HYBRID SIMULATION FOR RESOLVING RESOURCE CONFLICTS IN TRAIN TRAFFIC RESCHEDULING

Train traffic rescheduling is characterized by its dynamic, ever chang ing nature. Because of its complexity and dynamic behavior, simulation modelling is becoming a popular method to evaluate the effectiveness and appropriateness of train traffic rescheduling proposals and strate gies. Simulation can even be a powerful tool to support resolving reso urce conflicts in train traffic rescheduling. Normal event-driven simu lation and network-based simulation methods have been proposed for tra in traffic rescheduling. The former is a traditional event-based discr ete simulation method; the latter simulates through calculating the lo ngest path to every node in a network that is transformed from a norma l train traffic diagram. For more efficiently resolving resource confl icts in train traffic rescheduling, this paper proposes a hybrid metho d of the network-based simulation and the event-driven simulation. The proposed method attempts to obtain the advantages from both previous methods and reduce their shortcomings. Simple examples are used as cas e studies to illustrate the proposed method, and the effect of the alg orithm of the proposed method on overall performance is examined by ex periments. The experiment results show that the proposed method can be more easily and efficiently applied to resolve resource conflicts in train traffic rescheduling. (C) 1998 Published by Elsevier Science B.V . All rights reserved.