Railway traffic controllers face the problem that trains are often not operated as planned in timetables due to perturbations such as unexpected, degraded operations, and technical failures. This paper deals with the problem of train rerouting and rescheduling faced by the controllers of different regional railway control centers that must coordinate their decisions to minimize the impact of perturbations over the whole network. The authors split this real-time railway traffic management problem into two decision levels. At the lower level, the local area controllers, the dispatchers, manage train schedules and routes in their control areas from a microscopic perspective. At the higher level, a coordinator ensures the compatibility of dispatchers' decisions over two or more areas from a macroscopic perspective. To solve the problem, the authors propose an iterative optimization algorithm. The preliminary experimental results show that the proposed algorithm can reduce the delay propagation.