No abstract

This paper presents an event-driven simulation-based optimization method for solving the train timetabling problem to minimize the total traveling time in the hybrid single and double track railway networks. The simulation approach is well applied for solving the train timetabling problems. In present simulation model, the stations and block sections of the railway network are respectively considered as the nodes and edges of the network model. The developed software named SIMARAIL has the capability of scheduling trains in large scale networks respecting the capacity constraints and infrastructure characteristics. This simulation model for railway timetabling is based on a detailed microscopic infrastructure model, which includes the most detailed infrastructure information. This research is based on integration of a discrete event simulation and GA meta-heuristic algorithm to generate near optimal train timetable. In other words, the simulation model is used to construct a feasible solution for train timetabling problems.

This paper presents a simulation-based optimization approach for railway timetabling, which is made interesting by the need for trains to stop periodically to allow passengers to pray. The developed framework is based on integration of a simulation model and an evolutionary path re-linking algorithm with the capability of scheduling trains, subject to the capacity constraints in order to minimize the total waiting times. A customized deadlock avoidance method has been developed which is based on a conditional capacity allocation. The proposed look-ahead deadlock avoidance approach is effective and easy to implement in the simulation model. A case study of the Iranian Railway (RAI) is selected for examining the efficiency of the meta-heuristic algorithm. The result shows that proposed algorithm has the capability of generating good quality solution in real-world problems.

The process of disruption management in rail transit systems faces challenging issues such as the unpredictable occurrence time, the consequences and the uncertain duration of disturbance or recovery time. The objective of this chapter is to adopt a discrete-event object-oriented simulation system, which applies the optimization algorithms in order to compensate the system performance after disruption. A line blockage disruption is investigated. The uncertainty associated with blockage recovery time is considered with several probabilistic scenarios. The disruption management model presented here combines short-turning and station-skipping control strategies with the objective to decrease the average passengers' waiting time. A variable neighborhood search (VNS) algorithm is proposed to minimize the average waiting time. The computational experiments on real instances derived from Tehran Metropolitan Railway are applied in the proposed model and the advantages of the implementing the optimized single and combined short-turning and stop-skipping strategies are listed.