The handling of emergencies like severe wind and foreign matter collision in high-speed railways is a typical human-in-the-loop complex task. Conventional approaches find it challenging to satisfy the real-time, efficiency, and safety standards for reorganizing train schedules during such critical situations. This paper introduces a parallel railway traffic management (RTM) system to address the rescheduling of high-speed trains during such emergencies. The design and implementation of the parallel system are detailed using the ACP (Artificial systems, Computational experiments, and Parallel execution) methodology. An agent-based modeling approach is employed to construct the artificial RTM system. The computational experiments are designed and executed to forecast the system's status and assess the rescheduling strategies and algorithms utilizing computer-based simulation technology within the artificial RTM system. Through enabling real-time interaction and closed-loop feedback between the physical RTM system and the actual RTM system, the rescheduling strategies can be continuously assessed and iteratively optimized in real-time. Finally, a case study is conducted in two typical scenarios, involving temporary speed limit and complete blockage, to evaluate the performance of the proposed method through computational experiments. Timetable rescheduling during emergencies is formulated as a multi-objective optimization problem with various constraints. The hybrid strategy, First-Come-First-Served (FCFS) strategy, and First-Scheduled-First-Served (FSFS) strategy are applied to address the train rescheduling problem. The results demonstrate that the proposed method surpasses traditional approaches in managing train rescheduling during disruptions. It enhances the efficiency of emergency response and furnishes decision support for dispatchers.