Shortening average trip times by adjusting stopping and overtaking train stations

Abstract

We are studying railway operation optimization. Two kinds of trains, local and rapid, are under consideration to investigate the optimization of rapid trains stopping stations by adjusting the stopping stations to reach optimum convenience and rapidity. However, the rapid train has to overtake local trains that are running ahead. Therefore, overtaking facilities need to be considered too. In this paper, we describe the relation between stopping and overtaking station combinations and operation effectiveness. First, the rapid train stopping and overtaking stations are determined by DP (Dynamic Programming), considering real overtaking facilities at stations. Second, the rapid train timetable is determined and drawn on the train diagram. Third, the local train timetable is determined based on overtaking stations. The diagram is drawn for both directions starting from the overtaking station. The evaluation value is the average trip time, and is calculated by the timetable (diagram) and the passenger OD (Origin and Destination) table. By computer simulation using an existing line, we show that the most effective conditions are fewer overtaking stations, and no simultaneous overtaking at a stopping station.

Reducing passengers’ travel time by optimising stopping patterns in a large-scale network: A case-study in the Copenhagen Region

“…Skip-stop optimisation in transit networks has attracted an increasing amount of attention in recent years. Existing methods aimed at minimising either passengers' travel time (e.g., Suh et al, 2002;Mesa et al, 2009;Jong et al, 2012;Sogin et al, 2012;Feng et al, 2013;Jamili et al, 2014;Katori et al, 2014;Lee et al, 2014) or a combination of passengers' travel time and operating costs (e.g., Leiva et al, 2010;Freyss et al, 2013;Lin and Ku, 2014;Chen et al, 2015). Existing methods were formulated typically as mixed integer linear problems that were computationally hard to solve for real-world networks, were tested usually on a single railway corridor for the purpose of reducing the complexity of solution algorithms, and did not include adaptation of passenger behaviour as a consequence of the changes in the timetables related to the modifications in the stopping patterns.…”

“…Jamili et al (2014) applied a fuzzy approach to reach a flexible solution with the aim of minimising travel time and maximising the train spread while relaxing the headway constraints. Katori et al (2014) applied dynamic programming to find optimal stopping patterns and potential overtaking stops after forming local timetables from train diagrams.…”

“…Static demand was used in the existing methods either as predetermined, often based on passenger counts or ticket sales (e.g., Suh et al, 2002;Mesa et al, 2009;Leiva et al, 2010;Jong et al, 2012;Feng et al, 2013;Jamili et al, 2014;Katori et al, 2014;Lee et al, 2014;Lin and Ku, 2014), or approximated, generally because of complete or partial data inavailability (Sogin et al, 2012;Freyss et al, 2013;Chen et al, 2015). Specifically, Freyss et al (2013) did not have data about passenger movements and hence assumed that the destinations of all boarding passengers were evenly distributed on all remaining stops, while Sogin et al (2012) and Chen et al (2015) had data only about travel time between stops and hence ignored waiting and transfer times.…”