Timetable Analysis based on Social Cost and Simulation
Time: Thu 2022-06-09 10.15
Location: Kollegiesalen, Brinellvägen 8, Stockholm
Video link: https://kth-se.zoom.us/j/65106856604
Language: English
Subject area: Transport Science, Transport Systems
Doctoral student: Jennifer Warg , Transportplanering
Opponent: Professor Paola Pellegrini, Université Gustave Eiffel
Supervisor: Docent Markus Bohlin, Transportplanering; Docent Oskar Fröidh, Järnvägsgruppen, JVG, Transportplanering; PhD Carl-William Palmqvist, Transportplanering, Lunds tekniska högskola
QC220511
Abstract
The timetable is essential for railway traffic. Its design affects rail operations – and by extension all operators, passengers, and freight clients. International practice for capacity allocation varies and there is no common algorithm for designing and evaluating timetable changes. Especially challenging is coordinating heterogeneous services competing for track capacity.
Aim of this thesis is to improve efficiency and transparency in capacity allocation by new methods to assess timetable adjustments.
Although integration of capacity analysis and socio-economic evaluation is beneficial, combination of the two methods is rare. Because of that, two types of model are proposed in this thesis: The first combines capacity variables and social costs to represent the goodness of a timetable slot or a timetable using performance indices describing the travelling time experience of passengers riding the trains. Historical delay data from existing timetables provides information about the performance under operation. For strategies, a combined timetable and simulation approach is proposed to reveal variables for input to performance evaluation models. We show that microscopic simulation is advantageous for estimating operational data. Because departures ahead of schedule are common for freight trains suffers a simulation that does not include that feature of inaccuracies. Because of that, we suggest an approach for including early freight train departures to improve the accuracy of simulation results. Simulation with a macroscopic tool can speed up evaluation and improve user-friendliness of the method. For relatively small changes such as changed departure times and supplements which is focus in this work however, the microscopic tool shows greater advantage.
The second model type considers conflicts between commercial long-distance trains and publicly controlled short-distance traffic competing for track capacity. We propose a method to calculate the loss of societal benefits incurred by changing the commuter train timetable to accommodate a commercial train path request. The calculation of these benefits considers changes in commuter train passengers’ travel times, waiting times, transfers and crowding for all trips in the system in combination with changes in operating costs for operator(s) of the short-distance services.
All methods are applied in case studies and can (with adaptions) be used for other lines to improve the timetable allocation process, and to support transparent conflict solution. However, choosing an adequate model configuration is essential as it can affect results significantly.
Research was conducted for the two Swedish mainlines for passenger traffic (one articles includes an extension for a freight-focussed line), plus the Stockholm commuter network. The majority of the thesis concentrates on passenger traffic, and in particular, the conflict between long- and short-distance services.
In addition to methodological contributions, we provide several practical results. We present strategies for improving reliability and evaluate them. Simple small changes in the timetable (such as changed departure times or time supplements) can have significant impact on the goodness of one or several timetable slots.