Simulation of demand and supply of urban rail in a multimodal environment

High-frequency rail transit systems play a key role in dense urban transportation. As such, decisions made in design, implementation, and operation may precondition the successful performance of the overall urban system. To fully understand the impacts of these decisions there is a need for an integrated analysis, where rail is analyzed as a component of a larger and more complex transportation system. In this paper we present a new rail simulator as an integrated component of a comprehensive agent-based mobility simulation platform. By leveraging a multi-modal activity-based demand formulation and mesoscopic dynamic network supply simulation, our proposed design aims primarly to capture the impacts of complex dynamics between individual traveler behavior, the design and operation of a rail system, and the performance of all other modes. A second aim is flexibility; through the implementation of a flexible architecture, the simulation of a wide variety of different rail operational configurations and scenarios becomes feasible. As a demonstration of capabilities the simulator is used to replicate the operations and performance of a fully automated Mass Rapid Transit line in Singapore. The simulator was calibrated using a novel sequential calibration method with only automatic fare card data, without requiring automatic train control or manually collected data, and validated with another dataset. The proposed rail simulator can be used for the assessment of different rail control strategies under regular and disruption conditions, and to evaluate the performance of the road and other public transportation networks and users.