During the past decade, the rapid construction of rail transit in developing countries has had a great impact on bus transport, resulting in the waste of public transport resources and ineffective competition between bus and rail transit. Therefore, it is particularly important to explore ways to promote the integrated development of bus and rail transit by combining both the effects of the built environment and their competition. Using Wuhan city in China as a case study, this study develops gradient boosting decision trees (GBDTs) to examine the quantified threshold range of the built environment impacts on bus transit trip rates and to further explore the competitive impacts of rail transit on bus transit within these threshold ranges. The threshold range results show that when the number of service point of interests (POIs) within a 500-m radius of the bus stop reaches 350 and the distance to the destination is 5 km, the model split of bus transit is the highest. The competition impact analysis shows that rail transit promotes bus transit trip rates in urban fringe areas and locations, including a high proportion of 3- to 5-km middle-distance trips, but restricts bus transit trip rates in high-density central areas and the regions dominated by short-distance trips of less than 3 km or long-distance trips of more than 5 km. Strategies are proposed to optimize land use and the layout of bus and rail transit stations to mitigate ineffective competition and improve the efficiency of public transit to provide theoretical and technical support for optimizing urban spatial layouts through transit-oriented development.
Read full abstract