A Train Stop Deployment Planning Algorithm Using a Petri-net-based Modelling Approach

Abstract

One of the essential roles of railway operators is to maintain punctual transportation services and safety (as referred to in [1,2]). The dispatcher in charge of train rescheduling in the train traffic control room must stop related trains immediately to ensure safety if an accident occurs on a railway line. On the other hand, particularly if an accident causes blockage of the line for a long time, the dispatcher must decide which trains should be running and which should be stopped. Moreover, the locations of the trains to be stopped should be decided appropriately at that point. This decision-making process is referred to as train stop deployment planning.A basic requirement in train stop deployment planning is to accommodate the train at a location with a platform. If a train stops before entering a station, the passengers on board cannot disembark to take alternative means of transport (such as buses or trains on other railway lines) to reach their destinations. It is therefore desirable to hold the train at a location with a platform.Another requirement is that a stopped train should not block following trains that are scheduled to run onto another line. If stopped trains occupy all tracks at a station, the following trains cannot pass through. If a following train has a destination on another line, the delay propagates to that line. To avoid such a scenario, train dispatchers must maintain routes for trains bound for other lines.A number of constraints exist in train stop deployment planning. A train must not be stopped at a station where the length of the platform is insufficient to accommodate it, as its rear end will extend onto the track. Onboard hardware such as signalling system units limits which tracks a train can enter.To prepare the deployment of train stops, these requirements and constraints must be considered. Simply moving trains to the nearest station can cause the problem of local trains blocking following express services.Extensive research has been carried out in the area of train timetabling problems [3][4][5], and train rescheduling research has recently attracted attention (as referred to in [6][7][8][9][10][11][12]). On the other hand, train stop deployment planning problems are reported only in passing in [13]. In that paper, the authors examined a problem on a line with only two train classes and a simple track layout. It is necessary to introduce sophisticated procedures for the professional treatment of today's complicated railway lines.In this study, we represented a train stop deployment planning problem as a Petri net model (referred to in [14,15]) and formalized it as an integer programming problem. Based on this modelling and formalization, we introduced an algorithm to solve the problem. The results of numerical experiments indicated that the algorithm is capable of finding a practical solution to the problem within a reasonable computing time.

Impact of railway disruption predictions and rescheduling on passenger delays

“…In particular, Yang et al (2013) and Yang et al (2014) model the disruption length as a fuzzy variable that reflects the estimation by expert judgment. Hirai et al (2009) and Zhan et al (2015) focus on the first phase where trains need to stop before the disruption area. However both approaches disregard the uncertainty regarding the disruption length and the consequences for passenger delay.…”

Railway disruption management challenges and possible solution directions

“…This recovery is modelled macroscopically as a mixed integer program (MIP) to calculate the best reinsertion of cancelled services into the network to fit the periodic timetable. Hirai et al (2009) used Petri nets and integer programming (IP) to formalize and solve the train stop deployment problem. The model determines the stop locations for trains that can no longer operate according to the timetable and need to be cancelled.…”

“…Hirai et al (2009) provide a model that can be used to calculate the stop positions for the trains that are affected directly so that the other trains could continue their trips conflict-free. The approach by Zhan et al (2015) also defines the waiting location for trains until the source of the disruption is over.…”

<b>Algorithm Based on Gust and Flood Forecasts for Managing Train Operation Control and Passenger Evacuation</b>

“…RTRI has in the past carried out studies on the subject [3], however, they examined how to stop trains with a view to continue train operations on a main line if train operations were accidentally halted on a branch line.…”