Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Railway operation analysis: Evaluation of quality, infrastructure and timetable on single and double-track lines with analytical models and simulation
KTH, School of Architecture and the Built Environment (ABE), Transport and Economics, Traffic and Logistics. (Tågtrafikgruppen)
2010 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

This thesis shows the advantages of simple models for analysis of railway operation. It presents two tools for infrastructure and timetable planning. It shows how the infrastructure can be analysed through fictive line designs, how the timetable can be treated as a variable and how delays can be used as performance measures. The thesis also gives examples of analyses of complex traffic situations through simulation experiments.

Infrastructure configuration, timetable design and delays play important roles in the competitiveness of railway transportation. This is especially true on single-track lines where the run times and other timetable related parameters are severely restricted by crossings (train meetings). The first half of this thesis focuses on the crossing time, i.e. the time loss that occurs in crossing situations. A simplified analytical model, SAMFOST, has been developed to calculate the crossing time as a function of infrastructure configuration, vehicle properties, timetable and delays for two crossing trains. Three measures of timetable flexibility are proposed and they can be used to evaluate how infrastructure configuration, vehicle properties, punctuality etc affect possibilities to alter the timetable.

Double-track lines operated with mixed traffic show properties similar to those of single-tracks. In this case overtakings imply scheduled delays as well as risk of delay propagation. Two different methods are applied for analysis of double-tracks: a combinatorial, mathematical model (TVEM) and simulation experiments.

TVEM, Timetable Variant Evaluation Model, is a generic model that systematically generates and evaluates timetable variants. This method is especially useful for mixed traffic operation where the impact of the timetable is considerable. TVEM may also be used for evaluation of different infrastructure designs. Analyses performed in TVEM show that the impact on capacity from the infrastructure increases with speed differences and frequency of service for the passenger trains, whereas the impact of the timetable is strongest when the speed differences are low and/or the frequency of passenger services is low.

Simulation experiments were performed to take delays and perturbations into account. A simulation model was set up in the micro simulation tool RailSys and calibrated against real operational data. The calibrated model was used for multi-factor analysis through experiments where infrastructure, timetable and perturbation factors were varied according to an experimental design and evaluated through response surface methods. The additional delay was used as response variable. Timetable factors, such as frequency of high-speed services and freight train speed, turned out to be of great importance for the additional delay, whereas some of the perturbation factors, i.e. entry delays, only showed a minor impact. The infrastructure factor, distance between overtaking stations, showed complex relationships with several interactions, principally with timetable factors.

Place, publisher, year, edition, pages
Stockholm: KTH , 2010. , 53 p.
Series
Trita-TEC-PHD, ISSN 1653-4468 ; 10-001
Keyword [en]
Railway operation, Railway capacity, Single-track, Double-track, Partial double-track, Inter-station distance, Delay, Delay propagation, Crossing, Overtaking, Timetable flexibility, Simulation, Experimental design
National Category
Civil Engineering
Identifiers
URN: urn:nbn:se:kth:diva-12727ISBN: 978-91-85539-50-5 (print)OAI: oai:DiVA.org:kth-12727DiVA: diva2:318296
Public defence
2010-05-28, E1, Lindstedtsvägen 3, KTH, Stockholm, 10:00 (English)
Opponent
Supervisors
Projects
Framtida infrastruktur och kvalitet i tågföring
Note
QC20100622Available from: 2010-05-12 Created: 2010-05-07 Last updated: 2010-06-22Bibliographically approved
List of papers
1. Influences of station length and inter-station distance on delays and delay propagation on single-track lines with regional rail traffic
Open this publication in new window or tab >>Influences of station length and inter-station distance on delays and delay propagation on single-track lines with regional rail traffic
2006 (English)In: Computers in Railways X: Computer System Design and Operation in the Railway and Other Transit Systems / [ed] Allan, J; Brebbia, CA; Rumsey, AF; Sciutto, G; Sone, S; Goodman, CJ, Southampton: WIT Press , 2006, 511-520 p.Conference paper, Published paper (Refereed)
Abstract [en]

Train services on single-track lines suffer from time losses due to crossings, imposed by the bidirectional traffic. The time losses, in this paper denoted crossing time, are caused by constraints in the infrastructure and delay propagation, which give a stochastic contribution that varies from one crossing situation to another.

Two examples of infrastructure improvements that decrease the crossing time are examined: partial double-track at the location of timetabled crossing and decreased inter-station distances. A mathematical model is used to evaluate these improvements.

Partial double-tracks seem to be very efficient when traffic intensity and delay variances are moderate. Shortened inter-station distances give less effect but are less sensitive to delay variance and give valuable additional line capacity.

The used model assumes independence between crossing trains, which imposes a moderate capacity utilisation. In more congested situations simulation methods are needed to make more complex crossing patterns possible.

Place, publisher, year, edition, pages
Southampton: WIT Press, 2006
Series
WIT TRANSACTIONS ON THE BUILT ENVIRONMENT, ISSN 1746-4498 ; 88
Keyword
single-track, delay propagation, partial double-track, inter-station distance
National Category
Reliability and Maintenance Civil Engineering
Research subject
Järnvägsgruppen - Kapacitet
Identifiers
urn:nbn:se:kth:diva-12726 (URN)10.2495/CR060511 (DOI)000240944200051 ()2-s2.0-36148979378 (Scopus ID)1-84564-177-9 (ISBN)
Conference
10th International Conference on Computer System Design and Operation in the Railway and Other Transit Systems. Prague, CZECH REPUBLIC. JUL 10-12, 2006
Projects
Framtida infrastruktur och kvalitet i tågföring
Note
QC 20100622 QC 20111006Available from: 2010-05-07 Created: 2010-05-07 Last updated: 2012-02-28Bibliographically approved
2. SAMFOST: a timetable-free way of analysing single-track railway lines
Open this publication in new window or tab >>SAMFOST: a timetable-free way of analysing single-track railway lines
2007 (English)In: Proceedings of 2nd International Seminar on Railway Operations Modelling and Analysis / [ed] Hansen, I.A., Radtke, A., Pachl, J.P., Wendler E., 2007Conference paper, Published paper (Refereed)
Abstract [en]

Operation of single-track railway lines is characterised by the stochastic crossing time. Important properties such as capacity, run times and overall punctuality are strongly dependent on the crossing time. In non-congested situations the crossing time can be modelled analytically. SAMFOST is a model that shows in a transparent way how the crossing time varies along a railway line. Including arrival delays, the mean crossing time (and standard deviation) can be calculated for different timetable crossing points. An important contribution is that the dependence between arrival delays and crossing time is shown explicitly.

SAMFOST has been successfully validated against the accepted simulation tool RailSys. The validation shows that SAMFOST has an acceptable accuracy as long as the crossing time depends only on the two crossing trains (non-congested situations).

Using SAMFOST, different types of measures to decrease the crossing time (and its variance) have been analysed. For example, a combined crossing and passenger stop gives a considerable decrease in the mean crossing time, but at the same time the variance increases. A combined crossing and passenger stop thus becomes more dependent on punctuality than a crossing at a station without passenger stop

Keyword
single-track, railway operation, delay
National Category
Civil Engineering
Research subject
Järnvägsgruppen - Kapacitet
Identifiers
urn:nbn:se:kth:diva-12720 (URN)
Conference
2nd International Seminar on Railway Operations Modelling and Analysis
Projects
Framtida infrastruktur och kvalitet i tågföring
Note
QC 20100622Available from: 2010-05-07 Created: 2010-05-07 Last updated: 2012-02-28Bibliographically approved
3. Crossing times on single-track railway lines: dependencies of different infrastructure and traffic factors
Open this publication in new window or tab >>Crossing times on single-track railway lines: dependencies of different infrastructure and traffic factors
2007 (English)In: Proceedings of Railway Engineering 9th International Conference and Exhibition / [ed] Ford, M.C., 2007Conference paper, Published paper (Refereed)
Abstract [en]

Operation of single-track railway lines is characterised by crossings. Due to disturbances the crossing time is stochastic. Important properties such as capacity, run times and overall punctuality are strongly dependent on the crossing time. In non-congested situations the crossing time can be modelled analytically. SAMFOST (Simplified Analytical Model For Single Tracks) is a model that shows in a transparent way how the crossing time varies along a railway line.

This paper shows how the mean crossing time function can be used to examine timetable flexibility, i.e. possibilities to change the timetable given constraints on the mean crossing time. Three different measures for timetable flexibility are proposed, showing how the available crossing points are spread along the line and the spread of mean crossing time for available crossing points.

Many factors such as infrastructure parameters, vehicle parameters, timetable, delays etc affect the crossing time. In many cases these factors interact in complicated ways. To show this, the results of a 26-factorial experiment are presented. Partial double-tracks and passenger stops at timetable crossing points are examples of measures that give geographically local effects with strong interactions, whereas inter-station distance, vehicle type and arrival punctuality are factors with a more general impact and weaker interactions.

Keyword
single-track, crossing, rialway operation, delay, timetable flexibility
Research subject
Järnvägsgruppen - Kapacitet
Identifiers
urn:nbn:se:kth:diva-12722 (URN)
Conference
Railway Engineering 9th International Conference and Exhibition
Projects
Framtida infrastruktur och kvalitet i tågföring
Note
QC20100622Available from: 2010-05-07 Created: 2010-05-07 Last updated: 2012-02-28Bibliographically approved
4. Analysis of capacity on double-track railway lines
Open this publication in new window or tab >>Analysis of capacity on double-track railway lines
2010 (English)In: Transportation planning and technology (Print), ISSN 0308-1060, E-ISSN 1029-0354Article in journal (Refereed) Accepted
Abstract [en]

In Sweden rail traffic is almost never separated according to speed. On several double-track lines the mix of heavy freight, regional and high speed trains imposes severe capacity problems. In order to evaluate the capacity for different traffic mixes a combinatorial model, TVEM (Timetable Variant Evaluation Model), has been developed. In this model both the infrastructure and the timetable are modelled as variables. The traffic is divided into train patterns according to a presumed regular timetable and then systematically scheduled in different time locations. The timetable variants are evaluated with regard to: mean values of capacity that give the number of trains/h for the required mix, variance measures that show how the capacity depends on the timetable and scheduled delays that show the extension of run times imposed by overtakings.

The paper shows how the important distance between adjacent overtaking stations can be sampled from Weibull distributions. TVEM has been used to evaluate three different operational cases with mixed traffic. The analysis shows that the impact on capacity from the infrastructure increases with speed difference and frequency of operation for the passenger trains, while the importance of the infrastructure decreases when traffic is more homogeneous. The impact from the timetable is strongest when the speed differences are low and/or the frequency of passenger trains is low. Capacity loss due to increased speed differences can be compensated for with additional overtaking stations. The slower trains suffer from a considerable increase in scheduled delays when speed differences increase.

Keyword
Double-track, railway operation, railway capacity
Research subject
Järnvägsgruppen - Kapacitet
Identifiers
urn:nbn:se:kth:diva-12723 (URN)
Projects
Framtida infrastruktur och kvalitet i tågföring
Note
QC20100622Available from: 2010-05-07 Created: 2010-05-07 Last updated: 2017-12-12Bibliographically approved
5. Validation of a simulation model for mixed traffic on a Swedish double-track railway line
Open this publication in new window or tab >>Validation of a simulation model for mixed traffic on a Swedish double-track railway line
2009 (English)In: Proceedings of Railway Engineering 10th International Conference and Exhibition / [ed] Ford, M.C., 2009Conference paper, Published paper (Refereed)
Abstract [en]

Simulation is an appropriate method of modelling complex railway operation systems. In a simulation tool it is possible to model interactions between trains and the delay propagation that follows on conflicts. This study describes a process for calibrating/validating a simulation model set up in the commercial simulation tool RailSys.

Response Surface Methodology is applied for a simultaneously calibration of seven factors. Latin Hypercubes are used as the experimental design since they support complex response surface metamodels.

The simulation model’s ability to react accurately to different operational conditions is tested. Two different traffic mixes are modelled and the validation is performed on a significantly different delay level (magnitude of disturbance).

The results show that it is possible to calibrate the model merely by adjusting the chosen factors and that these factors can compensate for simplifications made when the model was being set up. Simulated responses (mean and standard deviation measures) were compared to the real outcome and the relative differences were 2-19%. The validation generally showed greater differences.

Keyword
Validation, mixed railway traffic, simulation, double-track, railway operation
Research subject
Järnvägsgruppen - Kapacitet
Identifiers
urn:nbn:se:kth:diva-12724 (URN)
Conference
Railway Engineering 10th International Conference and Exhibition, London
Projects
Framtida infrastruktur och kvalitet i tågföring
Note
QC20100622Available from: 2010-05-07 Created: 2010-05-07 Last updated: 2012-02-28Bibliographically approved
6. Impacts of infrastructure, timetable and perturbations in operation of double-track railway lines with mixed traffic
Open this publication in new window or tab >>Impacts of infrastructure, timetable and perturbations in operation of double-track railway lines with mixed traffic
2010 (English)In: 12th World Conference on Transportation Research 2010, 2010Conference paper, Published paper (Refereed)
Abstract [en]

Delays play a central role in railway operation. They are of great importance both for customers and operators. They are direct measures of quality and reliability and hereby also an important factor for the competitiveness of the entire railway. Indirectly, the delays also affect quantitative factors such as capacity, i.e. the number of trains that can be (practically) operated. For these reasons, analysis of delays and delay propagation is an essential part of railway operations research.

The up-coming deregulation of railway traffic means that completely mixed traffic can be foreseen on the Swedish railway network. This article shows how the delays on a double-track railway line, operated with mixed traffic, are affected by infrastructure, timetable and primary delays. Experimental design, simulation and response surface metamodelling are applied in a multi-factor simulation experiment with nine factors.

The combination of simulation and experimental design makes it possible to draw general conclusions from a limited number of simulated variants and this type of multi-factor analysis is essential to an understanding of the railway as an operational system. The derived metamodels may also be used in different types of planning processes.

The metamodels show that speed and frequency factors have a great impact on delays. Freight train speed and the frequency of service of high-speed trains in particular turned out to be important. Perturbation factors, i.e. entry delays, were found to affect the delays less. Neither does the distance between adjacent overtaking stations in itself affect the delays. However, the inter-station distance still affects delays through interactions with other factors.

Keyword
Double-track, timetable, railway operation, railway capacity, experimental design, simulation, delays, reliability
National Category
Civil Engineering
Research subject
Järnvägsgruppen - Kapacitet
Identifiers
urn:nbn:se:kth:diva-12725 (URN)
Conference
12th World Conference on Transportation Research, Lisbon, Portugal
Projects
Framtida infrastruktur och kvalitet i tågföring
Note
QC20100622Available from: 2010-05-07 Created: 2010-05-07 Last updated: 2012-02-28Bibliographically approved

Open Access in DiVA

fulltext(490 kB)3892 downloads
File information
File name FULLTEXT01.pdfFile size 490 kBChecksum SHA-512
fff4f4444ee8abb5a35a388abe908c9e603ad8ee798fc868322411d451f7294bbd6a2413ca5328ad7cc605efea6ae6ba6f16ec3c6dc87343e9b52a0887860900
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Lindfeldt, Olov
By organisation
Traffic and Logistics
Civil Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 3892 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 999 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf