Identification of transit farebox data errors: impacts on transit planning

Shu Yang, Yao Jan Wu, Bernadette Marion, Isaac E. Moses

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


Transit agencies require a constant stream of operations performance data to support standard planning, scheduling and operations management activities. Ridership and revenue statistics play a critical role in strategic system design, policy development, and budgeting decisions at all levels of transit management. Many agencies rely on electronic fare collection devices as a primary source for ridership and revenue data. The quality of this data will greatly affect transit-related reporting and decision making. This study proposes a systematic, data-driven approach to process revenue and ridership data pulled off electronic farebox equipment installed on a bus fleet operating in the St. Louis region. Three major farebox data errors are identified and impacts of these data errors are further evaluated and discussed at the system and trip level. Results indicate ridership and revenue may be overestimated by up to 8.05 and 9.95 %, respectively, due to farebox data errors. The results of this development effort offer a range of low-cost error identification and processing techniques that transit staff could easily and quickly implement. Even though the St. Louis Metro Transit data was used for analysis, these proposed approaches can be considered as a general framework and used by other transit agencies.

Original languageEnglish (US)
Pages (from-to)457-473
Number of pages17
JournalPublic Transport
Issue number3
StatePublished - Dec 1 2015


  • Data quality assurance
  • Farebox data
  • Minimum covariance determinant
  • Ridership
  • Transit planning

ASJC Scopus subject areas

  • Information Systems
  • Transportation
  • Mechanical Engineering
  • Management Science and Operations Research


Dive into the research topics of 'Identification of transit farebox data errors: impacts on transit planning'. Together they form a unique fingerprint.

Cite this