A framework for railway transit network design with first-mile shared autonomous vehicles

Document Type

Journal Article

Publication Date


Subject Area

mode - rail, economics - operating costs, infrastructure - fleet management, ridership - commuting, ridership - demand, planning - methods, planning - network design


Railway transit network design, Shared autonomous vehicles, Dynamic travel demand, Vehicle relocations, Mixed-integer programming, Fixed-point method


Providing a railway transit system (RTS) in less populated areas is a challenging task for transportation agencies due to its high construction and operating costs. With the advent of automation, shared autonomous vehicles (SAVs) as an integral part of public transit services has the potential to enhance the design of transit systems. In this paper, we present a joint optimization framework of railway transit network design and SAV first-mile service that minimizes the total cost of the combined RTS-SAV services and commuters’ waiting time, while serving a dynamic travel demand in the network. The proposed model optimizes the SAV fleet size and the RTS alignment while enabling vehicle relocations to tackle the vehicle imbalance issue in the SAV service. Due to the non-linear and mixed-integer formulation, we develop a fixed-point algorithm for this joint RTS-SAV problem where we transform the original problem into a mixed-integer linear programming (MILP) formulation. Our results indicate that the joint RTS-SAV services can be constructed and operated at a lower cost than either of the RTS or SAV services alone. Furthermore, the resulting joint RTS-SAV services are underpinned by a shorter railway alignment and larger fleet size rather than a multi-link extension. Additionally, the joint RTS-SAV services is robust to the variation in total demand, with respect to the railway alignment, SAV utilization and commuters’ waiting time.


Permission to publish the abstract has been given by Elsevier, copyright remains with them.


Transportation Research Part C Home Page: