Published In

Support for this project was provided by the U.S. Department of Transportation and the Tri-County Metropolitan Transportation District of Oregon (Tri-Met).

Document Type


Publication Date



Tri-County Metropolitan Transportation District of Oregon, Transportation -- Oregon -- Portland -- Planning, Local transit -- Technological innovations, Intelligent transportation systems

Physical Description

25 pages


While automatic passenger counters (APC's) offer the potential for cait effective data recovery and management, they also introduce new complications in the data recovery process. This report addresses three issues associated with the implementation of APC's, based on an evaluation of the recent experiences of the Tri-County Metropolitan Transportation District of Oregon (Tri-Met). First is the issue of validation, which is concerned with both the recovery and accuracy of APC passenger data. The second issue concerns the development of a sampling methodology for APC's compatible with UMTA's Section 15 reporting requirements. Third is the issue of inferring system-level ridership from sample data in the presence of selective APC failures.

We find that the APC's are providing systematically accurate passenger counts. Analysis of the data recovered from September to November 1988 also shows that sampling was representative, based on the set of "trains" from which ridership data were successfully recovered. The initial selection/assignment of trains, however, was not representative.

Given that APC's record operating data for all bus trips comprising a train assignment, a cluster sampling method is formulated that ensures an overall random selection of bus trips via a random first stage selection of trains.

Selective data recovery failures can hamper the process of inferring system-level ridership from the sample estimates. For example, when failure rates vary by bus type or time of day, inferences drawn from the sample of recovered data may over or under-represent total system ridership. In such circumstances, post hoc stratification of the sample data may be required. We outline several alternative corrections based on a-priori knowledge of the mix of bus types and schedule characteristics in the system.


Final report. Catalog Number PR037.

A product of the Center for Urban Studies, College of Urban and Public Affairs, Portland State University.

Persistent Identifier