Using General Transit Feed Specification-Ride to Identify Riders’ Use of Transfer Opportunities

Abstract

Understanding how riders use a transit agency’s services is central to providing effective service. Although the ideal experience for riders may not include transfers, these may be necessary to connect them from their origin to their destination. Previous methods have identified key hidden transfer locations within transit networks. However, there has been little effort to develop tools that enable small- to mid-sized agencies that typically lack access to sophisticated data sources to conduct this analysis. This research introduces a methodology for identifying transfer opportunities using a combination of statistical analyses of nonindividually identified automated passenger counter data to compute a transfer metric representing the association one service has with another through shared passengers. A unique aspect of this work is the utilization of ridership data compliant with the data standard General Transit Feed Specification (GTFS)-ride, which captures historic states of the transit network with associated ridership levels. GTFS schedule data from an Oregon transit agency were employed to identify transfer opportunities by assessing the probability of a transfer based on transfer time, an upper limit on transfer distance, and a new metric that measures the geographic coverage gains made by a particular transfer. A final transfer metric was calculated and compared against recently collected survey data. The key contribution of this work is the identification of transfer opportunities that lie outside traditional transfer hub locations. The resultant transfer metric will enable transit service planners to conduct regular analyses of their network, identifying key transfer locations and opportunities for further development.