Calibrating Path Choices and Train Capacities for Urban Rail Transit Simulation Models Using Smart Card and Train Movement Data

Transit network simulation models are often used for performance and retrospective analysis of urban rail systems, taking advantage the availability extensive automated fare collection (AFC) vehicle location (AVL) data. Important inputs to such models, in addition origin-destination flows, include passenger path choices train capacity. Train capacity, which has been overlooked literature, is an important input that exhibits a lot variabilities. The paper proposes simulation-based optimization (SBO) framework simultaneously calibrate capacity systems using AFC AVL calibration formulated as problem with black-box objective function. Seven algorithms from four branches SBO solving methods evaluated. evaluated experimental design includes five scenarios, representing different degrees choice randomness crowding sensitivity. Data Hong Kong Mass Railway (MTR) system case study. data generate synthetic observations “ground truth.” results show response surface (particularly constrained surfaces) have consistently good under all scenarios. proposed approach drives large-scale applications monitoring planning.