Optimization of coordinated intermodal transit networks

A model consisting of two sub-models, for intermodal transit network optimization and for route coordination is developed for jointly optimizing the characteristics of a rail transit route and its associated feeder bus services in a heterogeneous urban environment whose demand and supply characteristics may vary arbitrarily among adjacent zones. Network characteristics (route and station locations) and operating headways are found which minimize total costs, including supplier (e.g., vehicle operating, rail line, and rail station costs) and user (e.g., access, wait/transfer, and in-vehicle costs) costs. The total system cost is optimized with and without capacity constraints. The optimal decision variables (e.g., rail station spacing, bus route spacing, bus stop spacing, rail headway, and bus operating headways) will be determined analytically, while the optimal rail line length and location will be determined numerically. The optimal results are obtained through an iterative convergence test. It is shown that schedule synchronization may significantly reduce the transfer delay at some transfer stations. Coordination within the intermodal transit network is optimized by adding slack times in the schedules of all coordinated routes in order to improve probabilities of successful transfers and reduce the total system cost. Three scenarios, including uncoordinated, fully coordinated, and partially coordinated intermodal transit system operations are analyzed. The optimal transit facility locations (e.g., rail station, bus route and bus stop locations) are determined through a discrete approximation approach which is proposed here to solve the location problem. The total system cost, including coordinated and uncoordinated costs in the route coordination model are minimized numerically. The optimal transit headways (e.g., rail line and feeder bus routes), which minimize total system cost, are determined for uncoordinated transit system operations. The slack times are optimized after the coordinated transit routes are determined. Irregular many-to-many demand distributions, zonal variations in route costs, probabilistic vehicle arrival patterns at transfer stations and stopping at stops, vehicle sizes, and vehicle capacity constraints are considered in the proposed sequential optimization process. The proposed intermodal network and route coordination optimization models, either separately or jointly, provide the capabilities for designing a new intermodal transit network, and for expanding or improving existing transit systems.