Transit Network Complexity in the Context of Transit Itinerary Inference using Smartphone Travel Survey Data

Researchers and planners have taken great interest in the rich data resource smartphone and GPS device travel surveys now represent. Developing ways of interpreting this data has become a hot topic recently with methods such as transit itinerary inference (TII) from GPS rider trajectories and GTFS becoming useful tools for travel behavior analysis related to planning. This exploratory research aims to provide practitioners with metrics to quantify a factor that complicates bus route inference in different TII processes: overlapping routes in GTFS data. Accurate route inference is difficult in regions where rider data coincides with overlapping routes and various TII approaches has been tested to address this issue. By applying existing network overlap indicators and two proposed measures of network complexity, a ranking of transit networks is generated according to a network’s propensity to generate what researchers have referred to as “route ambiguity.” Revealing overlap and quantifying the degree of overlap on road links is achieved through the application of two proposed measures: The Overlapping Routes on Links (OROL) index, and the Probability of Passage (POP) score. The latter’s output is seen as one way to improve route inference rates in TII. These measures build off the traditional Line Overlapping Index and improve upon it by providing previously unobtainable road-link level detail. To ensure accurate analysis within and between networks, an additional novel procedure is developed which converts GTFS data into a simplified representation, replacing the need for road network layers entirely.

  • Supplemental Notes:
    • This paper was sponsored by TRB committee ABJ30 Standing Committee on Urban Transportation Data and Information Systems.
  • Corporate Authors:

    Transportation Research Board

    ,    
  • Authors:
    • Davey, Marshall
    • Yazdizadeh, Ali
    • Patterson, Zachary
    • Fitzsimmons, Kyle
  • Conference:
  • Date: 2019

Language

  • English

Media Info

  • Media Type: Digital/other
  • Features: Figures; References; Tables;
  • Pagination: 20p

Subject/Index Terms

Filing Info

  • Accession Number: 01697521
  • Record Type: Publication
  • Report/Paper Numbers: 19-03089
  • Files: TRIS, TRB, ATRI
  • Created Date: Dec 7 2018 9:30AM