Rail Transit Ridership: Station-Area Analysis of Boston’s Massachusetts Bay Transportation Authority

Abstract

This paper examines factors associated with transit ridership in Boston, Massachusetts. With the use of automated fare card data and boarding counts, this paper specifies and estimates direct-demand models of station-level ridership for Boston’s urban rail system during different time periods (e.g., a.m. peak and off-peak). The models include transit infrastructure and service characteristics, such as number of bus connections, as well as station-area characteristics aiming to represent economic activity, the built environment, and socioeconomics and demographics. The results show that population and employment are important drivers of overall daily demand and also show the importance of street design characteristics, various transit levels of service indicators, transit network accessibility, and station location relative to the central business district. Some variation in influencing factors was detected across different time periods and for weekday versus weekend boardings. The models reasonably replicate total observed boardings, but important under- and overpredictions at particular stations suggest areas for additional research. Applying the model to a light rail extension currently under development in the region predicts total weekday ridership up to 47% higher than official forecasts. The paper concludes with areas for future research.