Collaborative Optimization of Community Bus and Flexible Feeder Bus Connection System

Along with the development of public transportation network in china, the connection between rail transit and conventional buses still needs further improvement. Designing an efficient feeder-bus network is essential for improving metro hubs’ distribution capacity. This paper divides the operation of the flexible feeder bus system into two stages. The first stage is to cluster travel demands with similar spatiotemporal characteristics, and then construct RLMP, SPS problems by using the column generation algorithm, which can reduce system travel cost and complete vehicle route planning at the same time. In the second stage, the mixed decision optimization of vehicle service attributes is carried out based on the reverse auction algorithm, which can minimize its impact on community bus and meet the demands of flexible bus operation. Finally, based on the bus operation data of three community bus routes near a subway station in Chengdu, it's verified that this model has good effects both on operational efficiency and benefits.