Tapping into Delay: Assessing Rail Transit Passenger Delay with Data from a Tap-In, Tap-Out Fare System

Abstract

How should customer delay on a rail transit system be measured? Research into quantifying reliability and passenger delay often defines “delay” as when speeds or travel times fall above or below a fixed threshold. When applied to a rail transit system, however, this technique may not fully capture the impacts on all transit riders. This paper proposes a new way to assess passenger delay by leveraging the power and richness of transaction-level fare system data not always available in other contexts. This method enables a true understanding of the full spectrum of the impact on transit passengers by an analysis of the entire distribution of passenger travel times rather than by a simple tally of individuals with travel times above a threshold. This method—comparing the cumulative distribution of travel times—combines concepts from traffic queueing theory and travel time reliability research in both the transit and the highway arenas. The proposed method was applied to fare system data from the Washington, D.C., Metropolitan Area Transit Authority’s Metrorail system to evaluate the delays that arose from using different strategies for providing continued service during rail system rehabilitation.