MODELING THE IMPACT OF TRANSIT SERVICE UNRELIABILITY ON DEMAND

This paper discusses the effect that bus transit service unreliability has on demand and presents the results of its study. In the research a set of regression models that estimate route-level demand were developed using data collected with Automatic Passenger Counters and Automatic Vehicle Location systems installed on buses, and demographic, socio-economic and land use information from other sources. Eight regression models were introduced according to time of the day and whether the route was inbound or outbound. Focusing on service related variables, especially bus departure delay variable, the regression results showed this turned out statistically significant in all four outbound routes. The estimation of the coefficients in this case showed a negative value, which in turn proved that in some cases the unreliability does result in a decreased demand.