Estimating Ridership of Rural Demand–Response Transit Services for the General Public

Abstract

The general objective of this study was to develop a model for estimating ridership for rural demand–response transit services for the general public. A lack of data for demand–response service characteristics and geographic coverage has limited the estimation of such models. This study developed and estimated two models. The first was estimated with data from the 2013 rural National Transit Database, and the second was estimated with more detailed service data collected from surveys of transit agencies. Results showed that in addition to total population, demographic characteristics were important. Ridership was found to increase significantly when the percentage of the population composed of older adults or people without access to a vehicle increased. Both models showed a negative effect of fares on ridership. The second model analyzed the effects of service span and reservation requirements on ridership. Results showed that providing more days of service had an expected positive effect on ridership; allowing users to reserve rides on shorter notice also had a significant positive effect. Compared with previous research, the inclusion of a greater number of variables and more specific service information improved the performance of the models.