Drivers of bus rapid transit systems – Influences on patronage and service frequency

This paper reports the findings of a comparative analysis of bus rapid transit (BRT) performance using information on cross-section data of 121 BRT systems throughout the world, in which random effects regression is employed as the modelling framework for stand alone patronage and ridership models, and 3SLS for joint models in which frequency is treated as an endoneous effect on patronage. A number of sources of systematic variation are identified which have a statistically significant impact on BRT patronage in terms of daily passenger numbers such as fare, frequency, connectivity, pre-board fare collection, and location of with-flow bus lanes and doorways of a bus. In addition to the patronage model, a bus frequency model is estimated to identify the context within which higher levels of service frequency are delivered, notably where there exists higher population density, more trunk lines, the corridor provides bus priority facilities such as priority lanes for many bus routes, and where there is the presence of overtaking lanes at more than half of all stations along the heaviest section of the corridor. The findings offer important insights into features of BRT systems that are positive contributors to growing patronage which should be taken into account in designing and planning BRT systems.