Enhancing Household Travel Surveys Using Smart Card Data

Household travel surveys are still one of the most important sources of information used to produce detailed profiles of travel behaviors. They often require a lot of resources and lead to the definition of an average day of travel. This average day integrates behaviors which were probed at different weekdays within the period of investigation. Until very recently, few datasets allowed going beyond such estimation. The implementation of smart card (SC) system changes this situation by making available continuous information on transit use. SC systems output data that can be used to observe transactions on the transit network at any moment, on a continuous basis. While lacking details on the traveler, SC data offer the opportunity to enhance some travel survey estimates. Using data from a small Canadian transit authority, this paper compares various indicators estimated with data from the household travel surveys (2005) and data from the SC fare collection systems for the same time period. This study begins with the identification of transportation objects. Then, several indicators are calculated for comparable elements: date, time, route and fare type. The average weekday that is obtained from the travel survey is compared to each day of operation of the SC system. Results show that there are large variations in transit network use between weekdays. For the most important routes (regular adult card holders), the travel survey matches SC data at 5% level. However, for the origin-destination matrix, SC data is more precise and reveals movements not reported in the survey.


  • English

Media Info

  • Media Type: DVD
  • Features: Figures; Maps; References;
  • Pagination: 15p
  • Monograph Title: TRB 88th Annual Meeting Compendium of Papers DVD

Subject/Index Terms

Filing Info

  • Accession Number: 01128710
  • Record Type: Publication
  • Report/Paper Numbers: 09-1229
  • Files: TRIS, TRB
  • Created Date: May 19 2009 7:48AM