approximate the locations of vehicles using the available real-time arrival estimates, through a smartphone application named TransitVision. By utilizing the SIRI standard(www.siri.org.uk), TransitVision is interoperable with other public transportation agencies.

A preliminary study on travelers' habits and perceived waiting time, created a solid foundation for TransitVision. This thesis also includes a study into the state-of-the-art, to examine other applications, technologies and discover limitations. The study included a thorough examination of several SIRI implementations, to consider how to incorporate them in TransitVision. Finally, 15 testers evaluated the application's usability through the System Usability Scale(http://www.measuringu.com/sus.php).

The preliminary study indicated that mobile transit application users believed they waited on average five minutes, which was about the same as those using non-digital retrieval methods. TransitVision was initially developed for Oslo, but was also tested on transit data from Tampere. It worked well in both cities, confirming TransitVision's interoperability. However, other cities require major modifications to TransitVision. TransitVision achieved a System Usability Scale score of 86, which is close to a superior score.