schlechte thomas

Upload: mhenley8059

Post on 03-Jun-2018

219 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/12/2019 Schlechte Thomas

    1/244

    Railway Track Allocation: Models andAlgorithms

    vorgelegt vonDipl.-Math. oec. Thomas Schlechte

    aus Halle an der Saale

    Von der Fakultat II Mathematik und Naturwissenschaftender Technischen Universitat Berlin

    zur Erlangung des akademischen Grades

    Doktor der Naturwissenschaften Dr. rer. nat.

    genehmigte Dissertation

    PromotionsausschussBerichter: Prof. Dr. Dr. h. c.mult.MartinGr otschel

    PD Dr. habil. Ralf BorndorferVorsitzender: Prof. Dr. Fredi Tr oltzsch

    Tag der wissenschaftlichen Aussprache: 23.12.2011

    Berlin 2012D 83

  • 8/12/2019 Schlechte Thomas

    2/244

  • 8/12/2019 Schlechte Thomas

    3/244

    Railway Track Allocation: Modelsand Algorithms

    Thomas Schlechte

  • 8/12/2019 Schlechte Thomas

    4/244

  • 8/12/2019 Schlechte Thomas

    5/244

    PrefaceThe heart of a railway system is the timetable. Each railway opera-tor has to decide on the timetable to offer and on the rolling stock tooperate the trips of the trains. For the railway infrastructure managerthe picture is slightly different trains have to be allocated to rail-way tracks and times, called slots such that all passenger and freighttransport operators are satised and all train movements can be car-ried out safely. This problem is called the track allocation problem . Mythesis deals with integer programming models and algorithmic solutionmethods for the track allocation problem in real world railway systems.

    My work on this topic has been initiated and motivated by the in-terdisciplinary research project railway slot allocation or in GermanTrassenborse.1 This project investigated the question whether a com-petitive marketing of a railway infrastructure can be achieved using anauction-based allocation of railway slots. The idea is that competingtrain operating companies (TOCs) can bid for any imaginable use of the infrastructure. Possible conicts will be resolved in favor of theparty with the higher willingness to pay, which leads directly to thequestion of nding revenue maximal track allocations. Moreover afair and transparent mechanism cries out for exact optimization ap-proaches, because otherwise the resulting allocation is hardly accept-able and applicable in practice. This leads to challenging questionsin economics, railway engineering, and mathematical optimization. Inparticular, developing models that build a bridge between the abstractworld of mathematics and the technical world of railway operationswas an exciting task.

    I worked on the Trassenborse project with partners from different ar-eas, namely, on economic problems with the Workgroup for Economicand Infrastructure Policy (WIP) at the Technical University of Berlin(TU Berlin), on railway aspects with the Chair of Track and Rail-way Operations (SFWBB) at TU Berlin, the Institute of Transport,Railway Construction and Operation (IVE) at the Leibniz Universit atHannover, and the Management Consultants Ilgmann Miethner Part-ner (IMP).

    1 This project was funded by the Federal Ministry of Education and Research(BMBF), Grant number 19M2019 and the Federal Ministry of Economics and Tech-nology (BMWi), Grant number 19M4031A and Grant number 19M7015B.

  • 8/12/2019 Schlechte Thomas

    6/244

    This thesis is written from the common perspective of all persons Iworked closely with, especially the project heads Ralf Borndorfer andMartin Gr otschel, project partners Gottfried Ilgmann and KlemensPolatschek, and the ZIB colleagues Berkan Erol, Elmar Swarat, andSteffen Weider.

    The highlight of the project was a cooperation with the SchweizerischeBundesbahnen (SBB) on optimizing the cargo traffic through the Sim-plon tunnel, one of the major transit routes in the Alps. This real worldapplication was challenging in many ways. It provides the opportunityto verify the usefulness of our methods and algorithms by computing

    high quality solutions in a fully automatic way.The material covered in this thesis has been presented at several in-ternational conferences, e.g., European Conference on Operational Re-search (EURO 2009, 2010), Conference on Transportation Schedulingand Disruption Handling, Workshop on Algorithmic Approaches forTransportation Modeling, Optimization, and System (ATMOS 2007,2010), International Seminar on Railway Operations Modeling andAnalysis (ISROR 2007, 2009, 2011), Symposium on Operations Re-search (OR 2005, 2006, 2007, 2008), International Conference on Com-puter System Design and Operation in the Railway and other Transit

    Systems (COMPRAIL), International Conference on Multiple CriteriaDecision Making (MCDM), World Conference on Transport Research(WCTR). Signicant parts have already been published in various ref-ereed conference proceedings and journals:

    Borndorfer et al. (2006) [34], Borndorfer et al. (2005) [33], Borndorfer & Schlechte (2007) [31], Borndorfer & Schlechte (2007) [30], Erol et al. (2008) [85], Schlechte & Borndorfer (2008) [188], Borndorfer, Mura & Schlechte (2009) [40], Borndorfer, Erol & Schlechte (2009) [38], Schlechte & Tanner (2010) [189]3, Borndorfer, Schlechte & Weider (2010) [43], Schlechte et al. (2011) [190],1 and Bornd orfer et al. (2010) [42]2.

    1 accepted by Journal of Rail Transport Planning & Management.2 accepted by Annals of Operations Research.3 submitted to Research in Transportation Economics.

  • 8/12/2019 Schlechte Thomas

    7/244

    Research Goals and ContributionsThe goal of the thesis is to solve real world track allocation problemsby exact integer programming methods. In order to establish a fair andtransparent railway slot allocation, exact optimization approaches arerequired, as well as accurate and reliable railway models. Integer pro-gramming based methods can provide excellent guarantees in practice.We successfully identied and tackled several tasks to achieve theseambitious goals:

    1. applying a novel modeling approach to the track allocation prob-lem called conguration models and providing a mathematicalanalysis of the associated polyhedron,

    2. developing a sophisticated integer programming approach calledrapid branching that highly utilizes the column generation tech-nique and the bundle method to tackle large scale track allocationinstances,

    3. developing a Micro-Macro Transformation, i.e., a bottom-up ag-gregation, approach to railway models of different scale to pro-duce a reliable macroscopic problem formulation of the track al-location problem,

    4. providing a study comparing the proposed methodology to formerapproaches, and,

    5. carrying out a comprehensive real world data study for the Sim-plon corridor in Switzerland of the entire optimal railway trackallocation framework.

    In addition, we present extensions to incorporate aspects of robustnessand we provide an integration and empirical analysis of railway slot

    allocation in an auction based framework.

    Thesis Structure

    A rough outline of the thesis is shown in Figure 1. It follows thesolution cycle of applied mathematics. In a rst step the real worldproblem is analyzed, then the track allocation problem is translatedinto a suitable mathematical model, then a method to solve the models

  • 8/12/2019 Schlechte Thomas

    8/244

    in an efficient way is developed, followed by applying the developedmethodology in practice to evaluate its performance. Finally, the loopis closed by re-translating the results back to the real world applicationand analyze them together with experts and practitioners.

    Main concepts on planning problems in railway transportation are pre-sented in Chapter I. Railway modeling and infrastructure capacity isthe main topic of Chapter II. Chapter III focuses on the mathematicalmodeling and the solution of the track allocation problem. Finally,Chapter IV presents results for real world data as well as for ambitioushypothetical auctioning instances.

  • 8/12/2019 Schlechte Thomas

    9/244

  • 8/12/2019 Schlechte Thomas

    10/244

  • 8/12/2019 Schlechte Thomas

    11/244

    AbstractThis thesis is about mathematical optimization for the efficient useof railway infrastructure. We address the optimal allocation of theavailable railway track capacity the track allocation problem . Thistrack allocation problem is a major challenge for a railway company,independent of whether a free market, a private monopoly, or a pub-lic monopoly is given. Planning and operating railway transportationsystems is extremely hard due to the combinatorial complexity of theunderlying discrete optimization problems, the technical intricacies,and the immense sizes of the problem instances. Mathematical modelsand optimization techniques can result in huge gains for both railwaycustomers and operators, e.g., in terms of cost reductions or servicequality improvements. We tackle this challenge by developing novelmathematical models and associated innovative algorithmic solutionmethods for large scale instances. This allows us to produce for therst time reliable solutions for a real world instance, i.e., the Simplon corridor in Switzerland.

    The opening chapter gives a comprehensive overview on railway plan-ning problems. This provides insights into the regulatory and technicalframework, it discusses the interaction of several planning steps, andidenties optimization potentials in railway transportation. The re-mainder of the thesis is comprised of two major parts.

    The rst part (Chapter II) is concerned with modeling railway sys-tems to allow for resource and capacity analysis. Railway capacity hasbasically two dimensions, a space dimension which are the physical in-frastructure elements as well as a time dimension that refers to thetrain movements, i.e., occupation or blocking times, on the physicalinfrastructure. Railway safety systems operate on the same principleall over the world. A train has to reserve infrastructure blocks forsome time to pass through. Two trains reserving the same block of the infrastructure within the same point in time is called block conict .Therefore, models for railway capacity involve the denition and cal-culation of reasonable running and associated reservation and blockingtimes to allow for a conict free allocation.

    There are microscopic models that describe the railway system ex-tremely detailed and thorough. Microscopic models have the advantage

  • 8/12/2019 Schlechte Thomas

    12/244

    that the calculation of the running times and the energy consumptionof the trains is very accurate. A major strength of microscopic modelsis that almost all technical details and local peculiarities are adjustableand are taken into account. We describe the railway system on a mi-croscopic scale that covers the behavior of trains and the safety systemcompletely and correctly. Those models of the railway infrastructureare already very large even for very small parts of the network. Thereason is that all signals, incline changes, and switches around a railwaystation have to be modeled to allow for precise running time calcula-tions of trains. In general microscopic models are used in simulationtools which are nowadays present at almost all railway companies allover the world. The most important eld of application is to validatea single timetable and to decide whether a timetable is operable andrealizable in practice. However, microscopic models are inappropriatefor mathematical optimization because of the size and the high levelof detail. Hence, most optimization approaches consider simplied, socalled macroscopic , models. The challenging part is to construct a re-liable macroscopic model for the associated microscopic model and tofacilitate the transition between both models of different scale.

    In order to allocate railway capacity signicant parts of the microscopicmodel can be transformed into aggregated resource consumption inspace and time. We develop a general macroscopic representation of railway systems which is based on minimal headway times for enteringtracks of train routes and which is able to cope with all relevant railwaysafety systems. We introduce a novel bottom-up approach to generatea macroscopic model by an automatic aggregation of simulation dataproduced by any microscopic model. The transformation aggregatesand shrinks the infrastructure network to a smaller representation, i.e.,it conserves all resource and capacity aspects of the results of the mi-croscopic simulation by conservative rounding of all times. The mainadvantage of our approach is that we can guarantee that our macro-

    scopic results, i.e., train routes, are feasible after re-transformation forthe original microscopic model. Because of the conservative roundingmacroscopic models tend to underestimate the capacity. We can con-trol the accuracy of our macroscopic model by changing the used timediscretization. Finally, we provide a priori error estimations of ourtransformation algorithm, i.e., in terms of exceeding of running andheadway times.

    In the second and main part (Chapter III) of the thesis, the optimaltrack allocation problem for macroscopic models of the railway sys-

  • 8/12/2019 Schlechte Thomas

    13/244

    tem is considered. The literature for related problems is surveyed. Agraph-theoretic model for the track allocation problem is developed. Inthat model optimal track allocations correspond to conict-free pathsin special time-expanded graphs. Furthermore, we made considerableprogress on solving track allocation problems by two main features anovel modeling approach for the macroscopic track allocation problemand algorithmic improvements based on the utilization of the bundlemethod.

    More specically, we study four types of integer programming modelformulations for the track allocation problem: two standard formula-

    tions that model resource or block conicts in terms of packing con-straints, and two novel coupling or conguration formulations. Inboth cases variants with either arc variables or with path variables willbe presented. The key idea of the new formulation is to use additionalconguration variables that are appropriately coupled with the stan-dard train ow variables to ensure feasibility. We show that thesemodels are a so called extended formulations of the standard packingmodels.

    The success of an integer programming approach usually depends onthe strength of the linear programming (LP) relaxation. Hence, weanalyze the LP relaxations of our model formulations. We show, thatin case of block conicts, the packing constraints in the standard for-mulation stem from cliques of an interval graph and can therefore beseparated in polynomial time. It follows that the LP relaxation of a strong version of this model, including all clique inequalities fromblock conicts, can be solved in polynomial time. We prove that theLP relaxation of the extended formulation for which the number of variables can be exponential, can also be solved in polynomial time,and that it produces the same LP bound. Furthermore, we prove thatcertain constraints of the extended model are facets of the polytope

    associated with the integer programing formulation. To incorporaterobustness aspects and further combinatorial requirements we presentsuitable extensions of our coupling models.

    The path variant of the coupling model provides a strong LP bound,is amenable to standard column generation techniques, and thereforesuited for large-scale computation. Furthermore, we present a sophis-ticated solution approach that is able to compute high-quality integersolutions for large-scale railway track allocation problems in practice.Our algorithm is a further development of the rapid branching method

  • 8/12/2019 Schlechte Thomas

    14/244

    introduced in Bornd orfer, Lobel & Weider (2008) [37] (see also the the-sis Weider (2007) [213]) for integrated vehicle and duty scheduling inpublic transport. The method solves a Lagrangean relaxation of thetrack allocation problem as a basis for a branch-and-generate procedurethat is guided by approximate LP solutions computed by the bundlemethod. This successful second application in public transportationprovides evidence that the rapid branching heuristic guided by thebundle method is a general heuristic method for large-scale path pack-ing and covering problems. All models and algorithms are implementedin a software module TS-OPT .

    Finally, we go back to practice and present in the last chapter severalcase studies using the tools netcast and TS-OPT . We provide a compu-tational comparison of our new models and standard packing modelsused in the literature. Our computational experience indicates thatour approach, i.e., conguration models, outperforms other models.Moreover, the rapid branching heuristic and the bundle method en-able us to produce high quality solutions for very large scale instances,which has not been possible before. In addition, we present results for atheoretical and rather visionary auction framework for track allocation.We discuss several auction design questions and analyze experimentsof various auction simulations.

    The highlights are results for the Simplon corridor in Switzerland. Weoptimized the train traffic through this tunnel using our models andsoftware tools. To the best knowledge of the author and conrmedby several railway practitioners this was the rst time that fully auto-matically produced track allocations on a macroscopic scale fulll therequirements of the originating microscopic model, withstand the eval-uation in the microscopic simulation tool OpenTrack , and exploit theinfrastructure capacity. This documents the success of our approachin practice and the usefulness and applicability of mathematical opti-mization to railway track allocation.

  • 8/12/2019 Schlechte Thomas

    15/244

    Deutsche ZusammenfassungDiese Arbeit befasst sich mit der mathematischen Optimierung zur ef-zienten Nutzung der Eisenbahninfrastruktur. Wir behandeln die op-timale Allokation der zur Verf ugung stehenden Kapazit at eines Eisen-bahnschienennetzes das Trassenallokationsproblem . Das Trassenallo-kationsproblem stellt eine wesentliche Herausforderung f ur jedes Bahn-unternehmen dar, unabh angig, ob ein freier Markt, ein privates Mo-nopol, oder ein offentliches Monopol vorherrscht. Die Planung undder Betrieb eines Schienenverkehrssystems ist extrem schwierig auf-grund der kombinatorischen Komplexit at der zugrundeliegenden dis-kreten Optimierungsprobleme, der technischen Besonderheiten, undder immensen Gr oen der Probleminstanzen. Mathematische Model-le und Optimierungstechniken k onnen zu enormen Nutzen f uhren, so-wohl f ur die Kunden der Bahn als auch f ur die Betreiber, z.B. in Bezugauf Kosteneinsparungen und Verbesserungen der Servicequalit at. Wirlosen diese Herausforderung durch die Entwicklung neuartiger mathe-matischer Modelle und der dazugh origen innovativen algorithmischenLosungsmethoden f ur sehr groe Instanzen. Dadurch waren wir erst-

    mals in der Lage zuverlassige Losungen f ur Instanzen der realen Welt,d.h. f ur den Simplon Korridor in der Schweiz, zu produzieren.

    Das einf uhrende Kapitel gibt einen umfangreichen Uberblick zum Pla-nungsproze im Eisenbahnwesen. Es liefert Einblicke in den ordnungs-politischen und technischen Rahmen, diskutiert die Beziehung zwischenden verschiedenen Planungsschritten und identiziert Optimierungspo-tentiale in Eisenbahnverkehrssystemen. Der restliche Teil der Arbeitgliedert sich in zwei Hauptteile.

    Der erste Teil (Kapitel II) beschaftigt sich mit der Modellierung des

    Schienenbahnsystems unter Ber ucksichtigung von Kapazit at und Res-sourcen. Kapazit at im Schienenverkehr hat grunds atzlich zwei Dimen-sionen, eine raumliche, welche der physischen Infrastruktur entspricht,und eine zeitliche, die sich auf die Zugbewegungen innerhalb dieser be-zieht, d.h. die Belegung- und Blockierungszeiten. Sicherungssysteme imSchienenverkehr beruhen uberall auf der Welt auf demselben Prinzip.Ein Zug muss Blocke der Infrastruktur f ur die Durchfahrt reservieren.Das gleichzeitige Belegen eines Blockes durch zwei Zuge wird Block-konikt genannt. Um eine koniktfreie Belegung zu erreichen, bein-halten Modelle zur Kapazit at im Schienenverkehr daher die Denition

  • 8/12/2019 Schlechte Thomas

    16/244

    und Berechnung von angemessenen Fahrzeiten und dementsprechendenReservierungs- oder Blockierungszeiten.

    Es gibt mikroskopische Modelle, die das Bahnsystem sehr ausf uhrlichund genau beschreiben. Mikroskopische Modelle haben den Vorteil,dass die Berechnung der Fahrzeiten und des Energieverbrauchs derZuge sehr genau ist. Eine groe Starke von mikroskopischen Model-len ist, dass nahezu alle technischen Details und lokalen Besonderhei-ten einstellbar sind und bei den Berechnungen ber ucksichtigt werden.Wir beschreiben das Bahnsystem auf einer mikroskopischen Ebene, sodass das Verhalten der Z uge und das Sicherheitssystem korrekt und

    vollst andig abgebildet sind. Diese Modelle der Schieneninfrastruktursind bereits f ur sehr kleine Netzausschnitte sehr gro. Der Grund ist,dass alle Signale, Neigungswechsel und Weichen im Vorfeld eines Bahn-hofes modelliert werden mussen, um pr azise Fahrzeitrechnungen zu er-lauben. Im Allgemeinen wird diese Art der Modellierung in Simula-tionssystemen benutzt, die nahezu bei jedem Bahnunternehmen rundum die Welt im Einsatz sind. Die bedeutenste Anwendung dieser Sy-steme ist einen einzelnen Fahrplan zu validieren und zu entscheiden, obein Fahrplan betrieblich umsetzbar und in der Realit at durchf uhrbarist. Mikroskopische Modelle sind jedoch aufgrund ihrer Gr oe und ih-rer hohen Detailtiefe ungeeignet f ur eine mathematischen Optimie-rung. Dementsprechend betrachten die meisten Optimierungsans atzevereinfachte, so genannte makroskopische , Modelle. Die Herausforde-rung besteht hierbei darin, ein zuverl assiges makroskopisches Modellf ur ein entsprechendes mikroskopisches Modell zu konstruieren und denUbergang zwischen beiden Modellen verschiedener Detailstufen zu er-leichtern.

    Zur Belelgung von Kapazit at im Bahnsystem k onnen signikante Teileder mikroskopischen Infrastruktur zu einem aggregierten Ressourcen-verbrauch in Raum und Zeit transformiert werden. Wir entwickeln eine

    allgemeine makroskopischen Darstellung des Schienensystems, die auf minimalen Zugfolgezeiten f ur das Einbrechen von Z ugen auf Gleisab-schnitten basiert und welche damit in der Lage ist, alle relevante Si-cherungssyteme im Schienenverkehr zu bew altigen. Wir f uhren einenneuartigen Bottom-up-Ansatz ein, um ein makroskopisches Modelldurch eine automatische Aggregation von Simulationsdaten eines mi-kroskopischen Modells zu generieren. Diese Transformation aggregiertund schrumpft das Infrastrukturnetz auf eine kleinere Darstellung, wo-bei alle Ressourcen- und Kapazit atsaspekte durch konservatives Run-den aller Zeiten erhalten bleiben. Der Hauptvorteil unseres Ansatzes

  • 8/12/2019 Schlechte Thomas

    17/244

    ist, dass wir garantieren k onnen, dass unsere makroskopischen Resul-tate, d.h. die Trassen der Z uge, nach der Rucktransformation auchim mikroskopischen Modell zulassig sind. Durch das konservative Run-den tendieren makroskopische Modelle die Kapazit at zu untersch atzen.Die Genauigkeit des makroskopischen Modells k onnen wir durch diegewahlte Zeitdiskretisierung steuern. Schlielich liefern wir eine a prio-ri Fehlerabsch atzung unseres Transformationsalgorithmus, d.h. in derBeurteilung der Uberschreitungen der Fahr- und Mindestzugfolgezei-ten.

    Im zweiten und Hauptteil (Kapitel III) der Dissertation wird das Pro-

    blem des Bestimmens optimaler Trassenallokationen f ur makroskopi-sche Bahnmodelle betrachtet. Ein Literatur uberblick zu verwandtenProblemen wird gegeben. F ur das Trassenallokationsproblem wird eingraphentheoretisches Modell entwickelt, in dem optimale L osungen alsmaximal gewichtete koniktfreie Menge von Pfaden in speziellen zeit-expandierten Graphen dargestellt werden k onnen. Des Weiteren er-reichen wir wesentliche Fortschritte beim L osen von Trassenallokati-onsprobleme durch zwei Hauptbeitr age - die Entwickling einer neuar-tigen Modellformulierung des makroskopischen Trassenallokationspro-blemes und algorithmische Verbesserungen basierend auf der Nutzung

    des Bundelverfahrens.Im Detail studieren wir vier verschiedene Typen von ganzzahligen Mo-dellformulierungen f ur das Trassenallokationsproblem: zwei Standard-formulierungen, die Ressourcen- oder Blockkonikte mit Hilfe von Pack-ungsungleichungen modellieren, und zwei neuartige Kopplungs- oderKongurationsmodelle. In beiden F allen werden Varianten mit ent-weder Bogen- oder Pfadvariablen pr asentiert. Die Kernidee dieser neu-en Modelle besteht darin, zus atzliche Kongurationsvariablen zu nut-zen, die, um Zulassigkeit zu sichern, mit den Standard Flussvariablender Zuge entsprechend gekoppelt werden. Wir zeigen, dass diese Model-le eine spezielle Formulierung, eine sogenannte extended formulation,der Standard Packungsmodelle sind.

    Der Erfolg eines ganzzahligen Programmierungsansatzes h angt ublicher-weise von der Starke der LP Relaxierung ab. Infolgedessen analysierenwir die LP Relaxierungen unserer Modellformulierungen. Wir zeigen,dass sich im Falle von Blockkonikten die Packungsbedingungen derStandardformulierung aus den Cliquen eines Intervallgraphen ergebenund diese sich deswegen in polynomieller Zeit bestimmen lassen. Wirbeweisen, dass die LP Relaxierung der extended formulation bei der

  • 8/12/2019 Schlechte Thomas

    18/244

    die Anzahl der Variablen exponentiell sein kann, ebenso in polynomi-eller Zeit gelost werden kann und dass diese Relaxierung diesselbe LPSchranke liefert. Des Weiteren beweisen wir, dass bestimmte Bedin-gungen der extended formulation Facetten des Polytops der entspre-chenden ganzzahligen Modellformulierung sind.

    Die Pfadvariante des Kongurationsmodells besitzt eine starke LP -Schranke, ist geeignet f ur Spaltenerzeugungstechniken und ist somitverwendbar zum L osen sehr groer Probleme. Des Weiteren pr asentierenwir ein fortgeschrittenen L osungsansatz, der in der Lage ist L osungenhoher Qualit at f ur groe Trassenallokationsprobleme zu berechnen. Un-

    ser Algorithmus ist eine Weiterentwicklung der rapid branching-Me-thode von Bornd orfer, Lobel & Weider (2008) [37] (siehe ebenso Wei-der (2007) [213]) zur Losung von integrierten Umlauf- und Dienstpla-nungsproblemen im offentlichen Personenverkehr. Die Methode l ost ei-ne Lagrange-Relaxierung des Trassenallokationsproblems als Grund-lage f ur einen branch-and-generate Algorithmus, der durch approxi-mative L osungen des Bundelverfahrens f ur das LP geleitet wird. Die-se erfolgreiche zweite Verkehrsanwendung liefert den Beleg, da dierapid branching-Methode ein vielversprechender allgemeiner Ansatzzum Losen groer Pfadpackungs- und Pfad uberdeckungsprobleme ist.Die neuen Modelle und Algorithmen sind im Software-Tool TS-OPTimplementiert.

    Abschlieend blicken wir zur uck zur praktischen Anwendung und pr a-sentieren im letzten Kapitel mehrere Fallstudien unter Verwendungder entwickelten Werkzeuge netcast und TS-OPT . Wir liefern einenausf uhrlichen Vergleich der Rechnungen unserer neuartigen Modellemit bekannten Standardmodellen aus der Literatur. Unsere Rechenre-sultate zeigen, dass der neuartige Ansatz, d.h. die Kongurationsmo-delle, andere Modelle in den meisten F allen ubertrifft. Zudem erm og-lichen uns die rapid branching-Heuristik und die B undelmethode

    qualitativ hochwertige L osungen f ur sehr groe Probleminstanzen zuproduzieren, was bisher nicht m oglich war. Daneben pr asentieren wirtheoretische und eher vision are Resultate f ur die Vergabe von Trasseninnerhalb eines Auktionsrahmens. Wir diskutieren verschiedene Frage-stellungen zur Auktionsform und analyzieren Simulationsexperimenteverschiedenener Auktionen.

    Den Hohepunkt bilden Resultate f ur Praxisszenarios zum Simplon Kor-ridor in der Schweiz. Nach bestem Wissen des Autors und best atigtdurch zahlreiche Eisenbahnpraktiker ist dies das erste Mal, dass auf ei-

  • 8/12/2019 Schlechte Thomas

    19/244

    ner makroskopischen Ebene automatisch erstellte Trassenallokationendie Bedingungen des urspr unglichen mikroskopischen Modells erf ullenund der Evaluierung innerhalb des mikroskopischen SimulationstoolsOpenTrack standhalten. Das dokumentiert den Erfolg unseres Ansatzesund den Nutzen and die Anwendbarkeit mathematischer Optimierungzur Allokation von Trassen im Schienenverkehr.

  • 8/12/2019 Schlechte Thomas

    20/244

  • 8/12/2019 Schlechte Thomas

    21/244

    AcknowledgementsFirst of all, I am very grateful to Prof. Dr. Dr. h. c. mult. M. Gr otschelfor having given me the possibility to stay at the Zuse Institute Berlinafter writing my diploma thesis. Thank you for the trust and thefreedom during these past years.

    A fundamental person all through the thesis work was my supervisorDr. habil. Ralf Borndorfer. You always had time for me even if youwere acquiring and heading thousands of projects. You always trustedme, taught me how to structure a project, how to get the big picture,how to identify open questions, where contributions are still neededwithout getting lost in all technical details, and many more. Specialthanks goes also to Dr. Steffen Weider who provided me his code of thebundle method and supported my adaption and further developmentof the rapid branching heuristic.

    Applied research is really applied only if it is done and evaluated in closecollaboration with an industrial and operating partner. Therefore, I amvery thankful for all discussions with external experts from Lufthansa

    Systems Berlin, DB Schenker, DB GSU, and in particular from SwissFederal Railways (SBB). Special thanks go to Thomas Graffagninoand Martin Balser for explaining various technical details from railwaysystems and discussing several results. In addition, I want to thankDaniel Hurlimann for his support for the simulation tool OpenTrack . Ialso greatly appreciated the contact with international colleagues fromAachen, Rotterdam, Delft, Bologna, Zurich, Chemnitz, Kaiserslauternand Darmstadt during several fruitful conferences.

    I would like to thank also all my colleagues at the department Op-timization that made my time as a PhD student so enjoyable. The

    vivid atmosphere of the Optimization group was also very enriching.In particular, the daily coffee breaks with - Kati, Stefan H., StefanV., Christian, Timo, Ambros, Jonas and all the others - has become akind of institution for reection and motivation. Furthermore, I wouldlike to thank Marika Neumann, Markus Reuther, R udiger Stephan, El-mar Swarat, Steffen Weider, and Axel Werner for proof-reading anddiscussing parts of my thesis. Last but not least I want to thank mygirlfriend Ina and my family for their patience and support.

  • 8/12/2019 Schlechte Thomas

    22/244

  • 8/12/2019 Schlechte Thomas

    23/244

    Table of Contents

    Table of Contents xix

    List of Tables xxiii

    List of Figures xxv

    I Planning in Railway Transportation 1

    1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

    2 Planning Process . . . . . . . . . . . . . . . . . . . . . . . . . 9

    2.1 Strategic Planning . . . . . . . . . . . . . . . . . . . . 12

    2.2 Tactical Planning . . . . . . . . . . . . . . . . . . . . . 122.3 Operational Planning . . . . . . . . . . . . . . . . . . 15

    3 Network Design . . . . . . . . . . . . . . . . . . . . . . . . . 17

    4 Freight Service Network Design . . . . . . . . . . . . . . . . . 19

    4.1 Single Wagon Freight Transportation . . . . . . . . . . 20

    4.2 An Integrated Coupling Approach . . . . . . . . . . . 21

    5 Line Planning . . . . . . . . . . . . . . . . . . . . . . . . . . 24

    6 Timetabling . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

    6.1 European Railway Environment . . . . . . . . . . . . . 28

    6.2 Periodic versus Trip Timetabling . . . . . . . . . . . . 336.2.1 Periodic Timetabling . . . . . . . . . . . . . . 346.2.2 Non periodic Timetabling . . . . . . . . . . . . 366.2.3 Conclusion . . . . . . . . . . . . . . . . . . . . 39

    xix

  • 8/12/2019 Schlechte Thomas

    24/244

    6.3 Microscopic versus Macroscopic Models . . . . . . . . 41

    7 Rolling Stock Planning . . . . . . . . . . . . . . . . . . . . . 42

    8 Crew Scheduling . . . . . . . . . . . . . . . . . . . . . . . . . 43

    8.1 Airline Crew Scheduling . . . . . . . . . . . . . . . . . 44

    8.2 Crew Scheduling Graph . . . . . . . . . . . . . . . . . 45

    8.3 Set Partitioning . . . . . . . . . . . . . . . . . . . . . . 46

    8.4 Branch and Bound . . . . . . . . . . . . . . . . . . . . 48

    8.5 Column Generation . . . . . . . . . . . . . . . . . . . 48

    8.6 Branch and Price . . . . . . . . . . . . . . . . . . . . . 51

    8.7 Crew Composition . . . . . . . . . . . . . . . . . . . . 52

    II Railway Modeling 54

    1 Microscopic Railway Modeling . . . . . . . . . . . . . . . . . 57

    2 Macroscopic Railway Modeling . . . . . . . . . . . . . . . . . 64

    2.1 Macroscopic Formalization . . . . . . . . . . . . . . . . 652.1.1 Train Types and Train Type Sets . . . . . . . 672.1.2 Stations . . . . . . . . . . . . . . . . . . . . . 682.1.3 Tracks . . . . . . . . . . . . . . . . . . . . . . 69

    2.2 Time Discretization . . . . . . . . . . . . . . . . . . . 75

    2.3 An Algorithm for theMicroMacroTransformation . . . . . . . . . . . . . . 83

    3 Final Remarks and Outlook . . . . . . . . . . . . . . . . . . . 88

    III Railway Track Allocation 90

    1 The Track Allocation Problem . . . . . . . . . . . . . . . . . 91

    1.1 Traffic Model Request Set . . . . . . . . . . . . . . . 92

  • 8/12/2019 Schlechte Thomas

    25/244

    1.2 Time Expanded Train Scheduling Digraph . . . . . . . 95

    2 Integer Programming Models for Track Allocation . . . . . . 106

    2.1 Packing Models . . . . . . . . . . . . . . . . . . . . . . 106

    2.2 Coupling Models . . . . . . . . . . . . . . . . . . . . . 111

    2.3 Polyhedral Analysis . . . . . . . . . . . . . . . . . . . 121

    2.4 Extensions of the Models . . . . . . . . . . . . . . . . 1262.4.1 Combinatorial Aspects . . . . . . . . . . . . . 1272.4.2 Robustness Aspects . . . . . . . . . . . . . . . 128

    3 Branch and Price for Track Allocation . . . . . . . . . . . . . 132

    3.1 Concept of TS-OPT . . . . . . . . . . . . . . . . . . . . 132

    3.2 Solving the Linear Relaxation . . . . . . . . . . . . . . 1343.2.1 Lagrangean Relaxation . . . . . . . . . . . . . 1353.2.2 Bundle Method . . . . . . . . . . . . . . . . . 136

    3.3 Solving the Primal Problem by Rapid Branching . . . 141

    IV Case Studies 148

    1 Model Comparison . . . . . . . . . . . . . . . . . . . . . . . . 148

    1.1 Effect of Flexibility . . . . . . . . . . . . . . . . . . . . 150

    1.2 Results for the TTPlib . . . . . . . . . . . . . . . . . . 153

    1.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . 1572 Algorithmic Ingredients for the (PCP) . . . . . . . . . . . . 158

    2.1 Results from the Literature . . . . . . . . . . . . . . . 159

    2.2 Bundle Method . . . . . . . . . . . . . . . . . . . . . . 161

    2.3 Rapid Branching . . . . . . . . . . . . . . . . . . . . . 166

    2.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . 170

  • 8/12/2019 Schlechte Thomas

    26/244

    3 Auction Results . . . . . . . . . . . . . . . . . . . . . . . . . 170

    3.1 The Vickrey Track Auction . . . . . . . . . . . . . . . 172

    3.2 A Linear Proxy Auction . . . . . . . . . . . . . . . . . 174

    3.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . 175

    4 The Simplon Corridor . . . . . . . . . . . . . . . . . . . . . . 176

    4.1 Railway Network . . . . . . . . . . . . . . . . . . . . . 176

    4.2 Train Types . . . . . . . . . . . . . . . . . . . . . . . . 178

    4.3 Network Aggregation . . . . . . . . . . . . . . . . . . . 179

    4.4 Demand . . . . . . . . . . . . . . . . . . . . . . . . . . 181

    4.5 Capacity Analysis based on Optimization . . . . . . . 183

    4.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . 189

    Bibliography 190

  • 8/12/2019 Schlechte Thomas

    27/244

    List of Tables

    I Planning in Railway Transportation 11 Planning steps in railroad traffic, source: Bussieck, Win-

    ter & Zimmermann (1997) [50]. . . . . . . . . . . . . . 102 Sizes of the solved instances in the literature for the TTP

    instance. . . . . . . . . . . . . . . . . . . . . . . . . . . 40

    II Railway Modeling 541 Technical minimum headway times with respect to run-

    ning mode . . . . . . . . . . . . . . . . . . . . . . . . . 702 Relation between the microscopic and the macroscopic

    railway model . . . . . . . . . . . . . . . . . . . . . . . 75

    III Railway Track Allocation 901 Denition of train request set . . . . . . . . . . . . . . 1042 Sizes of packing formulation for the track allocation prob-

    lem with block occupation . . . . . . . . . . . . . . . . 111

    IV Case Studies 1481 Size of the test scenarios req 36 . . . . . . . . . . . . . 1512 Solution statistic for model (APP) and variants of sce-

    nario req 36 . . . . . . . . . . . . . . . . . . . . . . . . 152

    3 Solution statistic for model (ACP) and variants of sce-nario req 36 . . . . . . . . . . . . . . . . . . . . . . . . 1524 Solution statistic of model (APP) for wheel -instances. 1535 Solution statistic of model (ACP) for wheel -instances. 1546 Solution statistic of model (APP) for hakafu simple -

    instances. . . . . . . . . . . . . . . . . . . . . . . . . . 1557 Solution statistic of model (ACP) for hakafu simple -

    instances. . . . . . . . . . . . . . . . . . . . . . . . . . 1568 Solution statistic of model (APP) for hard hakafu simple -

    instances. . . . . . . . . . . . . . . . . . . . . . . . . . 157

    xxiii

  • 8/12/2019 Schlechte Thomas

    28/244

    9 Solution statistic of model (ACP) for hard hakafu simple -instances. . . . . . . . . . . . . . . . . . . . . . . . . . 157

    10 Comparison of results for differrent models on the TTPlib-instances. . . . . . . . . . . . . . . . . . . . . . . . . . 158

    11 Solution statistic of TS-OPT and model (PCP) for wheel -instances. . . . . . . . . . . . . . . . . . . . . . . . . . 160

    12 Comparison of results for model (PPP) from Cacchi-ani, Caprara & Toth (2010) [ 54] for modied wheel -instances. . . . . . . . . . . . . . . . . . . . . . . . . . 160

    13 Statistic for solving the LP relaxation of model (PCP)with column generation and the bundle method. . . . . 163

    14 Solution statistic of bundle method and greedy heuristicfor model (PCP) for hakafu simple -instances. . . . 165

    15 Solution statistic of rapid branching with aggressive set-tings. . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

    16 Solution statistic of rapid branching with moderate set-tings. . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

    17 Solution statistic of rapid branching with default settings. 16918 Solution statistic of TS-OPT for model (PCP) for very

    large instances. . . . . . . . . . . . . . . . . . . . . . . 16919 Incremental auction with and without dual prices: prot

    and number of rounds until termination . . . . . . . . 17520 Statistics of demand scenarios for the Simplon case study 18121 Running and headway times for EC with respect to 18222 IP-Solution analysis of network simplon big with time

    discretization of 10s and a time limit of 24h . . . . . . 18423 Solution data of instance 24h-tp-as with respect to the

    chosen time discretization for simplon small . . . . 18624 Solution data of instance 24h-f15-s with respect to the

    chosen time discretization for simplon small . . . . 18625 Distribution of freight trains for the requests 24h-tp-as

    and 24h-f15-s by using network simplon big and a round-ing to 10 seconds . . . . . . . . . . . . . . . . . . . . . 187

  • 8/12/2019 Schlechte Thomas

    29/244

    List of Figures1 Structure of the thesis. . . . . . . . . . . . . . . . . . . v

    I Planning in Railway Transportation 11 Estimated demand for (freight) railway transportation in

    Germany, source: Federal Transport Infrastructure Plan-ning Project Group (2003) [87]. . . . . . . . . . . . . . 2

    2 Simplied routing network of Charnes & Miller (1956)[67]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

    3 Idealized planning process for railway transportation inEurope. . . . . . . . . . . . . . . . . . . . . . . . . . . 11

    4 Requested train paths at DB, source: Klabes (2010) [129]. 135 Possible train composition for track f = ( vr , 14, wb, 20, 4). 236 Visualization of line plan for Potsdam. . . . . . . . . . 257 Screenshot of visualization tool for public transport net-

    works. . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 Timeline for railway capacity allocation in Europe, source:

    Klabes (2010) [129]. . . . . . . . . . . . . . . . . . . . 319 Simple conict example and re-solution for track alloca-tion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

    10 Principal methods in the literature for macroscopic time-tabling by Caimi (2009) [57]. . . . . . . . . . . . . . . 33

    11 A partial cyclic rolling stock rotation graph visualized inour 3D visualization Tool TraVis using a torus to dealwith the periodicity. . . . . . . . . . . . . . . . . . . . 44

    12 Crew Scheduling Graph. . . . . . . . . . . . . . . . . . 4613 Set of legs (above) and a set of covering pairings (below)

    show as a Gant chart in the planning tool NetLine. . . 4714 General column generation approach to solve LPs with a

    large column set. . . . . . . . . . . . . . . . . . . . . . 49

    II Railway Modeling 541 Idealized closed loop between railway models of different

    scale for railway track allocation. . . . . . . . . . . . . 552 Detailed view of station Altenbeken provided by DB Netz

    AG, see Altenbeken [11] . . . . . . . . . . . . . . . . . 58

    xxv

  • 8/12/2019 Schlechte Thomas

    30/244

    3 Screenshot of the railway topology of a microscopic net-work in the railway simulator OpenTrack . Signals can beseen at some nodes, as well as platforms or station labels. 59

    4 Idea of the transformation of a double vertex graph to astandard digraph . . . . . . . . . . . . . . . . . . . . . 59

    5 Microscopic network of the Simplon and detailed repre-sentation of station Iselle as given by OpenTrack . . . 61

    6 Blocking time diagrams for three trains on two routesusing 6 blocks. In the lower part of the diagram twosubsequent trains on route r2 and at the top one train onthe opposite directed route r1 are shown. . . . . . . . . 62

    7 I/O Concept of TTPlib 2008 (focus on macroscopic rail-way model) . . . . . . . . . . . . . . . . . . . . . . . . 64

    8 Example of macroscopic railway infrastructure . . . . . 669 Example of aggregated infrastructure . . . . . . . . . . 6710 Train types and train sets dened as a poset . . . . . . 6811 Macroscopic modeling of running and headways times on

    tracks . . . . . . . . . . . . . . . . . . . . . . . . . . . 7212 Macroscopic modeling of a single way track . . . . . . 7213 Representation as event-activity digraph G = ( V N , AN ) 7314 Implausible situation if headway matrix is not transitive 7415 Transformation of running time on track A B for timediscretizations between 1 and 60 seconds . . . . . . . . 8016 Rounding error for different time discretizations between

    1 and 60 seconds, comparison of ceiling vs. cumulativerounding . . . . . . . . . . . . . . . . . . . . . . . . . 81

    17 Headway time diagrams for three succeeding trains onone single track ( j1, j 2) . . . . . . . . . . . . . . . . . . 83

    18 Constructed aggregated macroscopic network by netcast. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

    19 New routing possibilities induced by given routes . . . 8720 Macroscopic network produced by netcast visualize by

    TraVis . . . . . . . . . . . . . . . . . . . . . . . . . . 87

    III Railway Track Allocation 901 Concept of TTPlib 2008 (focus on train demand speci-

    cation and TTP ) . . . . . . . . . . . . . . . . . . . . . 922 Penalty functions for departure(left) and arrival(right)

    times. . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

  • 8/12/2019 Schlechte Thomas

    31/244

    3 Prot function w() depending on basic prot and depar-ture and arrival times. . . . . . . . . . . . . . . . . . . 94

    4 Explicit and implicit waiting on a timeline inside a station 985 Complete time expanded network for train request . . 1016 Irreducible graph for train request . . . . . . . . . . . 1027 Preprocessed time-expanded digraph D = ( V, A) of ex-

    ample 1.6 . . . . . . . . . . . . . . . . . . . . . . . . . 1058 Example for maximum cliques for block occupation con-

    icts. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1099 Example for an equivalence class and a hyperarc. . . . 113

    10 Example for the construction of a track digraph. . . . 11411 Example for a path which does not correspond to a validconguration if the headway times violate the transitiv-ity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

    12 Relations between the polyhedra of the different models. 12213 Idea of the extended formulation (PCP) for (PPP) . . 12414 From fragile q 1 and q 2 to robust conguration q 3. . . . 13015 Robustness function r of two buffer arcs. . . . . . . . . 13016 Pareto front on the left hand and total prot objective

    (blue, left axis) and total robustness objective (green,

    right axis) in dependence on on the right hand. . . . 13217 Flow chart of algorithmic approach in TS-OPT . . . . . . 13418 Cutting plane model f P,Q of Lagrangean dual f P,Q . . . 13819 The new solution sets at iteration k, source: Weider

    (2007) [213] . . . . . . . . . . . . . . . . . . . . . . . . 146

    IV Case Studies 1481 Infrastructure network (left), and train routing digraph

    (right); individual train routing digraphs bear differentcolors. . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

    2 Reduction of graph size by trivial preprocessing for sce-narios req 36 and = 20. . . . . . . . . . . . . . . . 151

    3 Artical network wheel , see TTPlib [208] . . . . . . . 1544 Solving the LP relaxation of model (PCP) with column

    generation and the barrier method. . . . . . . . . . . . 1615 Solving the LP relaxation of model (PCP) with the bun-

    dle method. . . . . . . . . . . . . . . . . . . . . . . . . 1626 Testing different bundle sizes. . . . . . . . . . . . . . . 164

  • 8/12/2019 Schlechte Thomas

    32/244

    7 Solving a track allocation problem with TS-OPT ; dual(LP) and primal (IP) stage. . . . . . . . . . . . . . . . 167

    8 Solving track allocation problem req 48 with TS-OPT . 1689 Auction procedure in general . . . . . . . . . . . . . . 17110 Micro graph representation of Simplon and detailed rep-

    resentation of station Iselle given by OpenTrack . . . . . 17711 Given distribution of passenger or xed traffic in the Sim-

    plon corridor for both directions. . . . . . . . . . . . . 17912 Traffic diagram in OpenTrack with block occupation for

    request 24h-tp-as . . . . . . . . . . . . . . . . . . . . . 18513 Comparison of scheduled trains for different networks

    (simplon ) for instance 24h-tp-as in a 60s discretisation 18714 Distribution of freight trains for the requests 24h-tp-as

    and 24h-f15-s by using network simplon big and a round-ing to 10 seconds . . . . . . . . . . . . . . . . . . . . . 188

  • 8/12/2019 Schlechte Thomas

    33/244

    List of Algorithms

    I Planning in Railway Transportation 1

    II Railway Modeling 541 Cumulative rounding method for macroscopic running time

    discretization . . . . . . . . . . . . . . . . . . . . . . . . 772 Calculation of Minimal Headway Times . . . . . . . . . 813 Algorithm for the Micro-Macro-Transformation in netcast

    . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

    III Railway Track Allocation 904 Construction of D. . . . . . . . . . . . . . . . . . . . . . 1005 Proximal Bundle Method (PBM) for (LD) of (PCP) . . 1396 Perturbation Branching. . . . . . . . . . . . . . . . . . . 145

    IV Case Studies 148

    xxix

  • 8/12/2019 Schlechte Thomas

    34/244

  • 8/12/2019 Schlechte Thomas

    35/244

    Chapter I

    Planning in Railway Trans-

    portation

    The purpose of our work is to develop mathematical optimization mod-els and solution methods to increase the efficiency of future railwaytransportation systems. The reasons for this is manifold; liberaliza-tion, cost pressure, environmental and energy considerations, and theexpected increase of the transportation demand are all important fac-tors to consider. Every day, millions of people are transported by trains

    in Germany. Public transport in general is a major factor for the pro-ductivity of entire regions and decides on the quality of life of people.

    Figure 1 shows the expected development of freight transportation inGermany from 2003 to 2015 as estimated by the Deutsche Bahn AG(DB AG). This estimate was the basis of the last German FederalTransport Infrastructure Plan 2003 (Bundesverkehrswegeplan 2003),see Federal Transport Infrastructure Planning Project Group (2003)[87]. It is a framework investment plan and a planning instrumentthat follows the guiding principle of development of Eastern Germanyand upgrading in Western Germany. The total funding available forroad, rail and waterway construction for the period from 2001 to 2015is around 150 billion euros.

    The railway industry has to solve challenging tasks to guarantee or evenincrease their quality of service and their efficiency. Besides the needto implement adequate technologies (information, control, and book-ing systems) and latest technology of equipment and resources (trains,railway infrastructure elements), developing mathematical support sys-tems to tackle decision, planning, and in particular optimization prob-lems will be of major importance.

    1

  • 8/12/2019 Schlechte Thomas

    36/244

    2

    Figure 1: Estimated demand for (freight) railway transportation in Germany,source: Federal Transport Infrastructure Planning Project Group(2003) [87].

    In Section 1 we will give a comprehensive introduction on the politicalenvironment and organizational structures because both directly affectthe planning and operation of railway transport. In addition, we willrefurbish an early publication from Charnes & Miller (1956) [ 67] thatdemonstrates prominently that railway transport is one of the initialapplication areas for mathematics in particular for discrete and linearoptimization.

    Only recently railway success stories of optimization models are re-ported from Liebchen (2008) [149], Kroon et al. (2009) [140], and Caimi(2009) [57] in the area of periodic timetabling by using enhanced inte-ger programming techniques. This thesis focuses on a related planningproblem the track allocation problem . Thus, Section 2 gives a generaloverview of an idealized planning process in railway transportation.We will further describe several other planning problems shortly in-

    cluding line planning in Section 5 and crew scheduling in Section 8 inmore detail. Mathematical models and state of the art solution ap-proaches will be discussed, as well as the differences to and similaritieswith equivalent planning tasks of other public transportation systems.Moreover, in Section 6 we will depict the requirements and the processof railway capacity allocation in Europe to motivate and establish ageneral formulation for the track allocation problem.

    We will show how to establish a general framework that is able tohandle almost all technical details and the gigantic size of the railway

  • 8/12/2019 Schlechte Thomas

    37/244

    3

    infrastructure network by a novel aggregation approach. Therefore,and to build a bridge to railway engineering, we explain the most im-portant microscopic technical details in Chapter II. Furthermore, weintroduce a general standard for macroscopic railway models which ispublicly available TTPlib [208] and develop a multi-scale approach thatautomatically transforms microscopic railway models from real worlddata to general macroscopic models with certain error estimations.

    Nevertheless, the resulting macroscopic track allocation problems arestill very large and complex mathematical problems. From a complex-

    ity point of view track allocation problems belongs to the class of NP -hard problems. In order to produce high quality solutions in reasonabletime for real world instances, we develop a strong novel model formu-lation and adapt a sophisticated solution approach. We believe thatthis modeling technique can be also very successful for other problems in particular if the problem is an integration of several combinatorialproblems which are coupled by several constraints. Chapter III willintroduce and analyze this novel model formulation called congura-tion model in case of the the track allocation problem. Furthermore,we will generalize and adapt the rapid branching heuristic of Weider

    (2007) [213]. We will see that we could signicantly speed up ourcolumn generation approach by utilizing the bundle method to solvethe Lagrangean relaxation instead of using standard solvers for the LPrelaxations.

    Finally, to verify our contributions on modeling and solving track allo-cation problems in Chapter IV, we implemented several software toolsthat are needed to establish a track allocation framework:

    a transformation module that automatically analyses and simpli-es data from microscopic simulation tools and provides reliablemacroscopic railway models ( netcast ),

    an optimization module that produces high quality solutions (to-gether with guaranteed optimality gaps) for real world track al-location problems in reasonable time ( TS-OPT ),

    and a 3d-visualization module to illustrate the track allocationproblem, to discuss the solutions with practitioners, and to au-tomatically provide macroscopic statistics ( TraVis ).

  • 8/12/2019 Schlechte Thomas

    38/244

    1 Introduction 4

    1 Introduction

    Railway systems can be categorized as either public or private. Pri-vate railway systems are owned by private companies and are with afew exceptions exclusively planned, built, and operated by this sin-gle owner. Prominent examples are the railway systems in Japan andthe US, see Gorman (2009) [102]; Harrod (2010) [112]; White & Krug(2005) [215]. In contrast, public railway systems are generally fundedby public institutions or governments. In the past an integrated rail-way company was usually appointed to plan, build, and operate therailway system. Now the efforts of the European Commission to seg-regate the integrated railway companies into a railway infrastructuremanager (network provider) and railway undertakings (train operatingcompanies) shall ensure open access to railway capacity for any licensedrailway undertaking. The idea is that competition leads to a more ef-cient use of the railway infrastructure capacity, which in the long runshall increase the share of railway transportation within the Europeanmember states. However, even in case of an absolute monopoly theplanning of railway systems is very complex because of the technicali-ties and operational rules. This complexity is further increased by thevarying requirements and objectives of different participating railwayundertakings in public railway systems.

    The focus of this work is capacity allocation in an arbitrary railwaysystem. In a nutshell, the question is to decide which train can usewhich part of the railway infrastructure at which time. Chapter Iaims to build an integrated picture of the railway system and railwayplanning process, i.e., we will illuminate the requirements of passengerand freight railway transportation. In Chapter II resource models willbe developed that allow for capacity considerations. Based on one

    of these railway models, i.e., an aggregated macroscopic one, we willformulate a general optimization model for private and public railwaysystems in Chapter III, which meets the requirements of passenger andfreight railway transportation to a large extent.

    Several railway reforms in Europe were intended to promote on-railcompetition leading to more attractive services in the timetable. How-ever, even after the reforms were implemented, the railways continuedto allocate train paths on their own networks themselves. Discrimi-nation was thus still theoretically possible. However, competition can

  • 8/12/2019 Schlechte Thomas

    39/244

    1 Introduction 5

    only bring benets if all railway undertakings are treated equally whenseeking access to the infrastructure.

    Switzerland has been pioneer in introducing competition in the use of the rail networks. The three different Swiss railway network providers,SBB, BLS and SOB, outsourced the allocation of their train paths to a joint independent body. Accordingly, at the beginning of 2006, and inconjunction with the Swiss Public Transport Operators Association,these railways together founded the Trasse Schweiz AG (trasse.ch).

    By outsourcing train path allocation to a body which is legally in-dependent and independent in its decision making, the three largestSwiss standard gauge railways together with the Swiss Public Trans-port Operators Association reinforced their commitment to fair on-railcompetition. This institution ensures that the processes to prepare forthe timetable are free of discrimination. Trasse Schweiz AG coordi-nates the resolution of conicts between applications and allocate trainpaths in accordance with the legislation. One of their principles is:

    We increase the attractiveness of the rail mode by makingthe best use of the network and optimizing the applicationprocesses.

    That statement essentially summarizes the main motivation of thisthesis.

    An initial publication on applying linear optimization techniques comesfrom railway freight transportation. Charnes & Miller (1956) [67] dis-cussed the scheduling problem of satisfying freight demand by traincirculations. The setting is described by a small example in Figure 2.In a graph with nodes #1 , #2, and #3 a directed demand which hasto be satised is shown on each arc. The goal is to determine directedcycles in that graph that cover all demands with minimal cost, i.e.,

    each cycle represents a train rotation. For example choosing four timesthe rotation (#1 , #2 , #1) would cover all required freight movementsbetween #1 and #2. However the demand from #2 to #1 is only oneand therefore that would be an inefficient partial solution with threeempty trips called light moves in the original work. Charnes andMiller proposed a linear programming formulation for the problem enu-merating all possible rotations, i.e., ve directed cycles (#1 , #2 , #1),(#1 , #3 , #1), (#2 , #3 , #2) in Figure 2. Multiple choices of cycles thatsatisfy all demands represent a solution. Thus, for each rotation aninteger variable with crew and engine cost was introduced. The opti-

  • 8/12/2019 Schlechte Thomas

    40/244

    1 Introduction 6

    #1

    #2

    #3

    4 1

    6 6

    5 9

    Figure 2: Simplied routing network of Charnes & Miller (1956) [ 67].

    mization model states that the chosen subset has to fulll all demands.This was one of the rst approaches to solve real applications by meansof a set partitioning problem, i.e., to represent a solution as a set of sub-solutions, here cycles. Finally, they manually solved the instanceby applying the simplex tableau method.

    After that pioneering work on modeling it took many years of improve-ment in the solution techniques to go a step further and to support morecomplex planning challenges in public transportation and in particularin railway transportation by optimization.

    In fact the airline industry became the driving force of the development.One reason is the competitive market structure which leads to a highercost pressure for aviation companies. Therefore the airline industryhas a healthy margin in the implementation of automated processesand the evaluation of operations. Integrated data handling, measuringthe quality of service and controlling the planning and operation byseveral key performance indicators (KPI) are anchored in almost allaviation companies over the world. Nowadays in the airline industrythe classical individual planning problems of almost all practical prob-lem sizes can be solved by optimization tools. Integration of differentplanning steps and the incorporation of uncertainty in the input datacan be tackled. A prominent example for such robust optimization ap-

    proaches is the tail assignment problem which is the classical problemof assigning ights to individual aircraft. Nowadays robust versionscan be tackled by stochastic optimization, see Lan, Clarke & Barn-hart (2006) [144], or a novel probability of delay propagation approachby Bornd orfer et al. (2010) [41]. Suhl, Duck & Kliewer (2009) [205] usesimilar ideas and extensions to increase the stability of crew schedules.

    An astonishing situation happened in Berlin which somehow documentsthe challenges and problems that might result from the deregulation.The British Financial Times wrote on 27th of July 2009:

  • 8/12/2019 Schlechte Thomas

    41/244

    1 Introduction 7

    Concrete walls, watch-towers, barbed wire and armed bor-der guards for decades prevented Germans travelling acrossBerlin from the east to the west. But, as the German capi-tal gears up to celebrate 20 years since the fall of the BerlinWall, leftwing commentators are claiming that capitalism,not communism, is now keeping the two apart. For the S-Bahn - the suburban commuter railway running into andaround Berlin that became a symbol of the cold war divide- has come grinding to a halt.

    More than two-thirds of the networks 550 trains werewithdrawn from service last week and the main east-westline closed after safety checks following a derailment showedthat about 4,000 wheels needed replacing. Hundreds of thousands of Berliners have been forced to get on theirbikes or use alternative, overcrowded routes to work, whiletourists weaned on stereotypical notions of German punc-tuality and efficiency have been left inconvenienced and be-mused by the chaos. Deutsche Bahn, the national railwayoperator, is under re for cutting staff and closing repairworkshops at its S-Bahn subsidiary in an attempt to boostprotability ahead of an initial public offering, that has

    since been postponed.For businesses dependent on the custom of S-Bahn pas-

    sengers, the partial -suspension of services is no joke. Forthe past two or three days its been really bad. Customersare down by more than half, said an employee at a clothing-alteration service situated below the deserted S-Bahn plat-form at Friedrichstrasse station, in the former East Berlin.German trains are world famous. I didnt think -somethinglike this could happen.

    A columnist for Tagesspiegel, a Berlin-based newspa-

    per, drolely observed that the number of S-Bahn carriagesrendered unusable by management incompetence was onlyslightly less than the total number damaged by the RedArmy in 1945. Others note that even the Berlin Wall itself did not prevent S-Bahn passengers traveling between westand east, so long as they held a West German passport.The East German authorities continued to operate the S-Bahn in West Berlin after the partition of the city followingthe second world war until the 1980s. West Berliners even-tually boycotted this service in protest of the communist

  • 8/12/2019 Schlechte Thomas

    42/244

    1 Introduction 8

    regime. But now it is being claimed that capitalism is driv-ing passengers away.

    The chaos in the Berliner S-Bahn is a lesson in the con-sequences of capitalism. It is a graphic depiction of wheresubservience to nancial markets greedy pursuit of protultimately leads, Ulrich Maurer, chief whip of the radicalLeft party, said. Deutsche Bahn has apologized for the in-convenience but insists that cost-cutting was not the prob-lem and blames the train manufacturer instead. Even if wehad had twice as many employees and three times as manyworkshops . . . it would not have prevented these wheels frombreaking, a Deutsche Bahn spokesman said. Nevertheless,S-Bahn-Berlins entire senior management was forced to re-sign this month after it emerged that they had not orderedsufficient safety checks. The repairs, refunds, and lost farescould leave Deutsche Bahn up to 100 million euros out of the pocket, according to one estimate. A full service is notexpected to resume until December.

    The described situation documents that the railway system in Europehas to face huge challenges in implementing the liberalization. In ad-dition central topics of the railway system are often politically andsocially sensitive subjects. A detailed characterization of the recentpolitical situation of the German railway system, future perspectives,the role of the infrastructure, and other controversial issues can befound in G.Ilgmann (2007) [99]. All in all we hope and we believethat an innovation process in the railway system in Europe is going tostart. Major railway planning decisions can be supported by mathe-matical models and optimization tools in the near future, in particularthe almost manual construction of the timetables and track allocationswhich is often seen as the heart of the railway system.

    Due to the deregulation and the segregation of national railway com-panies in Europe the transfer of mathematical optimization techniquesto railway operations will proceed. In the future, competition willhopefully give rise to efficiency and will lead to an increasing use of information technology and mathematical models. Algorithmic deci-sion support to solve the complex and large scale planning problemsmay become necessary tools for railway transportation companies. Inthe future state of the art planning systems with optimization insidewill replace the manual solution. The key message is that optimiza-

  • 8/12/2019 Schlechte Thomas

    43/244

    2 Planning Process 9

    tion, i.e., mathematical models and solution methods, are predestinedto support railway planning challenges now and in the future.

    In the following section we will briey highlight several of these plan-ning problems from different transportation modes. We will presentmathematical models and discuss state of the art solution approachesto tackle real world applications, see Barnhart & Laporte (2007) [ 17]for an overview on optimization in transportation in general. We use inthis thesis the denitions and notation of Gr otschel, Lovasz & Schrijver(1988) [104] and Nemhauser & Wolsey (1988) [167] for graphs, linearprograms (LPs), and mixed integer programs (MIPs). Furthermore, we

    use the algorithmic terminology to LP and MIP solving of Achterberg(2007) [3].

    2 Planning Process

    Bussieck, Winter & Zimmermann (1997) [50] divide the planning pro-cess in public transport into three major steps - strategic, tactical andoperational planning. Table 1 shows the goals and time horizon of all steps. Public transport, especially railway transportation, is sucha technically complex and large system that it is impossible to con-sider the entire system at once. Also, the different planning horizons of certain decisions enforce a decomposition. Therefore a sequence of hier-archical planning steps has emerged over the years. However, in realitythere is no such standardization as we will explain it theoretically.

    Two important parties are involved in the railway transportation plan-ning process, i.e., train operating companies and railway infrastructureproviders. Following the terminology of the European commission,we will use the terms railway undertaking (RU) and infrastructuremanager (IM), respectively. Furthermore, several national and inter-national institutions have a huge political inuence on railway trans-portation, which is on the borderline between a social or public goodand a product that can be traded on a free liberalized market. Thespecial case of the changing railway environment in Europe will bediscussed in detail in Section 6.1.

    In contrast to railway undertakings fully private aviation or independ-ent urban public transport companies can perform the complete plan-ning process almost internally. In the airline industry the needed infras-tructure capacity, i.e., the slots at the airports, are granted by grandfa-

  • 8/12/2019 Schlechte Thomas

    44/244

    2 Planning Process 10

    level time horizon goal

    strategic 5-15 years resource acquisitiontactical 1-5 years resource allocationoperational 24h - 1 year resource consumption

    Table 1: Planning steps in railroad traffic, source: Bussieck, Winter & Zimmer-mann (1997) [50].

    ther rights, see Barnier et al. (2001) [ 21]; Castelli, Pellegrini & Pesenti(2010) [66]. Borndorfer, Gr otschel & Jaeger (2008) [36]; Borndorfer,Grotschel & Jaeger (2009) [39] and Hanne & Dornberger (2010) [108]give recent surveys about the potential of optimization for transporta-tion systems and the differences between the planning process in theairline industry, urban public transport, and the railway industry. Inthe case of urban public transport the planning process is discussed inWeider (2007) [213] and Bornd orfer, Gr otschel & Pfetsch (2007) [35]. Adetailed description of the process in the airline industry can be foundin Gronkvist (2005) [103] and Barnhart & Laporte (2007) [ 17]. Bussieck(1997) [49] describes the use of discrete optimization in the planningprocess of public rail transport in the case of an integrated system.Analogous considerations can be found in Liebchen (2006) [148] and

    Lusby et al. (2009) [159]. There the planning steps are classied withrespect to the time horizon and their general purposes.

    Strategic or long-term part concerns the issues of network design andline planning (resource acquisition), see Sections 3 and 5. On the tac-tical stage the level of services, usually a timetable has to be created,as well as the schedules for the needed resources (resource allocation).Finally, on the operational stage the resources, e.g., rolling stock, ve-hicles, aircraft and crews, are monitored in real operations (resourceconsumption).

    On the day of operation re-scheduling and dispatching problems haveto be faced. These kind of problems have a different avor than pureplanning tasks. Decisions must be made very quickly in the real-timesetting, but only limited information on the scenario is available.Usually data has to be taken into consideration in a so called onlinefashion. More details about this kind of problem can be found inGrotschel, Krumke & Rambau (2001) [105], Albers & Leonardi (1999)[9], and Albers (2003) [8]. Recent approaches are to establish fast meth-ods which bring the real situation back to the planned one when

  • 8/12/2019 Schlechte Thomas

    45/244

    2 Planning Process 11

    Railway undertakings (RU) Infrastructure manager (IM)

    Network Design

    Line Planning

    Timetabling Track allocation

    Rolling Stock Planning

    Crew Scheduling

    Real Time Management Re-Scheduling

    level

    strategic

    tactical

    operational

    Figure 3: Idealized planning process for railway transportation in Europe.

    possible, see Potthoff, Huisman & Desaulniers (2008) [ 177], Rezanova& Ryan (2010) [182], and Jespersen-Groth et al. (2009) [ 123].

    In Klabes (2010) [129] the planning process is newly considered for thecase of the segregated European railway system. In Figure 3 the novelprocess is illustrated for the segregated railway industry in Europe.

  • 8/12/2019 Schlechte Thomas

    46/244

    2 Planning Process 12

    2.1 Strategic Planning

    The responsibilities of the planning steps refer directly to either the rail-way undertaking or the infrastructure manager on behalf of the state.Nevertheless the long-term decisions in up- or downgrading the networkare highly inuenced by the railway undertakings and their demands.In case of passenger railway undertakings the desired timetable aimsto implement a given line plan. The timetable itself induces train slotsrequests which is one input for the track allocation problem. Theseare naturally very strict with respect to departure and arrival times inorder to offer and operate a concrete and reliable timetable. Furtherdetails on line planning and periodic timetabling are given in Section 5and Section 6.2, respectively.

    The requirements of train slot requests for cargo or freight railway oper-ators differ signicantly from slot requests for passenger trains becausethey usually have more exibility, i.e., arrival and departure are onlyimportant at stations where loading has to be performed. Section 3will describe the network design problem of the major European singlewagon railway transportation system. In general freight railway oper-ators need a mixture of annual and ad hoc train slots. The demand is

    of course highly inuenced by the industry customers and the freightconcept of the operating railway undertaking. We collected such datafor the German subnetwork hakafu simple to estimate the demandof the railway freight transportation, see Chapter IV Section 1 andSchlechte & Tanner (2010) [189].

    2.2 Tactical Planning

    The essential connection between all train slot requests is the step to

    determine the complete track allocation, which is the focus of this work.However, we primarily consider the point of view of a railway infras-tructure provider, which is interested in optimizing the utilization of the network. That is to determine optimal track allocations. This is incontrast to timetabling where one asks for the ideal arrival and depar-ture times to realize a timetable concept or a line plan. A timetablecan be seen as a set of train slot requests without exibility. Railwayoptimization from a railway undertakings point of view for passengertraffic is discussed in Caprara et al. (2007) [ 64]. State of the art model-ing and optimization approaches to periodic timetabling, which is the

  • 8/12/2019 Schlechte Thomas

    47/244

    2 Planning Process 13

    2003 2004 2005 2006 2007 2008 20090

    2

    4

    6

    8104

    year

    n u m

    b e r o f t r a i n s l o t r e q u e s t s

    totalDB railway undertakingsnon-DB railway undertakings

    2003 2004 2005 2006 2007 2008 20090

    50

    100

    150

    year

    r e j e c t e d

    Figure 4: Requested train paths at DB, source: Klabes (2010) [ 129].

    usual type of schedule for passenger railway traffic, is at length studiedby Liebchen (2006) [148].

    The induced competition for railway capacity allocation in public rail-way systems in Europe has a several impacts on the allocation pro-cedure. In the past, a single integrated railway company performedthe complete planning. Its segregation reduces the ability of the rail-way infrastructure manager to only perform network planning, capac-ity allocation, and re-scheduling with respect to infrastructure aspects.Thus, the infrastructure manager only has limited information duringthe planning process and needs to respect the condential informationof the railway undertakings. Moreover, new railway undertakings en-ter the market which increases the complexity of the planning process.Klabes (2010) [129] collected the relevant numbers from the DB Netzreports. On the left hand of Figure 4 the changing environment isillustrated by listing the growing number of train slot requests fromrailway undertakings independent from the former integrated railwaycompany Deutsche Bahn. On the right hand of Figure 4 the numberof rejected train slot requests for the same periods are shown. It canbe seen that at the start of the segregation from 2003 until 2006 a lotof requests had to be rejected by DB Netz. Efforts to decrease thesenumbers by providing alternative slots were apparently successful inthe following years.

    The business report for the year 2009 Trasse Schweiz AG [207] of theTrasse Schweiz AG documents the new challenges for constructingtrack allocations as well. In the Swiss network a lot of different railwayundertakings are operating, e.g., in 2009 there were 29 train operat-ing companies which submitted train slot requests. The geographicalposition in central Europe and the limited transportation possibilitiesthrough the Alps causes that. The future challenge for Switzerland

  • 8/12/2019 Schlechte Thomas

    48/244

    2 Planning Process 14

    will be to handle the complex track allocation process as the followingextract from the report 2009 already highlights:

    The regulation of the conicts arising in train slot ordersof the annual timetable 2010 was despite or even less be-cause of the nancial or economic crisis in comparison tothe last years extensive and time-consuming. Indeed thenumber of submitted train slot requests by cargo operatorsfor the annual timetable 2010 decreased up to 10 percent incomparison to the last year. However, railway undertakings(RM) concentrated her orders due to the cost pressure and

    competitive market conditions on the most attractive timewindows and stick much longer to their original requests.Nevertheless, we managed together with all infrastructureproviders 1 to nd for all conicts alternative train slots,which were accepted by the railway undertakings. No trainslot request had to be rejected. (translation by the au-thor).

    The competing railway undertakings should interact in a transparentand free market. The creation of such a market for railway capacity isa key target of the European Commission, hoping that it will lead to

    a more economic utilization of the railway infrastructure. Even moreliberalization of the railway system should lead to a growing marketand allow for innovative trends like in other old-established industries,i.e., aviation industry, telecommunication or energy market. After theacceptance of train slots each railway undertaking determines his par-tial operating timetable, which acts as input for the planning of theneeded resources. In case of a railway operator the rolling stock ro-tations have to be constructed, which is very complex problem dueto several regularities and maintenance requirements, see Fioole et al.(2006) [88],Anderegg et al. (2003) [12], Eidenbenz, Pagourtzis & Wid-

    mayer (2003) [80], and Peeters & Kroon (2008) [176].In public transport and in airline industry vehicle scheduling and air-craft rotation planning are the analogous tasks, see L obel (1997) [155]and Gr onkvist (2005) [103]. The major objective is to operate a re-liable timetable with minimum cost, which is in general minimizingthe number of engines, wagons, vehicles, aircrafts, etc. Another keyrequirement for planning railway rolling stock rotations is to provide

    1 There are three different railway infrastructure providers in Switzerland, i.e.,BLS, SBB, and SOB.

  • 8/12/2019 Schlechte Thomas

    49/244

    2 Planning Process 15

    regularity of the solutions. This means that a train that runs in thesame way every day of the week, will also be composed in the sameway every day of the week, always using the same cars from the samepreceding trains. Such a regime simplies the operation of a railwaysignicantly. However, the rule can not always be followed. Trains mayrun later on weekends, or not at all on certain days, e.g., in order toperform a maintenance operation. Although it is intuitively clear, it isnot easy to give a precise denition what regularity actually means.

    The output of rolling stock planning is to assign trains, i.e., specictrain congurations, to each passenger trip, to select deadhead trips,

    i.e., empty movements of the trains given by the constructed rollingstock rotation, and to schedule maintenances and turn around activitiesof trains. Passenger trips, that are trips of the published timetable, anddeadhead trips need to be assigned to crews, which have to executethem. We will describe this planning step in more detail in Section 8in case of an aviation company. This demonstrates the power of generalmathematical modeling and methodology to different applications andthat the authors experience about that planning step comes from airlinecrew scheduling, i.e., pairing optimization. However, recent work onrailway crew scheduling can be found in Abbink et al. (2005) [1] andBengtsson et al. (2007) [24].

    2.3 Operational Planning

    As already mentioned real time problems on the day of operation havequite different requirements, even if these problems can be formulatedvery similar from a mathematical modeling point. In railway trans-portation disruption and delay management is very difficult becauselocal decisions have a huge inuence on the complete timetable system.Nevertheless, easy and fast rules of thumb are used to decide, whichtrains have to be re-routed, have to wait, or even have to be canceled.DAriano et al. (2008) [72] and Corman, Goverde & DAriano (2009)[71] presented a real-time traffic management system to support localdispatching in practice. On the basis of this renewed timetable rollingstock rosters and crew schedules have to be adopted, see Clausen et al.(2010) [69]; Jespersen-Groth et al. (2007) [122]; Potthoff, Huisman &Desaulniers (2008) [177]; Rezanova & Ryan (2010) [182].

    Every single step in this idealized sequential planning process is a diffi-cult task by itself or even more has to be further divided and simplied

  • 8/12/2019 Schlechte Thomas

    50/244

    2 Planning Process 16

    into subproblems. We will discuss several of them in the following sub-sections, see how they can be modeled as combinatorial optimizationproblems, and solved by state of the art solution approaches.

    The main application of track allocation is to determine the best opera-tional implementable realization of a requested timetable, which is themain focus of this work. But, we want to mention that in a segregatedrailway system the track allocation process directly gives informationabout the infrastructure capacity. Imaging the case that two trains of a certain type, i.e., two train slots, are only in conict in one station.A potential upgrade of the capacity of that station allows for allocat-

    ing both trains. This kind of feedback to the department concerningnetwork design is very important. Even more long-term infrastructuredecisions could be evaluated by applying automatically the track allo-cation process, i.e., without full details on a coarse macroscopic levelbut with different demand expectations. Even if we did not devel-oped our models for this purpose, it is clear that suitable extensionsor simplications the other way around of our models could supportinfrastructure decisions in a quantiable way. For example major up-grades of the German railway system like the high-speed route fromErfurt to N urnberg or the extension of the main station of Stuttgartcan be evaluated from a reliable resource perspective. The billions of euros for such large projects can then be justify or sorted by reason-able quantications of the real capacity benet with respect to thegiven expected demand.

    An obvious disadvantage of the decomposition is that the in some senseoptimal solution for one step serves as xed input for the subsequentproblem. Therefore one cannot expect an overall optimal solutionfor the entire system. In the end, not even a feasible one is guaran-teed. In that case former decisions have to be changed, and a partor the complete process has to be repeated. Prominent examples are

    regional scenarios for urban public transportation where traditional se-quential approaches are not able to produce feasible schedules. Weider(2007) [213] demonstrates in case of vehicle and duty scheduling howintegrated models can cope with that and even more can increase theoverall planning efficiency. Nevertheless, hierarchic planning partitionsthe traffic planning problem into manageable tasks. Tasks lead directlyto quantiable optimization problems and can be solved by linear andinteger programming to optimality or at least with proven optimal-ity gaps. Problem standardization, automatization, organizing data,computational capabilities, mathematical modeling, and sophisticated

  • 8/12/2019 Schlechte Thomas

    51/244

    3 Network Design 17

    algorithmic approaches on a problem specic but also on a generallevel form the basis of optimization success stories in practice. As aprominent example for this we refer to the dutch railway timetable -the rst railway timetable which was almost constructed from scratch.In fact the entire planning process was decomposed and each planningproblem at Netherlands Railways (NS) was solved by the support of ex-act or heuristic mathematical approaches and sophisticated techniques,in particular linear, integer and cons