Feeder bus network design with modular transit vehicles

Abstract

Modular transit vehicles have the potential to transform public transport systems. Indeed, the possibility of dynamically adjusting capacity by assembling and disassembling multiple modular pods allows for improved transit service flexibility and efficiency, reducing operator and passenger costs. As such, modular buses are considered particularly advantageous for settings with large variations in passenger demand, permitting en-route capacity adjustment and seamless transfers. In this context, this study presents a model for the design of a feeder bus network operated by autonomous modular buses, accounting for en-route transfers, and a Genetic Algorithm to solve the non-linear mixed integer programming problem arising. Results for a case study in Athens, Greece for an area served by three metro lines are presented and discussed.