DEVELOPMENT OF TRAIN SCHEDULING MODEL ACCOMMODATING TRANSFER BEHAVIOR

Abstract

The purpose of this paper is the development of the operational strategies to maximize the rail network efficiency. The existing schedule optimization model, where the planning objective is operation cost minimization or travel time minimization, is extended to the new model by considering transfer behavior. The transfer penalty is introduced to constraints in the mixed integer nonlinear programming (MINLP) model for the analysis of the alternative with transfer between two types of trains. Because this type of the problem is NP-complete, so branch-and-cut approach, which is widely used for this type problem, is introduced to the solution procedure.
The practical availability of the rail operational strategy in this paper has evaluated and the numerical results show that the schedule model constructed in this paper can product the optimal schedule to utilize the completed rail network efficiently by introducing the transfer penalty into the framework. Afterwards, this study can make the basis of the improved operational strategy consisting of the rail transfer scheduling by including the studies about the more detailed travelers' behaviors.