Rolling Stock Scheduling by Mixed-Integer Linear Programming Considering Splitting and Combining

Abstract

Railway operating companies must prepare rolling stock schedules to realize train operations during a timetable revision. Companies create a roster, which includes a series of trains to be considered by the particular rolling stock with inspection interval constraints. In some Japanese railway lines, splitting and combining operations are often used to adjust the transportation capacity flexibly. However, shunting operations of splitting and combining are necessary under such operational conditions. These operations are labor-intensive and time-consuming, necessitating a reduction in the amount of splitting and combining. In this paper, we present a rolling stock scheduling based on a mixed integer linear programming model to reduce the amount of splitting and combining. Computational experimental results are presented based on real instances of several railway lines in Japan, indicating the computational effectiveness of the methodology.