Upgrade of Regional Railway Lines

Transcription

1 Upgrade of Regional Railway Lines - An investigation of integrating Lille Syd in the TEN corridor Master Thesis june 2015 Rie Jensen, s Mai-Britt Rasmussen, s092826

2

3 Upgrade of Regional Railway Lines - An investigation of integrating Lille Syd in the TEN corridor Main Report Authors Rie Jensen, s Mai-Britt Rasmussen, s Master Thesis 60 ECTS Points Supervisors Kim Bang Salling, DTU Transport Alex Landex, Rambøll Lars Wittrup Jensen, DTU Transport

4

5 Upgrade of Regional Railway Lines Rie Jensen, s092824, June 26th 2015 Mai-Britt Rasmussen, s092826, June 26th 2015

6

7 Preface This project constitutes the Master s Thesis of Mai-Britt Rasmussen, s and Rie Jensen, s The project is conducted at the Department of Transport of the Technical University of Denmark in the spring The project accounts for 60 ECTS points. The official supervisors for the project have been Associate Professor Kim Bang Salling, Chief Consultant Alex Landex, Rambøll and Ph.D. Student Lars Wittrup Jensen. We would like to extend our gratitude to Alex Landex for providing skillful guidance throughout the completion of project. Furthermore, we would like to thank Lars Wittrup Jensen for, in addition to guidance, also providing the project with results for the capacity analysis. We would also like to thank Jesper Thorsen for guidance in the completion of the proposed optimisation model and for implementation of the model. In addition, we would like to thank every one who has contributed with material, consultantance and guidance in the completion of this project. A Special thank is pointed to the entire Railway department at Grontmij, who has provided office facilities throughout the main part of the project. The project could not have been realised without their guidance and provision of project material. Finally, we will like to thank Rail Net Denmark for letting us use the provided project material. In addition to the physical main and appendix reports at hand, a CD is included with results of the main investigations, and an electronic version of the report in its full length. v Mai-Britt Rasmussen Rie Jensen Department of Transportation June 2015

8

9 Abstract Present thesis investigates the opportunity of integrating the Lille Syd line in the Copenhagen-Fehmarn corridor. The investigation is based on common European visions for increasing the rail transport in the European transport corridors, and the development of the Danish railway network in accordance hereto. An analysis of the network capacity of the Copenhagen-Fehmarn corridor concluded, that the line was highly utilised. The method used for analysing the corridor, is a newly proposed network capacity model proposed by Lars Jensen in his Ph.D. study, in To facilitate the capacity in the corridor, different infrastructural design solutions for integration of the Lille Syd line is investigated, with the purpose of relocating traffic to the line and thereby relieving the capacity. The relocation of traffic concerns two freight train paths and two passenger trains, in relation to a proposed timetable for 2027, based on the implementation of the Fehmarn Belt link. The design solutions all include an implementation of a second track along the line, as the existing single track line indicates a high utilisation. Besides the implementation of a second track, two speed upgrade solutions of 160 km/h on the entire line and 200 km/h respectively are proposed, to avoid sacrificing the travel time for high speed passenger trains. The network capacity analysis, with the integration of the Lille Syd line and relocation of traffic, indicated a significant improvement. In the investigation of the different speed profiles for the line, results showed that profiles of higher speed led to a poorer result for the network capacity. The poorer result is explained by the increase in heterogeneity, for running both slow freight trains and fast passenger trains on the line. To determine whether the different line solutions can be realised, the existing single track line is upgraded. Clarification of the Danish railway norms forms the foundation for the upgrade and construction of the second track. Comparisons with the TSI requirements indicate, that the Danish track geometry regulations in general are more restrictive, and that Danish exceptional regulations more or less reflects the TSI requirements. To facilitate the procedure for track upgrades, an optimisation model is conducted with the purpose of optimising the cant size according to the existing track layout. Different geometrical track design solutions are investigated, with the purpose of proposing two final speed profiles with a design speed of 160 and 200 km/h respectively. In reality, it was not possible to reach the respective design speeds in all line sections, which therefore led to a reduction in travel time savings according to the initial assumptions. An evaluation of the different track design solutions is carried out, in relation to the realised travel time savings contra the estimated construction costs. The estimated construction costs for the solutions are determined through a project description of the required work to be carried out. The findings in the project description indicates, that the implementation of a second track results in significant high costs for reconstruction of vii

10 VIII road and railway bridges. Besides the consideration of bridges, high costs for implementing the second track are found according to the track superstructure and the construction of embankments. The total costs for the two investigated alternatives resulted in 1.76 billion for the 160 km/h alternative, and 2.64 billion the 200 km/h alternative. The overall benefits of integrating the Lille Syd line in the corridor include; travel time savings, relief in network capacity and introduction of improved flexibility. The travel time savings for relocating the traffic via the shorter Lille Syd line, indicated very small savings for the passenger lines, and negative savings were actually obtained for the direct train, if the line was not upgraded to 200 km/h. The results for the freight trains were somewhat different. Regardless of the two speed upgrade initiatives, freight trains obtained travel time savings of up to 5.6 minutes, due to the reduction in travelled kilometres. Ideally, the network capacity analysis should be conducted again considering the actual obtained speed profiles. Due to time constraints, this has however not been possible. Assumptions of the development in the network capacity suggests, that a poorer result will not be obtained, due to the lack of reaching the design speed of 160 and 200 km/h on all line sections. The results from the earlier conducted network analysis showed, that slower speed profiles resulted in improved capacity. The flexibility parameter is difficult to measure, however, the introduction of an alternative suggests in a larger flexibility in the network. The benefits are weighted against the costs of the different alternatives and the findings conclude, that it will not be beneficial to increase the speed on the Lille Syd line, as passenger trains will not gain any significant travel time savings despite the reduction in route length. On the contrary, freight trains gained large travel time savings. Keeping the improved network capacity in mind with the benefit of separating slow freight trains from the fast corridor, the project concludes that an implementation of a second track along the line will benefit the total network. Based on this, current project suggests that a further investigation of implementing a second track should be initiated. To facilitate the strategy for upgrading railway lines, a process flow chart is designed. The flow chart is formed on the processes carried out in present project. An improvement in the process flow is suggested, by introducing greater integration between the operational planning and the infrastructural design processes. A model to facilitate this is proposed, with the focus of the general process flow to form into an iterative process. The technical design and implementation of the model is not developed, and the focus is therefore merely on the improved process flow.

11 Contents 1 Introduction Purpose Delimitation Reading Guide Background The Political Background The Trans-European Transport Network Green Transport Policy Togfonden DK The Traffical Background Freight Transport in Denmark Passenger Transport in Denmark Future transport needs and conflicts The Lille Syd Corridor Station Potential Interlinked Operation with the Local Line Summary of Background Capacity Theory The UIC 406 Method The UIC 406 Method Carried Out Congestion Evaluating the UIC 406 Method New Method for Assessment of Capacity Consumption in Railway Network Model Framework Stochastic Extension Analysing Model Results Summary of Capacity Theory Capacity Analysis Input for Initial Capacity Analysis Operation Plan Running Time Calculation Train Type Speed Profiles Running Time Supplement Stop Pattern Graphical Timetable ix

12 X CONTENTS Headway Capacity Analysis on Lille Syd Capacity Analysis on the Main Line Passing Ringsted The Transformation of the Network Network Input Results Edge Capacity Networks Capacity Validating Results Relocation of Traffic to the Lille Syd Line Revised Model Input Speed Profile Stopping Pattern Graphical Timetable Graphical Representation of Network Results Edge Capacity Network Capacity Potential Travel Time Savings Summary of Capacity Analysis Norm Foundation Standards The Danish National Railway Norms The Technical Specifications for Interoperability Application of the Railway Norms Summary of Standards with Respect to the Lille Syd Line Track Geometry Horizontal Curves Cant Cant Deficiency and Excess Requirement for Cant Size Requirements in Turnouts Transition Curves Abrupt Change of Cant Deficiency Straight Track Between Curves Longitudinal Profile Gradients Vertical Curves Summary of Track Geometry with Respect to the Lille Syd Line Ballast Profile Components within the Ballast Profile Upgrading New track along existing Summary of Ballast Profile with Respect to the Lille Syd Line Track Center Distance Fouling points Summary of Track Center Distance with Respect to the Lille Syd Line Structure Gauge Danish Requirements for Structure Gauge

13 CHAPTER 0 XI International Requirements for Structure Load Summary of Structure Gauge with Respect to the Lille Syd Line Platforms Danish Requirements for Platforms TSI Requirements for Platforms Safety Zones and Open Spaces Summary of Platforms with Respect to the Lille Syd line Level Crossings Danish Regulations for Level Crossings Risk Analysis Concerning Increased Speed in Level Crossings Allowed Speed in Level Crossings in Germany and Sweden International rules for level crossings Road Restriction Time - Another Aspect Summary of Level Crossings with Respect to the Lille Syd Line Norm Summary Optimisation Model Objective Model Framework Model Input Objective Function Constraints Model Implementation Field of Application Model Visions Summary Track Geometry Solution General Considerations within Speed Upgrades Strategy for Upgrading the Speed Profiles The Existing Track Geometry Upgrade of Existing Track to 160 km/h Upgrade to 160 km/h with Minimum Effort Upgrade of Existing Track to 200 km/h Upgrade to 200 km/h with Minimum Effort Comparison of Danish and EN Exceptional Regulations Summary of Track Geometry Solution Project Description Financial Setup and Budgeting Assumptions for the Project Description Traffic Assumptions Delimitations, Interfaces and Assumptions Inspiration and Procedure for Upgrading Projects Overview of the Disciplines Track Track Layout Speed Profile and Curvature Superstructure Track Construction - Upgrade

14 XII CONTENTS Track Construction - New Track Ballast Profile Earth Work Soil Handling - New Track Soil Handling - Upgrade Drainage Bridges and Constructions Bridges Platform Crossings Electrification System Electrification of the Existing Track in Relation to the Speed Upgrade Electrification of the New Track Power Supply Interlocking and Remote Control IT, Tele and Transmission Systems Constructions Platform Design for Speed Upgrades Areas Upgrade of Existing and Construction of Second Track to 160 km/h Upgrade and Construction of Second Track to 200 km/h Electrical Restrictive Covenants Forestry Additional Considerations Cross-disciplinary Costs Summary of Project Description Construction Cost and Project Evaluation Construction Cost Unit Cost Correction Supplement Cost Estimates Risk Plan Project Evaluation Recommendation Overall Process Flow Process Flow in Present Project Optimised Process Flow Discussion Conclusion 165 Appendices 174 A Learning Objectives B C Running Time Calculation Speed Profile

15 CHAPTER 0 XIII D Graphical Timetables E F Mail Correspondance UIC 406 Method on the Lille Syd Line G Edge Capacity H Matemathical Model I J Implementation of Mathematical Model Track Geometry Solutions K Earth Work Estimate L Bridge Estimate M Construction Cost Estimate N Hazard Identification O Flow Charts P Drawings

16

17 Chapter 1 Introduction In the recent decades, the European and Danish focus has been on paving the way for a significant shift in transport modes from road to rail. In 1992, the first European White Paper was formulated with the purpose of integrating national transport networks, to compose a Trans-European transport network. Visions of increasing rail transport has led to the establishment of key corridors, to combine the respective membership countries. Major infrastructural initiatives are introduced, and the Fehmarn Belt fixed link plays a significant role for connecting the continental Europe with Scandinavia. With the integration of the Danish route from Fehmarn to Copenhagen follows great visions of increasing both freight and passenger transport. The European visions are integrated in the Danish transport plan, and initiatives on upgrading and constructing new lines, are realised for the network to become integrated in the TEN corridor. Keeping this in mind, the present project focuses on introducing an alternative, shorter route parallel to the main TEN corridor. The integration of the Lille Syd line and relocation of traffic in the corridor, will possibly produce travel time savings, relief in capacity consumption on the main line, as well as contributing to a more flexible network. To support the relocation of traffic, the integration of the line requires an extension to double track, and a possible speed upgrade, to complement the high speeds on the main corridor. Different line upgrades with a simultaneous extension to double track on the Lille Syd line are investigated, to evaluate whether or not the assumed findings can be confirmed. The capacity consumption of the main line, with and without, the inclusion of the Lille Syd line is investigated by use of a new model composed by Lars Wittrup Jensen. The model evaluates the total network capacity, and can therefore clarify the impact of including the line. The different alternatives of upgrading the line are hereby investigated in a network capacity analysis. Furthermore, the alternatives are investigated by conducting an overall track design solution, and from this estimating the construction costs. The track design is determined according to the Danish railway norms, and parallels are drawn to the TSI requirements, to establish a sound norm foundation. To ease the geometrical track design process, an optimisation model is proposed for determining the optimised cant in relation to the possible highest speed profile. Based on the proposed geometrical track design solutions, a project description of the required construction work is conducted, to estimate the construction cost of the different solutions. These are held against the benefits from the 1

18 2 INTRODUCTION network capacity analysis and the obtained travel time savings, and an evaluation of the impacts of integrating the Lille Syd line in the corridor are made. An evaluation of the processes in the project leads to a preparation of a flow chart, to clarify and propose improvements in the general process flow for upgrading railway lines. The focus in the optimised flow is grater interconnection between the operational planning and infrastructural design phases. By implementing iterative processes for investigating different project solutions, early intermediate evaluations can be obtained with the purpose of complementing the screening phase level. 1.1 Purpose Based on the visions on introducing a shift in transport modes to facilitate rail transport and develop the Danish railway network, the purpose with present thesis is to investigate the following; The possibility of integrating the Lille Syd line in the TEN corridor from Copenhagen to Fehmarn, by determining the possible gains according to travel time savings, relief in capacity utilisation on the main corridor, and improved flexibility The impact in the network capacity on the Copenhagen-Fehmarn corridor by introducing the Lille Syd line and evaluating the capacity according to a newly proposed model Different infrastructure solutions for the future design of the Lille Syd line to accommodate visions for improved rail transport The gains compared to an estimated construction cost for the proposed solutions, and clarify whether a solution should be further investigated The strategy for upgrading railway lines in relation to introducing and improvement in process flow To investigate the above initiatives, the Danish national railway norms are clarified and compared with the TSI regulations to determine a norm foundation for the different project proposals. 1.2 Delimitation A set of delimiting initiatives are introduced, to bring the process of upgrading and integrating the Lille Syd line in the overall corridor into focus. The network capacity analysis conducted in the initial phase, solely considers the main line Copenhagen-Ringsted-Fehmarn and the integration of the Lille Syd line. No consideration is taken towards the remaining railway network. On the main line, it is expected that a fly-over in Ringsted is implemented, leaving no conflicting routes in the network capacity analysis. In relation to the proposed infrastructure solutions, the implemented second track along the Lille Syd line is coupled to the existing siding track with no consideration of the track geometry in these. Furthermore, no consideration is taken towards the integration of the Local line at Køge station, and either Køge or Næstved station are

19 CHAPTER 1 3 investigated. The speed upgrade and track design of the existing Lille Syd line fully reflects the design of the second track. This results in two identical speed profiles regardless of new or existing track. The implementation of the second track is carried out on an overall level, hence little attention is brought to the exact location and the influence on the local environments. The proposed infrastructure solutions are based on a screening phase level, hence the cost evaluation of these are based on extremely rough estimates. No projection of the infrastructure is carried out, and the proposed solutions are therefore based on a great deal of assumptions, to enable the execution of the project. These are presented throughout the report, and discussed in relation to the impacts on the results. 1.3 Reading Guide The report is structured in 6 sections according to Figure 1.1. The project is initialised with a presentation of the background and visions for the Danish railway network, in connection with a clarification of the overall purpose. The second part of the project investigates the network capacity on the main Copenhagen- Fehmarn line, by using a newly proposed capacity model. The impacts integrating an extended and upgraded Lille Syd line and relocating traffic hereto are analysed. The technical background establishes the norm foundation for upgrading and extending the Lille Syd line. In relation hereto, an optimisation model is developed to facilitate the work process for upgrading the speed profile. Based on the technical background, several proposed solutions for upgrading and expanding the line are investigated. The solutions are evaluated according to travel time savings, relief in network capacity and an estimated construction cost. Based on the evaluation, a project solution is proposed. An evaluation of the processes covered throughout project is presented in a process flow. The process flow is reviewed and a proposed improvement of the flow is presented. The final section concludes on the project and discusses the assumptions and findings of the analysis. Background Initial Capacity Analysis Technical Background Proposed Solutions Process Evaluation Rounding Introduction Background Capacity Theory Capacity Analysis Norm Foundation Optimisation Model Track Geometry Solution Project Description Construction Cost and Project Evaluation Process Flow Discussion Conclusion Figure 1.1: Overview of the structure in the Report

20 4 INTRODUCTION

21 Chapter 2 Background Present chapter covers a presentation and the motivation for recent European and Danish infrastructure initiatives. The chapter seeks to clarify current and future demand in passenger and freight transport as well as possible conflicts and bottlenecks. The route Copenhagen-Fehmarn is in focus due to it being one of the two passages for international trains between Sweden and Germany, the other crosses the Great Belt. The Lille Syd line is presented with the purpose of introducing an alternative, shorter route for both passenger and freight transport. Freight transport is the main focus as the demand for freight train paths is expected to increase significantly with the opening of the Fehmarn fixed link, however, an additional increase in passenger trains between Denmark and Germany is also expected (Rail Net Denmark, 2013b). The potential of running additional traffic on the Lille Syd line is investigated throughout the report. The chapter is divided in three sections concerning the political background, the traffic-related background and a presentation of the Lille Syd line. 2.1 The Political Background The European infrastructure is undergoing numerous of changes to establish a common transport network. The White Paper formulated by the European Commission states objectives for the development. Nine interoperable transport corridors will connect 28 member states in the European Union. The Scandinavian-Mediterranean corridor runs through Denmark and ensures an interoperable passage from Palermo in Italy to Stockholm in Sweden (Commission, 2014). The White Paper does not form the exact requirements for the development of the Danish network, however the main goals and initiatives must be considered in a strategic plan as means for reaching the goals are considered carried out. In continuation hereof, the Danish transport policy is formulated in accordance with the European goals The Trans-European Transport Network In 1992, the first White Paper was published by the European Commission to develop a common European transport policy. The main objective was to pave the way for an open transport market and relieve the difficulties of cross border transport. The idea of a common Trans-European Transport Network integrating national transport networks throughout Europe was born (European Commission, 2007). By the time of the new millennium, the goal was more or less achieved for the road traffic, but the railway sector was still characterised by national systems making it difficult to move across boarders. 5

22 6 BACKGROUND Then, in 2001, a new White Paper on transport was formulated treating issues such as road congestion and environmental aspects due to the increase in passenger and freight transport. Initiatives on shifting the share of road traffic to rail was launched as well as a strategy for making a more sustainable transport network (European Commission, 2011). In 2011, an evaluation of the accomplishments since 2001 was carried out; the market opening of the road transport had been successful, however the rail transport had not followed suit and more important, the transport system was not sustainable. Based on this, the European Commission formulated a long-term strategy for a common European transport sector to meet future transport demands whilst promoting sustainability. Improvements of the European railway network to enhance the attractiveness of rail transport over road was an overall objective. A goal of having more than 30 % of medium passenger trips and 30 % of all truck transport over 300 km shifted to rail by 2030 as well as reaching a shift in freight transport from road to rail of 50 % by 2050 was formulated (European Commission, 2011). The main mean of achieving the long-term strategy is the introduction of an interoperable TEN core network that will link the major cities in the East, West, North and South of Europe together. 30 key priority projects are formulated, of which 18 solely focus on the development of the railway infrastructure, see figure 2.1 below. The railway network should be founded on common European norms and standards and the implementation of a common signalling programme will establish interoperability throughout Europe. The major goals are to remove bottlenecks, promote interoperability and finally construct missing cross-boarder connections. The core network will ensure smother and quicker travels, and will contribute to more safe and less congested routes. Furthermore, there is focus on enhancing the competition within the railway sector by opening the market for various railway operators. In addition, another important interest is the promotion of international rail freight corridors. By giving high priority to freight transport in accordance with market needs, within freight corridors, a transfer of freight from road to rail will be accomplished (European Commission, 2013). Figure 2.1 illustrates the 9 corridors within the core network. In the central part of Europe several corridors run parallel contributing to flexibility in the network. One of the key corridors, the pink corridor from Stockholm to Palermo, runs through Denmark via the existing route through Jutland and crossing the Great Belt and via the route across Fehmarn Belt. A key project within the corridor is therefore the implementation of a fixed link across Fehmarn Belt. The corridor considers both passenger and freight transport and is known as the Scandinavian-Mediterranean corridor. In 2013, two regulations formulated by the European Union (1315/2013 and 1316/2013) related to the TEN railway network come into effect. The regulations state the priorities for the development of the TEN network. This results in the fact that all requirements related to the core network must be complied with by the end of 2030, while the requirements dealing with the remaining network must be complied with by 2050 (Rail Net Denmark, 2014b) Green Transport Policy The European visions was set off into the Danish transport policy and in 2009, the largest parties in Denmark entered into an agreement for developing a green transport policy.

23 CHAPTER 2 7 Figure 2.1: The 9 key corridors of the Trans-European Transport Network, the Scandinavian- Mediterranean corridor is highlighted in pink (European Commission, 2014) The agreement known as Green Transport Policy was made to ensure shorter travel times between the largest cities in Denmark by upgrading the railway network. The One-Hour Model, ensuring a travel time of one hour between the largest cities, and the first high speed line Copenhagen-Ringsted was presented. Furthermore, the transport policy will ensure the railway of being independent of fossil fuels and at the same time reduce the CO 2 emission (Ministry of Transport, 2009). An implementation of the signalling system ERTMS level 2 will additionally harmonise the Danish and the European railway network and countenance of later electrification the network Togfonden DK Another agreement was made in January 2014, between the Government, the Red-Green Alliance and the Danish People s Party. This agreement known as Togfonden DK gives financial assurance for improving the public transport network by increasing the taxation of oil companies operating in the Danish part of the North Sea billions is set aside to modernise the railway and a number of key objectives are formulated in the agreement. A main objective is the implementation of the One-Hour Model in addition to initiatives of speed upgrades of regional lines, electrification of the main network and several new railway lines. Another focus in the agreement is the increase in freight transport. The implementation of the signalling program and the electrification of the main network will result in a higher railway capacity benefiting the accessibility of the freight transport. By establishing a fixed link via Fehmarn Belt, all transit railway freight transport between Sweden and Germany is planned to be redirected via this route. Also, a number of the inter-

24 8 BACKGROUND national passenger trains are assumed to pass Denmark via the Fehmarn Belt link. The fixed link will increase the competitive advantage for transit freight transport passing Denmark compared to road traffic and the existing ferries. To ensure the maximum benefit for the freight transport an analysis concerning additional bottlenecks is carried out as well as an analysis of the potential for increasing freight transport with the associating initiatives. The analysis is assumed to be completed in 2015 (Ministry of Transport, 2014). 2.2 The Traffical Background As described, the Danish railway network is currently being developed to become a part of the TEN-network. In continuation of the European goal of connecting Europe and the Danish goals of renewing the Danish railway system, ensure shorter travel times between key cities and increasing freight transport, the Danish Transport Authority has conducted a traffic plan for the years 2012 to 2027, in relation to future freight and passenger transport in Denmark Freight Transport in Denmark Freight transport by train in Denmark has decreased up to 2010 and is now increasing. It is especially transit freight trains and an increase in number of operators causing the development. Induced traffic is expected with the opening of the Scandinavian- Mediterranean corridor due to tendering of attractive paths, modernisation of terminals, and initiatives forming increased capacity. Contributing to this is the opening of the Fehmarn fixed link, the new signalling system, and the high speed line Copenhagen- Ringsted. The route via Fehmarn Belt is 160 km shorter than the route via the Great Belt, and characteristics such as congestion and energy costs will play an impact. (Rail Net Denmark, 2012b) In 2010, a capacity analysis of the Danish railway network was carried out, declaring the central lines in Copenhagen congested (Rail Net Denmark, 2010). It is a well-known fact that the capacity utilisation of the Danish railway network is high, and the network is one of the most highly utilised networks in Europe (Landex, 2008). Currently 1-2 freight train paths (per hour and in each direction) are reserved for freight trains via the Great Belt. In 2012, Rail Net Denmark presented the report Increased freight transport formulating the obstacles and challenges, the network faces before an increase in freight transport can take place. In accordance with the political goals of increasing freight transport, Rail Net Denmark has formulated a strategy providing an infrastructure with the potential of doubling the passengers and freight transport in From 2027, three freight paths/hour/direction through Denmark via the Scandinavian-Mediterranean corridor will be available and maybe four outside rush hours (Rail Net Denmark, 2012b). Two are reserved for the route via Fehmarn Belt, whereas one is reserved via the Great Belt link, see figure 2.2 below. However, it is questionable if this is sufficient to handle the future demand for freight paths. ERNCF (Regulation for a European Rail Network for Competitive Freight) requires 4 freight train paths through Denmark (Rail Net Denmark, 2012b). The European Commission s White Paper from 2011 presents a goal of having 159 freight trains running through Denmark via Fehmarn and the Great Belt in Based on this, Rail Net Denmark has performed a calculation of the demand for freight train paths

25 CHAPTER 2 9 Figure 2.2: Rail corridors in Denmark, illustrated by the authors to comply with the European goals. All in all, the analysis concludes that the demand for freight trains daily in both directions is 164. The practical capacity via Fehmarn is 78 trains daily in both directions and 40 trains via the Great Belt, resulting in a shortage of 46 trains on a daily basis running through Denmark. In addition hereto, Rail Net Denmark presents a plan of action in the report suggesting, that an increase in freight capacity can be derived from financially controlling the freight train path demand according to time, as well as introducing infrastructural initiatives on capacity increase. In relation to this, allowing longer freight train lengths (up to 1000 m), larger axle loads and more efficient terminals will contribute to this. (Rail Net Denmark, 2012b) The question regarding expansion of freight terminals has surfaced the railway sector in recent years. In recent years rail freight companies have experienced the closure of sidings and freight tracks, however this development is expected to change (Østergaard, 2012). Rail Net Denmark owns three international combi-terminals in Denmark; one in Padborg, Taulov and Høje Taastrup. Besides these, medium-sized harbour terminals and other medium-sized terminals are located in Aarhus, Esbjerg, Aalborg and Køge (Rail Net Denmark, 2012b). In 2009, the political initiative Green Transport Policy contained initiatives on strengthening the freight transport by introducing new terminals in Esbjerg and Køge and reopening and renewing freight tracks. The terminal in Esbjerg became a reality in 2015, however, the new terminal in Køge is not implemented (Rail Net Denmark, 2012b). Since the Green Transport Policy, the majority of the initiatives have been abolished to big regret for especially the national freight transport (Kien, 2013). In the decision process concerning the new Copenhagen-Ringsted line, a new combi-freight terminal in Køge was on the drawing board. The Danish Transport Authority conducted a market analysis

26 10 BACKGROUND of the potential for constructing the terminal, however, the conclusion came out negative (Danish Transport Authority, 2010c). In relation to this, it was stated that the increase in competitive performance of the railway might possibly result in the need for expansion of terminals in Zealand. An expansion will either concern an upgrade of the terminal in Høje Taastrup or a new terminal in Køge (Danish Transport Authority, 2014a). Contributing to the Køge alternative are factors such as Køge harbour and the location of the Scandinavian Transport Center close to the railway Passenger Transport in Denmark In 2012, the Danish Traffic Authority published a traffic plan for the Danish railway for the years 2012 to The traffic plan considers the enacted railway projects and gives an overview of the possible traffic and passenger related effects. The Green Transport Policy from 2009 formulated the goal of increasing passenger kilometres from 6.5 to 13 billion per year. To reach the goal, initiatives on reduced travel times, better comfort and increased capacity are crucial. The Fehmarn Belt fixed link and the additional upgrade and electrification of the line Ringsted-Fehmarn will result in significant travel time savings. The fixed link will ensure an increase from 6 to 20 passenger trains running daily in each direction between Denmark and Germany in These will run both via Fehmarn and the Great Belt. The travel time between Hamburg and Copenhagen will be reduced by 1 hour and 40 min, a reduction of 37 %. Travel times between Copenhagen and Nykøbing Falster and Næstved is also reduced significantly. (Rail Net Denmark, 2013b) The shorter travel times and higher frequency will lead to a significant increase in passenger transport. Although the fixed link is planned to be implemented by 2021, it is still unknown exactly when the planned German double track from Fehmarn to Bad Schwartau will be implemented. However, the treaty between Denmark and Germany states that the double track is to be implemented no later than 2028 (Larsen, 2015). The capacity on the existing single track line limits the number of daily trains to 20 passenger trains and 24 freight trains in each direction. Currently passengers crosses the Danish-German border by train annually. 80 % of the passengers travels to/from destinations in Jutland and 20 % travels to/from Funen and Zealand. The Danish and German initiatives will shorten the travel time between Hamburg and Copenhagen, which will entail an increased attractiveness compared to flight travel (Danish Transport Authority, 2013) Future transport needs and conflicts In a paper from 2009, assistant professor Alex Landex discusses the conflicts for the existing freight corridor in Denmark. The lack of capacity to fulfil the demand for running freight trains was already in 2008 a well-known issue. Landex points to the future major bottlenecks of Kastrup and the Drogden tunnel connecting Amager and Peberholm, which play a major role in the current corridor through Denmark, see Figure 2.3. In Figure 2.3 red lines illustrate high capacity utilisation. With the implementation of the Fehmarn Belt link future potential bottlenecks include Ringsted and Ny Ellebjerg (Landex, 2009b). A flyover at Ny Ellebjerg will eliminate the crossing of freight and passenger

27 CHAPTER 2 11 trains between the Øresund line and the new line Copenhagen-Ringsted, and funding for the initiative is set aside in the Togfonden agreement (Marfelt, 2014). Figure 2.3: Kastrup and the Drogden tunnel connecting Peberholm and Amager are among others highly utilised in the capital region (ngselskommissionen, 2013) An implementation of a flyover in Ringsted is not yet decided, as doubts about the location have postponed the law enactment. The implementation of the flyover is essential to eliminate the crossing of the two lines towards Fehmarn and the Great Belt, respectively. Another aspect in relation hereto is the fact that fast passenger trains towards the Great Belt are forced to decrease the speed in Ringsted to 120 km/h, which is conflicting with the One-Hour Model. It can, however, be assumed that additional upgrades on the line Copenhagen-Odense possibly will be sufficient for reaching this goal. Considering the expected increase in number of trains in the corridor Copenhagen-Ringsted-Fehmarn, a flyover in Ringsted might be difficult to sacrifice unless part of the traffic is relocated to alternate routes. In the article Landex argues in relation to the future capacity issues for the opportunity of running freight trains along the Lille Syd line. For this to succeed, factors such as electrification and exchange of the signalling system on the line are required, initiatives which currently are being implemented (Landex, 2009b). Danish Professor in Traffic Modelling Otto Anker Nielsen assesses the expediency of reserving land use in relation to future infrastructure initiatives in a minute from Based on a Danish-Swedish project known as IBU-Øresund long-term infrastructure planning in the Oresund region is considered. One of the main objectives is to develop a Danish-Swedish railway plan to ensure a doubling in the capacity between Stockholm and Hamburg (Nielsen, 2011). The project deals with several rail related transport issues and considers initiatives treating these. A suggestion on upgrading the Lille Syd line in relation to the electrification contributes to the suggestion of running freight trains in the corridor. An argument for the upgrade is that freight trains will save up to 5 minutes due to the fact that the line is close to 10 kilometres shorter than via Ringsted. Another argument is the capacity relieve

28 12 BACKGROUND of the new line Copenhagen-Ringsted. By running freight trains along the Lille Syd corridor the share in infrastructure with the fast passenger trains will only include the stretch from Køge Nord to Copenhagen. By implementing shunts in Køge and Næstved it will be possible to reach a speed of 250 km/h in these locations. Due to a less complicated track geometry along the Lille Syd line, an upgrade to 200 km/h seems possible. This will enable fast passenger trains between Hamburg and Copenhagen to use the line and at the same time create a significant travel time saving for the regional lines (Nielsen, 2011). 2.3 The Lille Syd Corridor The single track line Roskilde-Køge-Næstved known as Lille Syd opened in 1870 and has a total length of 61 kilometres. The line has a maximum speed of 120 km/h and 13 associated stations. The daily operation consists of two trains in each direction every hour. With the opening of the new Copenhagen-Ringsted line a new station Køge Nord is being implemented as a prolongation of the Lille Syd line see figure 2.4. This will divide the original line in two parts; Roskilde-Køge and Køge-Næstved, and ensure a direct connection between Næstved and Copenhagen. In relation to this, initiatives on upgrading the southern part of the Lille Syd line has been enacted. Figure 2.4: The Copenhagen-Ringsted line (green) and the Lille Syd line (red)(møller, 2012) The southern part of the Lille Syd line, from now on referred to as the Lille Syd line, is approximately 44 kilometres. The line has nine stations with Køge Nord and Næstved as terminal stations. In 2010, the annual number of passengers on the line was 0.7 billion, which is expected to increase to 1.1 billion in The number of trains running per hour in each direction is 2 and no further trains are expected. The line is highly utilised however, according to the traffic plan , there is little need for an upgrade of the line, due to the fact that no additional trains are expected (Danish Transport Authority, 2013). Despite the immediate lack of need for further improvement, see Figure 2.5, initiatives on upgrading the line have started. In 2009, the law concerning the Signalling Programme

29 CHAPTER 2 13 Figure 2.5: Capacity utilisation of Danish railway lines at Zealand, Lille Syd is enhanced by the blue ring (Danish Transport Authority, 2013) was enacted and in 2016 it will be put into operation on the Lille Syd line, being one of the first lines in Denmark operating with the system. In 2012, COWI conducted a screening on whether it will be profitable to upgrade the regional lines in Denmark from 120 km/h to 160 km/h. The findings of the analysis were positive for the Lille Syd line, and by upgrading the line to 160 km/h, a positive socio-economic benefit can be achieved (COWI, 2012). In 2013, the Ministry of Transport enacted a law concerning electrification of the line from Køge to Næstved. The electrification will be put into operation by 2018, and results in a travel time saving of 6 minutes between Køge and Næstved. In continuation hereof, the political agreement of Togfonden DK encompassed a speed upgrade of the line allowing a speed of 160 km/h. The speed upgrade is carried out simultaneously with the electrification and is expected to finish in 2018 (Rail Net Denmark, 2013d). Besides the future initiatives, a track renewal project on the Lille Syd line carried out in The renewal project concerns the exchange of rails and sleepers as well as ballast cleaning. The renewal will ensure a better punctuality on the line due to fewer speed restrictions (Rail Net Denmark, 2013a). In connection with the electrification of the line Tommy O. Jensen, Atkins, explains the benefits of the electrification in an article from Jensen points towards a future time table where trains between Nykøbing Falster and Copenhagen will run on the Lille Syd line. In addition to this, Jensen addresses the opportunity of running freight trains along the Lille Syd line in case of maintenance work at Ringsted station (Møller, 2012). The idea of using the Lille Syd line as an alternate route between Næstved and Copenhagen is not new. In 2009, a large track work was initiated on the South Line from Ring-

30 14 BACKGROUND sted to Nykøbing Falster alternately blocking the tracks. In rush hours trains opposite the main passenger flow were redirected along the Lille Syd line. All trains, including international trains, between Næstved and Ringsted were redirected via the Lille Syd line outside rush hours. In the most busiest hours of the day seven additional trains were inserted on the line (DSB Kommunikation og Banedanmark Kommunikation, 2009). Also, the corridor along the Lille Syd line has earlier been used for freight transport. In 2007 the traffic committee within the national parliament stated that 13 % of the total freight transport between Zealand and Jutland/Germany ran on the Lille Syd line from Ringsted to Køge via Næstved (Folketingets Trafikudvalg, 2007) Station Potential In 2008, DSB performed a counting of passengers in Eastern Denmark. The counting was performed during one day and carried out to map the passengers travel pattern. Based on the counting Holme-Olstrup and Tureby Station were ranked as being in the top five stations having the lowest number of passengers, counting 280 and 161 respectively (DSB, 2008). Table 2.1: Daily passenger arrivals and departures on the Lille Syd line Stations/year Køge Nord Ølby Køge Herfølge Tureby Haslev Holme-Olstrup Næstved Nord Næstved In a screening analysis from 2008, the Danish Transport Authority examined the potential of opening new stations and closing less used stations on the state railway network. In general stations having less than 100 departures and arrivals were examined with the potential of getting closed. The assessment of whether to close a station is based on balancing the advantage that passengers on the particular station experience contra the disadvantage for the remaining passengers due to longer travel times. 38 stations were examined,of which none of them are located on Zealand (Danish Transport Authority, 2008). In general the number of passengers in Eastern Denmark is much higher than in Western Denmark, which also explains the higher number of passengers using the minor stations. In an analysis of the station structure from 2014 the Danish Transport Authority suggests that further attention is directed towards opening a new station in Darup 2 km south of Roskilde and in Hastrup 2 km south of Køge, due to possitive net time profit. The stations are estimated to generate approximately 550 and 450 daily trips, hence more than

31 CHAPTER 2 15 twice the number in Holme-Olstrup. Due to the relatively low number of passengers on the line, the benefits for the new passengers exceed the drawbacks for the existing passengers. In both cases, the municipalities are planning to develop the town areas and the report concludes that a further analysis of the potential is to be initiated (Danish Transport Authority, 2014b). Addressing the potential for opening and closing stations is outside the scope of this analysis, however the aspect is kept in mind during the analysis Interlinked Operation with the Local Line In 2013 the Ministry of Transport initiated a study on how the regional network can be best utilised. The study concluded that the best utilisation is achieved by extending the traffic on the local line Østbanen to continue to Køge and Roskilde. This will result in a direct connection between Rødvig, Faxe Ladeplads and Roskilde. An interlinked operation is assumed to become effective in 2018 when the new line Copenhagen-Ringsted is put into service (Ministry of Transport, 2015). Figure 2.6: Interlinked operation of the line Roskilde-Køge and the local line towards Rødvig and Fakse Ladeplads (Danish Transport Authority, 2013) 2.4 Summary of Background The preceding chapter presents the relevant current developments the Danish railway infrastructure is undergoing. The Danish political decisions are closely related to the European goals of constructing a common interoperable infrastructure throughout Europe. The goal of increasing the number of, especially freight, train paths and shifting a major share of both the passenger and freight transport to rail requires renewal and construction of new lines. Especially the demand for freight train paths is increased with the implementation of the Fehmarn fixed link, and an analysis of the available freight train paths compared to the European Commission s demand through Denmark concludes a shortage of capacity for 46 trains. In relation to the passage Copenhagen-Ringsted-Fehmarn an additional route via the Lille Syd line is suggested. The Lille Syd line will, with the implementation of the new Copenhagen-Ringsted line, be directly connected to the TEN

32 16 BACKGROUND passage through Denmark and is possible to function as a shorter alternative route. The Lille Syd line is currently undergoing a speed upgrade in connection with the implementation of the electrification and signalling programme however, the traffic on the line is not expected to increase according to the traffic plan Suggestions of running additional traffic on the line as an alternative route between Copenhagen and Nykøbing Falster is expressed by both Professor in Traffic Modelling Otto A. Nielsen and electrification expert Tommy O. Jensen, Atkins. This is investigated further in the following chapters in relation to the capacity consumption of the corridor between Copenhagen- Ringsted-Fehmarn.

33 Chapter 3 Capacity Theory Based on the findings in chapter 2, the capacity utilisation of Copenhagen-Ringsted- Fehmarn and the Lille Syd line is investigated. Present chapter describes the three different types of methods used for determining the capacity. The capacity utilisation of the two line sections are used to determine, whether relocating traffic to the Lille Syd line will relieve the consumption on the main line. The methods include; the UIC 406 method (International Union of Railways, 2004), an extension of UIC 406 method used on single track lines proposed by Landex (Landex, 2009a), and a new model assessing the capacity consumption of an entire network proposed by Jensen (Jensen, L. W., Landex, A. and Nielsen, 2015). The UIC 406 method is currently valid for determining the capacity utilisation of European railway lines. Landex discusses the method and proposes an extension of adding dummy trains to eliminate a paradox within the method. Finally, Jensen proposes a new method for evaluating the capacity consumption of an entire network by introducing a model, that can be used in early strategic planning, when uncertainty within the planned timetable is large. The chapter is divided in three sections and presents the capacity theory used in the further analysis of the Copenhagen-Fehmarn corridor. 3.1 The UIC 406 Method In 2004 the International Union of Railways (UIC) published a leaflet presenting a general method for calculating the capacity consumption of railway lines; the UIC 406 leaflet concerning capacity. The leaflet states that Railway capacity in itself does not exist as it depends on the utilization of the infrastructure according the timetable. The subsequent sections are based on the UIC 406 leaflet, unless else is stated (International Union of Railways, 2004). Capacity is defined as an interdependence between the following parameters: The number of trains running on the line The average speed of the trains The stability of the timetable. In order to ensure that smaller delays in the timetable do not scatter to the rest of the trains, buffer times are added to the timetable The heterogeneity of the timetable reflecting the type of train runs according to average speed and stop pattern. A difference in speed and stop pattern for two trains 17

34 18 CAPACITY THEORY running simultaneously on the same track will result in greater track occupation than two trains running with the same running time Figure 3.1: Capacity Balance (International Union of Railways, 2004) Figure 3.1 explains the balance of the capacity. Taking the example of a railway line operating mixed-traffic, the heterogeneity is high due to the mix of passenger and freight transport and the relatively high average speed. The high heterogeneity causes a respectively lower amount of train runs and therefore adds to the lower stability, as delays will scatter throughout the network. This is true, as delayed trains are not easily cancelled if the frequency is low. In Denmark the current operation is characterized as being more homogeneous compared to other European countries, as no high speed lines and only limited freight trains are operated. Freight trains run with an average lower speed than the regional trains, but the frequent stopping pattern contributes to regional trains having longer running times. The ongoing speed upgrades of Danish lines and the fact that high speed and intercity trains are not separated from freight and regional trains contribute to heterogeneous operation, which is expected to increase with the new high speed railway line from Copenhagen to Ringsted. The aim of the UIC 406 method is to provide railway infrastructure managers with a common method on how to determine the capacity consumption of a railway line based on common international criteria and methodologies. The purpose is to develop a general way of evaluating and comparing railway lines in the European network (International Union of Railways, 2004). The UIC 406 leaflet presents different views on capacity according to market groups; customer needs, infrastructure and timetable planning and operation. These groups have different views on how capacity is reflected in their requirements, as customers are interested in short journeys and care less for maintenance strategies represented by the infrastructure planning. The timetable planners are interested in time supplements and well-planned connections at stations whereas the operators focus on different types of delays. Even though the four groups represent different actors in the industry some views are shared, but the capacity-requirement result also differs. Based on the views of the different market groups and capacity-relevant constraints such as the timetable structure, the UIC 406 leaflet defines capacity as the total number of possible paths in a defined time window... (International Union of Railways, 2004).

35 CHAPTER The UIC 406 Method Carried Out The UIC 406 method is based on a pre-constructed timetable, which graphically is compressed within a specific line section in a determined time interval. The capacity consumption is derived from the infrastructure occupation and a buffer time added, to account for stabilisation of the timetable and maintenance requirements. The infrastructure occupation is derived from a compression of the timetable considering the minimum headway times related to the infrastructure specifics of interlocking systems and train characteristics. The network is divided into smaller line sections at junctions, crossing stations, etc., and the capacity consumption is then derived from the compressed timetable divided by the cycle time (Landex, 2007). The compressed timetable is evaluated for a peak hour in order to reflect the most busy time of the day. In some cases the limit of capacity consumption is not obtained, and excess capacity in the network is left. It is not always possible to utilise unused capacity for operating additional trains. The possibility of operating additional trains depends on the network and desired level of punctuality. If excess capacity can not be used for allowing additional train paths, it will be defined as lost capacity. The capacity consumption is determined on every separate line section on the line or route, and the highest value determines the overall consumption of the line. The compression of the timetable is carried out by only allowing a minimum headway time between all single train paths e.g. the separate line sections, see Figure 3.2. The length of the line sections may vary from network to network; however, it is recommended to divide the sections between two neighbouring stations without the possibility for crossings or overtaking. The running times, overtaking, crossings or stopping times must not be changed during the compression. Figure 3.2: Compression of a graphical timetable (International Union of Railways, 2004) A train path usually consists of several block sections; hence, the block occupation time is decisive. The block occupation time depends on the signalling system, the block systems and the safety technology. The occupation time reflects the actual running time of the train according to train characteristics and the infrastructure. Time supplements are used for smaller irregularities in the operation, scheduled waiting times to compensate for regular-interval timetables and finally time for clearing and releasing the route, see Figure 3.3.

36 20 CAPACITY THEORY Figure 3.3: Elements of block occupation (International Union of Railways, 2004) In addition to the block occupation time, further time supplements for stabilisation of the the timetable can be added such as buffer times due to reduction of scattering of delays and crossing buffer times e.g. only used on single track lines, see Figure 3.4. The capacity consumption is calculated according to the formulas below (International Union of Railways, 2004). The total consumption time: k = A + B + C + D, k: the total consumption time [min] A: the infrastructure occupation [min] B: the buffer time [min] C: the supplements for single track lines [min] D: the supplements for maintenance [min] The capacity consumption: K = k 100 U, K: the capacity consumption [%] U: the time interval [min]

37 CHAPTER 3 21 Figure 3.4: Elements of Capacity Consumption (International Union of Railways, 2004) 3.3 Congestion The capacity consumption of a railway line can be evaluated from Table 3.1 below. The general UIC recomendations suggest that for mixed-traffic lines, which the main network in Denmark is characterized as, the maximum capacity consumption should be 75% in peak hours. The maximum capacity consumption differs from daily periods, as it is accepted to run a higher risk of not being able to recover from delays in shorter time spans during the day. The difference according to the type of line is related to the traffic running on the line. For heterogeneous traffic there is a larger probability of faster trains catching up with slower trains and hereby experiencing secondary delays (Landex, 2007). However, if the number of trains operating on the line is low, the delays will not scatter so intensively, and a higher utilisation is allowed. The infrastructure is not allowed to have a 100% utilisation, as delays will spread to the rest of the network. Table 3.1: UIC intervals for maximum capacity consumption (International Union of Railways, 2004) Type of line Peak hour Daily period Comments Dedicated suburban passenger traffic Dedicated high speed lines 85% 70% 75% 60% Mixed-traffic lines 75% 60% The possibility to cancel some services allows for high levels of capacity utilisation Cen be higher when number of trains is low (lower than 5 per hour) with strong heterogeneity

38 22 CAPACITY THEORY 3.4 Evaluating the UIC 406 Method The UIC 406 method has, since its publication, been used for analysing the capacity consumption on both single- and multiple track lines. Yet, discussions of its applicability have surfaced. The leaflet suggests that single track line sections must be divided between all stations. This entails that the timetables are compressed between all stations without considering the bindings, which appears in single track operation and turnarounds in multiple track operation. This yields a much lower capacity consumption than considering the entire single line operation. It is therefore debatable, if this methodology reflects the actual capacity consumption. This has given rise to a discussion on alternative ways of decomposing the infrastructure, given that different decompositions will yield different results. Applying different methods will, however, lead to incomparable results. In the newest version of the UIC 406 leaflet from 2013, it is recommended to consider entire routes when decomposing the timetables (Jensen, L. W., Landex, A. and Nielsen, 2015). Another discussed strategy is to divide the line sections at crossing stations, and also add a crossing buffer time to the capacity consumption as a supplement for trains running bidirectional, see Figure 3.4. In an article from 2008 concerning the evaluation of the UIC 406 capacity method used on single track operation, Landex states that the method can only be applied with success to single track lines, when at least more than one train runs simultaneously in the same direction. To streamline the execution of the analysis, Landex introduces an extended version of the UIC 406 method used on single track lines, see sections below (Landex, 2009a). Comparing the typical operation between a double track- and single track railway line results in a major difference in the possible number of trains able to run on the lines. The factors of a closely planned timetable and the location of crossing stations are of high importance when operating a single line railway. The double track lines are on the contrary not as dependent of crossing stations. The operation can be divided according to running direction and hereby allowing a significantly larger amount of trains. In general the operation of single line railways are characterized as being homogeneous according to stop pattern and identical number of trains running in each direction. In the article Methods to estimate railway capacity and passenger delays, Landex presents a paradox of the UIC 406 method; the issue that operating more trains on the single line railway can result in less capacity consumption. By operating more trains additional crossings are needed, hence, the line is divided into smaller sections allowing a larger compression of the timetable. A suggested way of overcoming the paradox is to add dummy trains to the existing timetable, to identify the relevant crossing stations and hereby the division of the line into sections. This variation of the analysis is used in Denmark for calculating the capacity on single track lines. The type of dummy trains can vary; however, initially trains similar to the slowest train is added followed by trains similar to the second slowest and so on. Furthermore, the trains must be added simultaneously in bidirectional order. The capacity consumption is calculated for each line section according to the UIC 406 method, and the section with the highest capacity consumption determines the total capacity consumption of the line (Landex, 2008).

39 CHAPTER New Method for Assessment of Capacity Consumption in Railway Network Using the UIC 406 method requires input of a predefined timetable. Producing timetables is, however, a very time consuming activity. In addition, the dynamic phase of strategic planning of future railway operation will often result in high levels of uncertainty in the proposed timetables. It is therefore desirable to evaluate the capacity consumption in a railway network without applying timetables, to optimise the use of resources and reduce the cost during strategic planning. This efficiency has Jensen strived to achieve through his Ph.d project in Jensen has developed a model, which calculates capacity consumption distributions without applying timetables. The aim of the model is to assist decision makers with a tool for evaluating the capacity consumption as well as the robustness of a future railway network. The model is presented in the article; Assessment of Stochastic Capacity Consumption (Jensen, L. W., Landex, A. and Nielsen, 2015), which is used as a source for the following description. The model evaluates the capacity consumption on every permutation of train sequences based on a given operation plan of number and types of trains. No exact timetable is therefore required, but only running times and a minimum headway time for the trains within the network. By applying traditional stochastic simulation it is furthermore possible to model delays and thereby derive the uncertainties of normal railway operations. This enables the model to account for robustness in the network Model Framework The model is built by a two-layer representation of the infrastructure. First layer consists of a model for calculating running times and minimum headways based on predefined routes and train characteristics. Second layer stores the most important aggregated information from the first layer. The first layer can be derived in any possible way, yet, a microscopic model is preferred, since the model will never be better than its input. During strategic planning it can, however, be impossible to derive such specific data in an future network. In such case, Jensen proposes microscopic modelling for the existing infrastructure, while a simpler model or solely desired headways and speed profiles are applied for future extensions of the network. The second layer of the model is a mesoscopic model, which yields fast computation. This two-layer framework enables detailed and precise modelling, while maintaining the computational complexity at a comprehensive level. The mesoscopic model framework contains a graph G = (V, E), where edges e E can be either directional, bidirectional (single track) or pseudo bidirectional (at-grade junctions). Figure 3.5 shows an example of how a schematic track layout is represented in the model. The input parameters consist of a set of trains s S, where each train is assigned to a route r R and a running time T s. The routes are composed by a set of continuously connected edges E r E, and a train leaves the network when completing its route. Each edge is assigned a minimum headway represented by a n-by-n matrix, H e, with n trains in the network. Finally, the network can be divided into partitions, P, to enable the model

40 24 CAPACITY THEORY Figure 3.5: Schematic track layout transformed into Jensen s proposed model (Jensen, L. W., Landex, A. and Nielsen, 2015) to only consider parts of the network. The partition p P is a set of connected edges p E. Multiple partitions can overlap, which must be considered when assessing the capacity consumption of a given route. The user of the model must therefore predefine which edges applies in a route and the associated running time on every edge for each train. Overtaking is therefore predefined as well. Ideally the model would be able to adjust the route choice to account for new overtaking. It is expected that this change will make the network more flexible, thus, leading to less capacity consumptions. Jensen works currently towards incorporating this in a new version of the model. For this analysis it is, however, necessary to predefine all location of overtakings. The model calculates the capacity consumption as the ratio between the required time for completing a given train sequence, t, and the considered time interval, T, which is normally set to one hour. The model finds the required time, t, by compressing the given train sequence with regards to the minimum allowed headway. In cyclic operation, this is the duration of minimum cycle time, while acyclic operation calculates elapsed time as the timespan between the first and last train. This method is similar to the UIC 406 method, yet instead of decomposing the network, it allows for network limitations by considering entire routes or the total network in accordance with the recommendation of the newest version of the leaflet from 2013 described in section 3.4. Furthermore, the model calculates shadow capacity, which is the capacity consumption found by compressing the timetable on every edge without regards to the entire network. The method for calculating the shadow capacity is therefore more similar to the UIC 406 method however, it differs in the decomposition of the network which can lead to incomparable results. The capacity consumption is calculated for all permutations of train sequences resulting in a distribution of (n 1)! and n! values for cyclic and acyclic operations, respectively.

41 CHAPTER 3 25 To simplify the calculation of a very large network, the model additionally enables calculation of samples of train sequences. In both cases, the model allows for train sequences with a more or less degree of bundling of slow and fast trains, which especially for heterogeneous traffic, has high influence on the capacity consumption. Hence, a high degree of bundling of trains entails lower capacity consumption than no bundling of trains. This strategy of investigating all or a sample of permutations of train sequences entails, that the model can spare an exact timetable. Generating all permutations of train routes is, however, slightly unrealistic given that not all train sequences will make sense. Jensen therefore has developed a new edition of the model, which allows for this. By eliminating the bundling routes to a certain extent, the results are more realistic, however, the capacity consumption is increased. A flowchart of the model is shown in Figure 3.6. Figure 3.6: Lars Wittrup Jensen s proposed model framework. The step of Calculate Delay Propagation is only relevant for the stochastic model (Jensen, L. W., Landex, A. and Nielsen, 2015) Stochastic Extension To sum up, the input of the basic model is the infrastructure formulated as a graph, routes, a plan of operation and partitions. Based on this, the model calculates the capacity consumption for all or a sample of permutations of train sequences. In this way, the model will generate the same output every time, hence, producing deterministic results.

42 26 CAPACITY THEORY In reality a given network will from time to time be exposed to external factors causing delays. These delays can have large consequences for a tight planned timetable, and it is therefore important to include the aspect of robustness in a capacity assessment. In recent work, Jensen aims to improve the foundation for assessing robustness in the phase of early strategic planning, by adding a stochastic simulation to the model Analysing Model Results The model generates outputs based on the total network utilisation and also the utilisation distributed on edge level. Previously, Jensen has not addressed the edge based output, however, this is examined in current analysis. The edge based output can be used to evaluate the capacity consumption on each edge in the network. In this way, it is possible to investigate how the capacity distribution is affected, when the network is extended or the traffic is dismantled through alternative corridors. The result concerning the utilisation of the total network capacity is also assessed, by comparing the amount of feasible timetables to the total amount; hence, the consumption of train sequences with a maximum capacity consumption of 100 %. Looking at the UIC intervals in Table 3.1, it can be argued that the capacity consumption in the deterministic model should be less than 75 %, as recommended for mixed-traffic lines during peak hours to ensure robustness in the timetables. Yet, these numbers are based on the UIC 406 leaflet, where the network is decomposed at all overtaking stations. In this model, the capacity consumption is determined with regards to network limitations. The capacity consumption is therefore determined by compressing entire network. This entails an increase in the calculated capacity consumption compared to the total decomposed line, which should be considered when analysing the model results. 3.6 Summary of Capacity Theory The previous chapter presented the three capacity methods, which are used in the further capacity analysis of the Copenhagen-Fehmarn corridor. The UIC 406 method divides the railway line into sections at crossing stations and junctions, and the capacity consumption is calculated by compressing the graphical timetable (International Union of Railways, 2004). The method gives rise to a paradox; less capacity utilisation with the addition of additional trains. Landex introduces an extension to the method by adding dummy trains to the timetable, while maintaining the crossings and stopping times (Landex, 2009a). A third method is introduced by Jensen in his Ph.d. dissertation; a new model for determining the capacity utilisation within a network. The model calculates the capacity consumption of a network by using inputs of running and headway times on edges. The model generates the capacity consumption for all permutations of train sequences, and an evaluation of the utilisation of the network can hereby be obtained (Jensen, L. W., Landex, A. and Nielsen, 2015). The model requires no specific timetable input and can therefore be used in overall strategic planning processes. The following chapter presents the capacity analysis for the Copenhagen-Ringsted-Fehmarn corridor and the Lille Syd line, both separately and with an integration of the Lille Syd line in the corridor.

43 Chapter 4 Capacity Analysis Based on the theory presented in chapter 3, a capacity analysis of the corridor from Copenhagen to Fehmarn is conducted grounded on year The aim of the analysis is to investigate the possibility of relocating traffic from the main line passing Ringsted to the Lille Syd line, and clarify how this will affect the capacity consumption of both lines. To limit the analysis without compromising the results, it is decided only to analyse the capacity consumption from Copenhagen to Nykøbing Falster Station. Initially, the basic assumptions are described, including how the input used for analysing the capacity consumption of the base scenario is derived. Based on this, the UIC 406 method is used to determine the future capacity consumption on the Lille Syd single track line, while the deterministic model, proposed by Lars Wittrup Jensen, is applied to evaluate the capacity consumption of the main line passing Ringsted for year Effort is made to analyse the capacity consumption of both the edge level and the total network. Subsequently, train lines which potentially can be relocated to the Lille Syd line is identified. By assuming the Lille Syd line is extended from single to double track, Jensen s model is used to evaluate the capacity consumption on both the Lille Syd line containing additional traffic, and the relieved main line passing Ringsted. Furthermore, different speed profiles for the Lille Syd line is investigated, to clarify how these will influence the overall capacity consumption. Finally, the determined capacity consumptions in all scenarios are used to assess the potential of involving the Lille Syd line in the corridor between Copenhagen and Fehmarn, to achieve a more harmonised distribution of capacity consumption in the entire network. 4.1 Input for Initial Capacity Analysis To carry out the capacity analysis, it is necessary to obtain information on the amount of trains and their associated travelling time in the network in Below is described the foundation for the analysis, and the procedure for deriving the required model input Operation Plan 2027 With respect to the future projects described in chapter 2, Rail Net Denmark has proposed an operation plan for the Danish passenger lines in year 2027, see the enclosed CD. Figure 4.1 shows an overview of the line variants, which are planned to run on the lines 27

44 28 CAPACITY ANALYSIS Figure 4.1: Line patterns in the corridors between Copenhagen Central Station and Nykøbing Falster Station proposed in Rail Net Denmark s operation plan for 2027, see the enclosed CD. Illustrated by the authors between Copenhagen Central Station and Nykøbing Falster Station. The stop pattern of each line is illustrated as black dots on the line. In general, all lines will run at a frequency of 1 train per hour per direction with exception of line 20 (purple), the direct line between Copenhagen and Hamburg, which only departs every second hour. This analysis will focus on the hour where line 20 departs, to capture the most trafficked hour. The grey lines in Figure 4.1 indicate the two planned freight paths for These will separate from the remaining traffic after Ny Ellebjerg Station and continue towards Kastrup and Sweden. Rail Net Denmark s operation plan for 2027 contains no exact time for departures or arrivals of freight trains. Figure 4.2: Overview of the coupling between the Northern and Southern part of Lille Syd. For the entire drawing see TTA placed in Appendix P The red line variants 61 and 62 run on the Northern part of the Lille Syd line between Roskilde and Køge. The coupling of the Northern and Southern part of Lille Syd is

45 CHAPTER 4 29 planned to occur just before Ølby Station, as shown in Figure 4.2 (For entire drawing see enclosed drawing TTA placed in Appendix P. This means, that line 61 and 62 will run on Lille Syd between Ølby and Køge. As described in chapter 2, it is suggested that the local line; Østbanen, in the future will take ownership of the line between Køge and Roskilde and thereby establish a direct connection between Hårlev and Roskilde. Based on the current planned track layout, the local line will, however, still be forced to run on the Lille Syd line between Køge and Ølby, creating a potential bottleneck. This line is not included in the electrification programme, and the rolling stock used for these line variants are therefore assumed to be diesel run Running Time Calculation The operation plan for 2027 contains timetables of all passenger lines in the content of departure and few arrival times given in minutes. However, to conduct the capacity analysis at hand, it is necessary to obtain more precise information of the running times of all line variants. The running times are therefore calculated by applying the running time calculation programme developed by Lars Wittrup Jensen during his Master Thesis in This programme estimates acceleration and braking conditions for a given train type, and can be used as a rough estimate of the required running time. The total running time is then calculated by adding a running time supplement. The programme is sufficient for initial studies, while more precise simulation programmes such as Railsys or TPS are recommended for detailed timetable planning. The programme requires inputs of; speed profile, stop pattern, train type and running time supplement. The formulas supporting the running time calculations of the programme are enclosed in Appendix B (Jensen, 2012) Train Type The rolling stock s ability to accelerate and decelerate has large influence on the total running time, and the driving characteristics of the train must therefore be established to carry through the running time calculations. Attached to Jensen s running time calculation programme is a list of different train types containing information on the driving characteristics, see Table 4.1. This includes data on start acceleration a 0, maximum speed V max, braking percentage C, and braking ratio c, all required to calculate the running time according to the formulas listed in Appendix B. The braking percentage C is the ratio of train weight and braked weight, and is additionally provided as parameters in the table. Furthermore, the maximum allowed speed is defined, to ensure an upper limit for the rolling stock s speed ability. Finally, the train type parameters contain information on the length of the rolling stock (length of locomotive, length of train/unit, and number of units). This information is required to ensure, that the rolling stock at no time exceeds the desired speed profile. The train must therefore never accelerate, before the total train has passed the lower speed restriction. Information on the train length enables the model to determine the position, where the train is permitted to accelerate. (Jensen, 2012)

46 30 CAPACITY ANALYSIS Table 4.1: Characteristics of the train types applied for the operation plan of 2027 for passenger trains, (Jensen, 2012) Description Symbol Unit Litra ET Velaro Freight Lint Train/unit length - m Length of locomotive* - m N/A N/A N/A Number of units* N/A 1 Start acceleration a 0 m/s Maximum speed v max m/s Allowed max speed - km/h Braking percentage C Braked weight - ton Train weight - ton Braking ratio c Table 4.2: Overview on applied train types Route Line Pattern Color Train Type Comment Kh-KjN-Næ 35 ET Running at Lille Syd Kh-KjN-Næ 36 ET Running at Lille Syd Kh-Nf 21 ET Passing Ringsted Kh-Rg 15/16 ET IC Towards Funen and Jutland Kh-Rg 11 Velaro IC-Lyn towards Funen and Jutland Kh-Rg 12 Velaro IC-Lyn towards Funen and Jutland Kh-Rg 13 Velaro IC-Lyn towards Funen and Jutland Kh-Nf 20 Velaro Passing Ringsted Rg-Nf 22 ET - Rg-Nf 23 ET - Olb-Kj 61 Lint Running at non-electrified line Olb-Kj 62 Lint Running at non-electrified line Kh-Nf Channel 1 Freight - Kh-Nf Channel 2 Freight - There is still no final decision on which type of rolling stock DSB will set into operation to realise the operation plan of However, a fair assumption is, that DSB will stake on off-the-self items. This capacity analysis is, to the extent possible, built on the same assumptions made by DSB in their equipment plan for 2014, where running time calculations are performed for the One-Hour Model. Here, the well tested Oresund s train (ET Litra) is applied for conventional lines, yet, with a top speed of 200 km/h instead of 180 km/h (DSB, 2014). This assumption is further supported by COWI in the report Screening of Regional Railway Lines for 160 km/t, carried out in 2012, where it is stated, that the ET

47 CHAPTER 4 31 Litra train contains a level of driving abilities expected for the future regional network by 2020 (COWI, 2012). The Velaro train is applied for the direct trains travelling towards Jutland and Hamborg (IC-lyn), to benefit the top speed of 250 km/h for the new Copenhagen-Ringsted high speed line. The non-electrified trains, which only include line variant 61 and 62, running on the line between Ølby and Køge are assumed to be of the Lint type. Finally, freight trains are assumed to have a maximum allowed speed of 100 km/h and a length of 725 meters. This length must be considered later, when designing tracks used for overtaking. Train characteristics used within this analysis are listed in Table 4.1. An overview of the applied train types for the different line variants is provided in Table Speed Profiles Another decisive factor for the running time is the train speed. In general, the train speed will strive to follow a given speed profile, however, factors as acceleration and deceleration or the upper bound of the maximum possible speed can entail the rolling stock to run at a lower speed. The corridor from Copenhagen Central Station (Kh) to Nykøbing Falster station is composed by several lines, namely; the existing infrastructure, the new Copenhagen-Ringsted (Kh-Rg) high speed line, the Ringsted-Fehmarn line and potentially the Lille Syd line, as shown in the overview map in Figure 4.3. Evaluation of the capacity consumption for year 2027, requires speed profiles for the railway lines as they are planned to function in the associated year. Thus, effort is put into clarifying the future speed profile based on the upgrading and renewal projects planned to be carried out on each of the lines. Figure 4.3: Overview of the corridor from Copenhagen to the fixed link at Fehmarn, illustrated by the authors Speed Profile for the Lille Syd Line Currently, the Lille Syd line is in the process of being upgraded from 120 km/t to 160 km/h on open lines as a result of the electrification programme and a speed upgrade project carried out in the years (Rail Net Denmark, 2015a). In this analysis, it is assumed that these projects are finalised. The base scenario is therefore based on the speed profile presented by Grontmij in the Programming Report from March For detailed information on the composition of the speed profile, see Appendix C. The speed profile for the Lille Syd line is shown in Figure 4.4. It is noted, that the Programming Report only focus on the section between Næstved and Ølby, hence the profile covering the area between Køge Nord and Ølby is not part of the Grontmij analysis.

48 32 CAPACITY ANALYSIS 170 Speed Profile Lille Syd Line KjN Ølb Kj Hf Th Hz Ol Næ Næn Speed in [km/h] Stationing TIB-04 [km] Stations Lille Syd Figure 4.4: Planned speed profile for the Lille Syd line, based on (Grontmij, 2014) Speed Profile for the Main Line Passing Ringsted The line variants running between Copenhagen Central Station and Nykøbing Falster will use the existing line, the new Kh-Rg line and the Ringsted-Fehmarn line. The speed profile of the existing line is found in the track documentation TIB-06, the route between Kh and Vigerslev Alle (Vgt) (Rail Net Denmark, 2015b). Hereafter, the speed profile of the new Kh-Rg line takes in. This profile is based on a drawing of the schematic track layout from the tender documentation, see the enclosed CD. The new Kh-Rg line terminates east of Ringsted Station, where the future track layout still is undecided as described in chapter 2. Investigations of the station capacity in detail are considered outside the scope of this analysis. The following assumptions are therefore drawn for the further analysis; the double track from the Kh-Rg line is extended towards the station and subsequently continued towards Næstved. The schematic track layout shows the Kh-Rg speed profile until stationing , see enclosed CD. From here and till Ringsted Station, the speed is assumed to be 120 km/h in accordance with the existing infrastructure shown in TIB-1 (the route passing Roskilde) (Rail Net Denmark, 2015b). The remaining speed profile between Ringsted and Nykøbing Falster Station is determined according to the Ringsted-Fehmarn project. This project prepares the land-based facilities towards the Fehmarn Belt fixed link. This includes a speed upgrade up to 200 km/h, electrification, and extension of single to double track on the sections from Vordingborg towards Fehmarn (Rail Net Denmark, 2013b). A long political debate in relation to carrying through the project of the fixed link has taken place. However, the upgrade of the Ringsted-Fehmarn land-based facilities is only relevant in case the fixed link is enacted. The capacity analysis carried out here therefore assumes, that the project is finalised. The speed profiles of the three lines are composed into one with upward stationing, to

49 CHAPTER 4 33 enable calculations of running times by Jensen s programme. Detailed description on how the speed profiles are derived and composed is placed in Appendix C. The profile used is shown in Figure 4.5. Figure 4.5: Planned speed profile for the entire corridor from Copenhagen to Nykøbing Falster passing Ringsted, for further description of the composition see Appendix C Due to simplification in the project material, no distinction is made between the speed of the right and left track, which is assumed to be equal. When calculating the running time in the reverse order (from Nf to Kh), the profile is therefore simply mirrored. An overview of the geographical placement of stations, and how the original stationing is converted is shown in drawing TTA placed in Appendix P Running Time Supplement As described in chapter 3, the running time supplements are added to the timetables to absorb minor delays. The size of the running time supplement is a trade-off between; at one hand aiming for a high regularity, which is easier obtained when applying large supplements, and on the other hand aiming for low travel times. In Table 4.3 the running time supplements applied by Rail Net Denmark are listed together with the ones recommended by UIC. Both are given in percentage, but the UIC adds additionally a supplement of 1.5 minutes per 100 km. which makes it difficult to make a direct comparison (Schittenhelm, 2011). The percentage is used to allow for varying distances between stops. Furthermore they increase in correlation to speed whereas

50 34 CAPACITY ANALYSIS high speed on lines will result in shorter travel times opposed to lower speed, in relation hereto a larger supplement is added to ensure a sufficient supplement. Table 4.3: Running Time Supplements used by Rail Net Denmark and recommended by UIC norms (Schittenhelm, 2011) Speed Interval Time Supplement km/h Rail Net Denmark UIC Recommendations V 75 3 % 3 % min/100 km 75 < V % 3 % min/100 km 100 < V % 3% min/100 km 120 < V % 3% min/100 km 140 < V % 4% min/100 km 160 < V % 5% min/100 km 180 < V % 5% min/100 km 200 < V % 6% min/100 km 250 < V % 7% min/100 km In general, Table 4.3 shows that Rail Net Denmark applies larger time supplements as recommended by the UIC norms. This shows, that Rail Net Denmark ascribe great importance in achieving high regularity on the cost of short travel times. This strategy is debated by experts in the Danish railway sector, who accuse the railway operators of adding a wasteful amount of running time supplements on the expense of increased travel times (Østergaard, 2013). This discussion is, however, outside the context of this capacity analysis. Still, it is important to underpin that a reduction in travel time can be obtained by reducing Rail Net Denmark s time supplements. In the long run this could reduce the amount of work required for reducing travel times on the Danish railway lines. Yet, the assumption for this analysis is to comply with the standards set by Rail Net Denmark. These running time supplements are therefore applied when estimating the running times Stop Pattern The final input required by Jensen s running time calculation programme is the stop patterns. For passenger trains, this is read off in the operation plan for 2027, see the enclosed CD. For freight trains, another approach is needed since the operation plan does not provide any information on these. As point of departure, freight trains are not required to stop, given that they contain transit goods. In reality, it is, however, necessary to be able to overtake these trains, due to their much lower speed (maximum 100 km/h) compared to the fast passenger trains. Since the capacity model proposed by Jensen is not able to adapt overtaking in the current train sequence, it is necessary to identify where in the

51 CHAPTER 4 35 network overtaking will take place. This is done by examining the schematic track layout of the entire network. Figure 4.6 shows an overall drawing of the position of overtaking tracks between Copenhagen and Nykøbing Falster on the line passing Ringsted, including their design speed. Based on the assumption of maximum allowed speed of 100 km/h for freight trains, it is, however, not possible to exploit the design speed of 120 km/h at KjN and Oh. This is taken into account when using Jensen s running time calculation programme. A more detailed overview plan of the schematic track layout containing stationing names is subjoined as drawing TTA placed in Appendix P. Figure 4.6 shows that overtaking are possible at the stations; Køge Nord (KjN), Næstved (Næ) and Lundby (Lu), when travelling towards Nykøbing Falster. In the reverse direction it is additionally possible to overtake at Vordingborg (Vo) and Ringsted Station (Rg). Furthermore the Ringsted-Fehmarn land-based facility project prepares two new overtaking tracks on the open line at Møllebækken (Mb) and Ore Hoved (Oh), which can be entered from both directions. Figure 4.6: Position of overtaking tracks for the line passing Ringsted, illustrated by the authors As a basis for planning the freight path, Rail Net Denmark has provided a graphical timetable for the section between Ringsted and the fixed link at Fehmarn, see the enclosed CD. This graph is used to read off the position of overtaking of freight trains on the section between Ringsted and Nykøbing Falster Station. For the rest of the network, it is assumed that all freight trains are overtaken at Køge Nord Station. The stop pattern of the freight paths varies every other hour, as a result of the direct passenger line towards Hamburg (Line 20) only runs every second hour. Because the model only considers one hour is a distinction made between the first and second hour, whereas this analysis focus upon the second, given that this hour contains more traffic. For the direction towards Nykøbing Falster, the enclosed graphical timetable only shows one freight path, stopping at Mb in the first hour (G4), and at Mb and Lu in the second hour (G6). To obtain two freight paths in accordance with the visions for 2027 an extra freight path is enclosed in the timetable. This runs every hour and stops only at KjN (GX). For the direction towards Copenhagen, the enclosed graphical timetable already shows two freight paths. In the first hour, a freight train runs without stop between Nykøbing Falster and Ringsted Station (G1). In the second hour, the freight train is overtaken at Lu (G3). Furthermore, a freight train runs every hour with stop at Næ(G5). An illustration of the stopping pattern for the freight trains passing Ringsted is shown in Figure 4.7

52 36 CAPACITY ANALYSIS Figure 4.7: Planned overtakings for the line variants passing Ringsted, illustrated by the authors The stopping time can vary given the fact that, additional time must be added to achieve a timetable without conflicts. However, when applying Jensen s new model for determining the capacity consumption without using timetables, only the minimum required stopping time is relevant. By considering the size of the passenger exchange on all stations, it is decided to apply the minimum stopping times listed in Table 4.4. These time includes supplement for start acceleration. Table 4.4: Minimum stopping times including start acceleration supplement. *The minimum stopping time for freight trains applies at all overtaking tracks. Passing both lines Passing Ringsted Passing Lille Syd Station Min. Stop Station Min. Stop Station Min. Stop Ny Ellebjerg (Nel) 30 sec. Ringsted (Rg) 30 sec. Ølby (Ølb) 30 sec. Køge Nord (KjN) 30 sec. Glumsø(Gz) 30 sec Køge (Kj) 60 sec. Næstved (Næ) 60 sec Lundby (Lu) 30 sec. Herfølge (Hf) 30 sec. Freight Trains* 180 sec Vordingborg (Vo) 60 sec. Tureby (Th) 30 sec. Nørre Alslev (Nv) 30 sec. Haslev (Hz) 60 sec. Eskildstrup (Ek) 30 sec. Holme Olstrup (Ol) 30 sec. Tingsted (Tn) 30 sec. Næstved Nord (Næn) 30 sec Graphical Timetable Based on the input described above, the running time for all lines is calculated. The graphical timetables shown below are based on departure times given in the operation plan for The operation plan contains no detailed information on arrival and departure times, and it has therefore been necessary to add extra stopping time to eliminate obvious conflicts. As a point of departure, it was believed, that by applying the departure times stated in Rail Net Denmark s operation plan for 2027, it would be possible to add two freight paths and obtain timetables without conflicts. This is unfortunately not the case, particularly when observing the graphical timetable from Nykøbing Falster passing Ringsted.

53 CHAPTER 4 37 This is possibly caused by the uncertainties of the position of overtaking. Another reason could be the assumption of a maximum speed of 100 km/h for freight trains, where Rail Net Denmark might has allowed a speed of 120 km/h. No detailed attention has been granted towards creating timetables without conflicts, given that the most essential knowledge for the further analysis is the travel times. Figure 4.8 and Figure 4.9 show the graphical timetables for the route passing Ringsted respectively, while Figure 4.10 shows the graphical timetable planned for the Lille Syd Line, also placed in Appendix D Figure 4.8: Graphical timetable running from Copenhagen and passing Ringsted, for larger figure see Appendix D Figure 4.9: Graphical timetable running from Nykøbing Falster and passing Ringsted, for larger figure see Appendix D

54 38 CAPACITY ANALYSIS Figure 4.10: Graphical timetable for Lille Syd Headway The last parameter required for conducting the capacity analysis is a predefined headway. The main reason for having a headway of several minutes is due to the safety systems. The exact position and speed of the train is difficult to determine, and it is therefore estimated within the total block section having a range of km. In the current signalling system is the position of trains only updated once every minute resulting in longer block occupation than necessary. In the new signalling System is the position and train speed updated continuously, which can result in less required headway time. (Landex, 2013) As a rule of a thumb uses Rail Net Denmark currently a headway of four minutes to obtain a proper level of regularity. However, in the most trafficked sections this can be reduced down to 2.5 minutes, according to the received from Ib Flod placed in Appendix E. Within this analysis it is decided to use headway times of 3 minutes in the entire network to allow for the improved technology in the new signalling system, which is assumed to be fully deployed in In a more detailed analysis it would be interesting to conduct a sensitivity analysis of the applied headway, to investigate the capacity consumption when an increased headway of 4 minutes is applied. Another approach would be to differentiate the headway throughout the network, so for instance the heavy trafficked sections near Copenhagen is assigned a lower headway than the rest of the network. 4.2 Capacity Analysis on Lille Syd The capacity consumption on the Lille Syd single track line is analysed by using the UIC 406 method with a headway of 3 minutes. When compressing the timetable the following two assumptions are made;

55 CHAPTER 4 39 All crossing stations can handle parallel movement hence, two trains can approach the station simultaneously Crossing trains can arrive and depart from the same station at the exact same time These assumptions are justified by the new signalling system, which is believed to increase the flexibility when trains meet at stations. Based on the discussion from chapter 3, different approaches are applied for the decomposition of the network to evaluate the influence of the determined capacity consumption. In Figure 4.11, the network is decomposed at all possible crossing stations according to the original UIC 406 method, while Figure 4.12 shows an alternative decomposition namely, a decomposition at all crossings in the predefined timetable. The third option for decomposing the network introduced in chapter 3 includes insertion of dummy trains, to obtain more crossings in the timetable and thereby allow a smaller decomposition. However, given that the stopping time must be maintained, it is found impossible to fit in any such dummy trains. This approach is therefore omitted. Figure 4.11: Original UIC 406 method for composition of the network at all crossing stations The double track of the Kh-Rg line is according to the schematic track layout drawing continued until stationing The compression of timetables at this section is therefore made between the point of termination of the double track and Kj. The decomposition of both approaches are shown in detail in Appendix F. Here, the figures of the decomposed timetables are applied to read-off the time intervals the compressed timetables occupy the network (k), after which they are put into the formula (4.1) to calculate the capacity consumption K. The considered time interval U is set to one hour. K = k 100 (4.1) U

56 40 CAPACITY ANALYSIS Figure 4.12: Alternative UIC 406 method for composition of the network at crossings The determined capacity consumption for all decomposed sections are shown in Table 4.5. The longest section most often results in the largest occupation time, and thereby the highest capacity consumption on the evaluated line. Thus, this section indicates the total capacity consumption on the line. When decomposing the network at all crossing stations, the highest capacity consumption of 41.8 % is caused by the Hz-Ol section where the travel time is 25.1 minute, see Table 4.5. This is far below the UIC recommendation of 75 % during peak-hours for mixed traffic lines stated earlier in Table 3.1, indicating a very low capacity consumption on the line. This is further supported by the fact, that single track lines often accept higher capacity consumption than recommended by the UIC. When decomposing the network at all crossings, in relation to the predefined timetable, a very different picture is seen. Here the Kh-Hz section triggers a capacity consumption of % due to a running time of 63.3 minutes, see Table 4.5. Opposed to earlier, this result indicates a total break down on the line due to the capacity consumption exceeding 100 %. The actual capacity consumption should be lower than 100 %, given that the analysis is based on Rail Net Denmark s proposed operation plan for 2027, which can be run without conflicts. One reason for obtaining a higher capacity consumption can be inconsistency between running times applied within this analysis and by Rail Net Denmark. This is, however, not the case, given that the calculated running times in present project are lower than the ones found by subtracting departure time between adjacent stations in Rail Net Denmark s operation plan. To ensure consistency, the applied stopping times have been adjusted to the departure times stated by Rail Net Denmark. The capacity consumption exceeding 100 % must therefore be caused by the chosen decomposition of the network. This example shows how the capacity consumption can be interpreted very differently, when applying different methods for decomposing the network. It is therefore of great

57 CHAPTER 4 41 Table 4.5: Capacity Consumption (CC) Decomposed at all crossing stations Decomposed at Crossings Section Occupation CC Section Occupation CC KjN-Kj 24.5 min % KjN-Kj 24.5 min % Kj-Hf 22.6 min % Kj-Hz 63.3 min % Hf-Th 22.6 min % Hz-Næ 37.3 min % Th-Hz 18.8 min % Hz-Ol 25.1 min % Ol-Næ 21.2 min % importance to streamline the methods, when comparing the capacity consumption of several single track lines. The question now comes down to; which methods is most applicable to reflect the actually capacity consumption? The fact that no dummy trains could be added to the timetable, indicates a relative tight timetable also supported by the two crossings. This calls for a high capacity consumption on the line. On the other hand, the graphical timetables indicate, that the timetable can be run without conflicts, thus the actual capacity consumption is below 100 %. Furthermore, it is not unusual to run two trains per hour and direction on a single track line. In cases where the crossing stations are located close to each other and the timetable is well-planned, it is possible to run up to six trains per hour in each direction. An example is Nærumbanen, which operates with a 10 minutes frequency per direction in rush hours. The high frequency is mainly possible due to a well-planned timetable and a totally homogeneous operation. Another example of a single track line operation, is the line between Vamdrup and Vojens. As of 2014, the line is being upgraded to double track, however, 40 passenger trains and 40 freight trains were previously operated daily. This is far above the theoretical capacity, and the timetable is only possible due to extended travel times, and no opportunity for train-free intervals for maintenance (Nielsen, 2004). Both examples above indicate, that it will be doable to operate the expected traffic on the Lille Syd line. Still it is assessed that the actual capacity consumption is relative high, thus closer to % than 41.8 %. Based on this it can be concluded that the Lille Syd single track line must be expanded, if the line is to carry more traffic than currently allowed for in the 2027 operation plan, to avoid obtaining too high capacity consumption which can result in delays. 4.3 Capacity Analysis on the Main Line Passing Ringsted The focus is now drifted towards evaluating the capacity consumption on the main line passing Ringsted, which currently is expected to run all traffic between Copenhagen and the fixed link at Fehmarn. The method is based on the new model developed by Lars Wittrup Jensen for analysing

58 42 CAPACITY ANALYSIS the capacity consumption, without applying a predefined timetable as an input introduced in chapter 3. Effort is made in relation to evaluating the overall network capacity consumption and the distribution of capacity consumption on edge basis The Transformation of the Network Initially the network is transformed into a graph G = (V, E) composed by vertices and edges. The entire transformed network is shown in drawing TTA placed in Appendix P. A smaller segment of the drawing is shown in Figure In the bottom, the stationing names are shown for the direction towards Nykøbing Falster, while the stationing names for the reverse direction are written at the top. A new edge is defined every time the train can change path, for instance at Ringsted Station all trains arrive from Copenhagen at edge 3 after which some leave the network and travel towards Funen and Jutland, while others continue at edge 4 towards Næstved. The edges are shown by different color codes to clarify the beginning and ending of an edge. Figure 4.13: Segment of drawing TTA showing the network transformed into a graph applicable for Jensen s model for estimating capacity consumption Network Input Subsequently the model is fed with data on the amount of trains and their running times on each edge respectively. Only freight trains run on the overtaking tracks in accordance with the predefined stopping pattern. The distribution of trains on the remaining edges are defined in accordance with the operation plan for The running time on every edge is determined by applying Jensen s running time calculation programme, and the input described above. The output of the programme is the braking procedure of the entire line pattern. The edge occupation time is then found by subtracting the arrival time from the departure time on every edge. If the exact stationing is not provided in the braking procedure, a constant linear correlation between the

59 CHAPTER 4 43 Figure 4.14: Overview of the number of trains running in the network arrival times of the two adjacent stationing are assumed. The applied travel time input can be found in the enclosed CD. The deterministic model edition is applied in this capacity analysis. The running time supplements are therefore simply added to the timetable. In a further analysis it would be interesting to test the stochastic extension, however, this has not been possible in this project Results The model output is used to evaluate the network capacity and edge capacity of the base scenario. It is decided only to evaluate the capacity on the direction towards Nykøbing Falster, even though results for both directions are obtained. This is caused by the fact, that the speed profile and line pattern in general is mirrored in the two directions. The only difference is found in relation to the location of overtaking of freight trains. The location of overtaking tracks can affect the overall result to a smaller or larger extent. However, given that these positions are estimated by a best guess procedure as described earlier, the effect of different overtaking positions is not considered within this analysis Edge Capacity The model takes the entire network into account when compressing the timetable, and the result is provided as compressed graphical timetables of 0th, 50th and 100th percentiles on every edge. The 0th percentile therefore shows the result of lowest capacity consumption, the 50th percentile shows the median, while the 100th shows the run with the highest capacity consumption.

60 44 CAPACITY ANALYSIS The graphical timetable outputs are applied for determining the largest occupation time on every edge after which, the same formula as for the UIC 406 (4.1) is applied for calculating the capacity consumption on the particular edges. As an example is drawn edge number 3 shown in Figure Due to the fact that this analysis focus on acyclic operation, the dashed line is ignored given, that it indicates when a new cycle begins. Instead, the duration of the time is read-off at the blue circles. The largest time occupation at link 3 for the 0th percentile is found at the end at 40.5 minutes (the numbers supporting the graph is applied to determine the exact timespan). The capacity consumption of link 3 can then be calculated as (4.2) K = = 67.5% (4.2) Figure 4.15: Output of edge 3, placed between node 9 and 67 which correspond to stationing and , according to drawing TTA placed in Appendix P By calculating the edge capacity on each edge for both the 0th, 50th, and 100th percentile, it is possible to show the distribution of the capacity consumption of the entire network in the base scenario, see Figure The graph shows that the 0th percentile capacity consumption is below the UIC recommended capacity consumption of 75 %. This recommendation is, however, based on the original UIC 406 method, where a double track network is decomposed at the center of overtaking tracks if used. A direct comparison can therefore not be made, however, given that the UIC recommendation is the only available guideline for the evaluation, it seem reasonable to glance at this factor. The capacity consumption of the 50th percentile is in general above 100 %, and the graph shows a small rise between KjN and Rg. Same rise is observed for the 100th percentile yet, this maintains a higher level throughout the rest of the network. In the worst case the capacity consumption exceeds 200 %. These results indicate that the capacity consumption on the main line passing Ringsted will reach its capacity limit in 2027, due to the fact that only few timetables can be accomplished or schedules waiting time has to be added to make more feasible timetables.

61 CHAPTER 4 45 Figure 4.16: Distribution of the capacity consumption in the base scenario for the line passing Ringsted Networks Capacity To evaluate the overall network capacity, the deterministic model provides results of the accumulated distribution of the capacity consumption. This result shows the amount of train sequences, which can be run with a capacity consumption below 100 %. Figure 4.17 shows the result for the base scenario. From here it is found, that only 1.0 % of all train sequences achieve a capacity consumption of up to 100 %. This entail that only very few timetables feasible in the operation plan for 2027, indicating a heavy overutilisation of the future network. Figure 4.17: Cumulative distribution of the capacity consumption for the base scenario on the main line passing Ringsted

62 46 CAPACITY ANALYSIS Validating Results The model applied for estimating the capacity consumption is still in the process of being developed. There is therefore still no precise method on how to interpret the results. Before concluding that the network in 2027 will reach outraged levels of capacity consumption, it is therefore necessary to look at the network from an overview perspective. The most heavy trafficked section is between Ny Ellebjerg and Køge Nord, where a maximum of 10 trains/hour/direction will run after which 8 trains continue towards Ringsted. A rule of a thumb states, that a double track conventional railway line can run trains/hour/direction (Jensen, 2013). It is therefore a puzzle how the 8-10 trains/hour/direction can trigger such high capacity consumption. The reason for the high capacity consumption must therefore be found in the high level of heterogeneity on the line. By running high speed trains with a maximum speed of 250 km/h and freight trains at 100 km/h on the same line, a long occupation of the block sections is expected. This suggests that even though the maximum amount of trains is relative low, the heterogeneity can cause an extremely high capacity consumption. It is therefore worthy to investigate the possibility of removing some of the traffic from the Kh-Rg line to the Lille Syd line, with the purpose of relieving the new Kh-Rg high speed line at the section between KjN and Rg. 4.4 Relocation of Traffic to the Lille Syd Line Looking at the operation plan for 2027 it is clear that four line patterns can be relocated to the Lille Syd line without compromising the stopping service at stations. Significant line upgrades as the extension of Lille Syd from single to double track in combination with a speed upgrade can cause the entire operation plan to shift. However, in this simplified analysis the original operation plan for 2027 is kept as point of departure. The line pattern which can be removed to the Lille Syd line includes; line 20 (the direct train towards Hamburg) line 21 (the regional train towards Nykøbing Falster stopping at Køge Nord and Næstved) and finally the two freight paths. An overview of the new operation plan is shown in Figure In order to ensure sufficient capacity on the Lille Syd line, it is assumed that the entire line is extended from single to double track. It thereby becomes possible to apply Jensen s proposed model, to evaluate the capacity consumption on both the main line passing Ringsted with relieved traffic load, and on the new double track Lille Syd line with increased traffic load. The frame of reference therefore becomes consistent ensuring more comparable results. One purpose of relocating the lines is to release capacity on the high speed line. This is especially relevant for the freight paths, given that these entail large occupation of the line section. Another gain of relocating the traffic is the potential travel time savings due to the fact, that the Lille Syd section is 9.24 km shorter than the line passing Ringsted. Because the line speed of the line passing Ringsted is relatively higher than the Lille Syd

63 CHAPTER 4 47 Figure 4.18: Modified operation plan for 2027 where line 20, 21 and the two freight paths are relocated to the Lille Syd line, illustrated by the authors line this is, however, not necessarily true. The extended analysis therefore includes the following three scenarios for the double track Lille Syd line; Existing speed profile Upgrade to 160 km/h at stations Upgrade to 200 km/h on the entire line The difference between the first and second scenario is, that in the current speed profile planned for Lille Syd, the design speed is maintained at 120 km/h at all stations, while 160 km/h is permitted on the open line. The original operation plan only suggests line 35 and 36 running on the Lille Syd line. Due to the fact that both line 35 and 36 stop at every station, an upgrade of the speed along the stations will be wasteful, given that neither of the trains will be able to exploit the increased speed. When relocating traffic to the Lille Syd line, the speed limitation along stations will, however, cause unwanted braking resulting in prolonged travel times. In relation hereto, it is interesting to investigate the costs and benefits of upgrading the speed profile along the stations to 160 km/h. The last scenario investigate a more comprehensive speed upgrade of the entire line to 200 km/h making it more comparable to the remaining land-based facilities towards Fehmarn. Finally, a gain of upgrading the Lille Syd line is to obtain a more flexible railway network with the possibility of running traffic both on the line passing Ringsted and at Lille Syd. This will be favourable in case of a break down or maintenance requirements on one of the two lines. Furthermore, this is relevant in the long-term perspective, where the need

64 48 CAPACITY ANALYSIS of several freight paths can appear as described in chapter Revised Model Input To apply Jensen s proposed model to estimate the capacity consumption on the Lille Syd line after relocating traffic, new model input is required. This includes alternative running times, and a graphical representation of the network passing the Lille Syd line. Furthermore, the position of overtaking freight trains must be adapted to the new shorter line, resulting in a new stopping pattern for freight trains Speed Profile In order to calculate alternative running times, new speed profiles are composed in the same manner as for the main line passing Ringsted. The procedure of composing the different line elements are described in Appendix C, and summarised in the the enclosed drawings TTA and TTA placed in Appendix P. The drawings show a geographical overview of the entire corridor. In the three scenarios, the speed profile at the Lille Syd line is changed, while the remaining part of the route between Copenhagen and Nykøbing Falster is maintained. The three different speed profiles used in the further analysis are shown in Figure Figure 4.19: Composition of speed profiles applied for running time calculations when relating lines to Lille Syd Stopping Pattern The stopping pattern for the passenger lines is maintained regardless of whether the lines are run across Lille Syd or the main line passing Ringsted. However, when relocating the freight trains, the placement of overtaking must be reassessed due to the alternative route and thereby a new travel time.

65 CHAPTER 4 49 The new stopping pattern is assessed on the basis of the previous. All freight trains will therefore still be overtaken at Køge Nord. This is however not possible with the current track layout, given the overtaking track is placed towards Ringsted after entering the Lille Syd line. The overtaking track is therefore moved just east of Køge Nord Station to enable overtaking. The length of the new overtaking track is set to 1,400 meters in accordance with the longest overtaking track on the main line passing Ringsted. The freight trains running towards Copenhagen are currently stopping at Lundby and Næstved Station as shown in the base scenario in Figure These stations are located before the trains enter the upgraded Lille Syd line, and these positions are therefore kept. The freight trains running towards Nykøbing Falster are in the base scenario overtaken at Møllebækken and Lundby (G6), and only Lundby (G4), in addition to the established stop at Køge Nord. After being relocated to the Lille Syd line, the overtaking at Møllebækken is moved to Næstved. This change is found acceptable due to the shorter distance to Næstved when running along the Lille Syd line compared to the main line passing Ringsted. Furthermore, the extra overtaking for G6 is moved from Lundby to OreHoved, to maintain a proper distance between the second and third position of overtaking. The chosen positions for overtaking of freight trains are shown in Figure Figure 4.20: New position of overtaking for freight paths when running at the Lille Syd line, illustrated by authors Graphical Timetable The travel time for the line patterns running at the Lille Syd line can now be calculated based on the introduced speed profiles and new stopping patterns for freight trains. The travel time is then illustrated in graphical timetables. These are made by fixating the arrival and departure times at Copenhagen Central Station, given that the line section close to Copenhagen contains the most traffic. Aside from this, no additional steps are taken to

66 50 CAPACITY ANALYSIS eliminate conflicts in the graphical timetables, since this is considered outside the scope of this project. The graphical timetables for the lines passing Lille Syd is placed in Appendix D in a larger format, while Figure 4.21 shows a minor preview of the graphical timetable of the right track. To support the conclusion of extending the Lille Syd line from single to double track, trains in both directions are inserted at the Lille Syd line, hence the only single line in the corridor. As shown in Figure 4.21, it would be extremely difficult to retain Lille Syd as a single track line while running the extra traffic. Figure 4.21: New position of overtaking for freight paths when running at the Lille Syd line Graphical Representation of Network The network passing the Lille Syd line is transformed into the graphical representation applicable for Jensen s deterministic model. This is conducted in same manner as for the network passing Ringsted. The composition of edges is shown in the enclosed drawing TTA placed in Appendix P. The running time on each edge within the new network is determined and prepared as model inputs. The graphical presentation of the network on the main line passing Ringsted is more or less reused, when investigating the capacity consumption after relieving the line. Yet, minor adjustments are required to ensure coherence on the lines running at Lille Syd. An extra edge is therefore inserted between Køge Nord and Næstved Station to allow

67 CHAPTER 4 51 for the travel time on the Lille Syd line. This dummy edge is required to make sure, the model allows for coherence between the trains running between Kh-KjN and Næ- Nf. Furthermore, the relocated lines are removed from the edges between KjN and Næ. Finally, new travel times are calculated for the freight trains due to the changed stopping pattern. The applied travel time input on each edge is provided on the enclosed CD Results Below is described the results derived when relocating traffic from the main line to the Lille Syd line in the three scenarios. The aim is to investigate how the capacity consumption is affected when traffic is relocated. On the main line passing Ringsted, it is possible to compare the base scenario with the three scenarios of relieved traffic. For the line passing Lille Syd it is not possible to compare with the base scenario, given this was derived by applying the UIC 406 method. Instead, a comparison is therefore made between the mutual scenarios. Finally is given a short account of the potential travel time savings as a result of relocating the traffic. In spite of this not being a part of the capacity analysis, it is interesting from a overall benefit point of view Edge Capacity Effort has been put into determining the edge capacity in all scenarios to see how this is influenced by relocating traffic and increasing speed. Graphs containing the distribution of edge capacity on both the main line passing Ringsted and the alternative route passing Lille Syd are placed in Appendix G. Unfortunately it has not been possible to interpret these due to large fluctuations in the results. As an example the distribution of capacity consumption for the 50th percentile on the main line passing Ringsted is withdrawn in Figure The capacity consumption for the base scenario is captured in the pink graph, whilst the grey, green and red graphs illustrate the capacity consumption when relieving the line. The overall tendency shows as expected, that the capacity consumption decrease when relocating traffic to the Lille Syd line. The scenarios containing the existing speed profile and the 160 km/h upgrade, show likewise reasonable results. The capacity consumption is in both cases lower than the base scenario and the development indicates a likely correlation. On the contrary, the capacity consumption for the 200 km/h shows large fluctuations throughout the entire network, and after passing Næstved Station it exceeds the base scenario. This tendency is found unexplainable, thus the results can not be interpreted. When examining the associated results for the 0th and 100th percentile same inconsistency is found. This applies too for the edge capacity found on the paths passing the Lille Syd line. Furthermore no consistency is found when comparing the edge capacity on edges reused in both the network passing Lille Syd and Ringsted. This inconsistency can be caused by the large variance in the applied train orders. When withdrawing results based on the 0th, 50th and 100th percentiles, it occurs that different train orders are compared causing interpretable results. To enable comparisons on the edge level it will be necessary to withdraw the same train order. This will, however, require an exact timetable which contradicts the original purpose of Jensen s model.

68 52 CAPACITY ANALYSIS Figure 4.22: Distribution of capacity consumption for the 50th percentile for the main line passing Ringsted Another way of obtaining more applicable results for evaluating the edge capacity will be to change some of the perquisite of the analysis. Instead of applying the compressed timetables, which allow for entire routes to calculate the capacity consumption, the corresponding graph showing shadow capacity could be applied. These graphs compress the timetables without considering the entire network. By composing the network in an alternative way so edges are placed in accordance to the compositions carried out in the UIC 406 method, it is possible to obtain results comparable with the guidelines in the UIC 406 leaflet. This extended analysis would be interesting to perform in a further study Network Capacity Instead, the network capacity is applied to interpret how the capacity consumption is affected, when relocating traffic in combination with speed upgrades of the Lille Syd line. These results are applicable, since the network capacity summarise all investigated train orders, causing less impact of the variance. The cumulative distribution of the capacity consumption is shown for all scenarios in Figure The dashed line indicates the network passing the Lille Syd line, while remaining lines illustrate the cumulative capacity consumption on the main line passing Ringsted. Overall, the largest capacity consumption is found in the base scenario, where only 1.0 % of the train sequences result in a capacity consumption below 100 %. By relocating the traffic to Lille Syd without upgrading the speed profile, it is possible to increase this number to 18.4 % in the main line and 12.6 % on the Lille Syd line. By increasing the speed at Lille Syd, these percentages of feasible train sequences will drop. In the scenario with 160 km/h, the main line passing Ringsted obtains 14.5 % feasible solutions while the Lille Syd line drop to 7.6 %. By increasing the line speed to

69 CHAPTER 4 53 Figure 4.23: Cumulative percentage of capacity consumption in the different scenarios 200 km/h, these numbers will drop even further. The main line will then obtain 10.8 % feasible solutions and the Lille Syd will correspondingly obtain 4.2 %. This shows that a speed increase will increase the capacity consumption of the entire network. This is most likely caused by a higher level of heterogeneity, given the freight trains will maintain a speed of 100 km/h regardless of the speed upgrade. Passenger trains will thereby more likely catch up with slower freight trains resulting in an increased capacity consumption Potential Travel Time Savings In addition to investigating the capacity consumption when relocating the traffic, it is also interesting to glance at the potential travel time savings, in order to obtain additional arguments for realising the project proposal. The travel time savings are shown in Figure 4.24 for the lines running towards Nykøbing Falster. Because the freight trains have a maximum speed of 100 km/h these will not gain any travel time savings when upgrading the Lille Syd line speed to 160 and 200 km/h respectively. These are therefore only treated in the base scenario. Looking at the scenario regarding the existing speed profile of the Lille Syd line freight trains achieve a time saving in the interval of minutes depending on their stopping pattern. Line 20 and 21 will on the contrary experience a prolongation of travel time if relocated to Lille Syd without upgrading the speed. For line 20, 4.1 minutes are added to the travel time, while 1.3 minutes are added to the travel time of line 21. Upgrading the entire line speed to 160 km/h entails a delay of only 0.7 minutes for line 20 while

70 54 CAPACITY ANALYSIS Travel Time Saving (Compared to base scenario) 5.5 Travel Time Saving [Min] Existing 160 km/h 200 km/h -4.5 Scenario /36 Freight X Freight G4 Freight G6 Figure 4.24: Travel time savings by relocating lines to the Lille Syd line line 21 will gain a time saving of 1.2 minutes. This will simultaneous give rise to a small travel time saving for the existing line 35 and 36 on 0.2 minutes. Finally a speed upgrade to 200 km/h on the Lille Syd line will entails travel time savings for all passenger lines, see Figure The line speed must therefore be upgraded in order to obtain travel time savings for the passenger lines. The freight trains will on the contrary experience a travel time saving of almost 5 minutes corresponding to 5 % of the total travel time by being relocated to Lille Syd. 4.5 Summary of Capacity Analysis In present chapter the capacity consumption is investigated for the corridor between Copenhagen and the fixed link at Fehmarn, with eyes on relocating parts of the traffic to the Lille Syd line, to harmonise the overall network capacity consumption. The capacity consumption on the Lille Syd line is initially investigated by applying the UIC 406 method. Here it is demonstrated how the method for decomposing the network has a high influence on the interpretation of the line capacity. Decomposing the network at all crossing stations, as described in the original UIC 406 method, yields as capacity consumption of 41.8 %. When alternatively decomposing the network at all crossings, according to the predefined timetable, a capacity consumption of % is obtained. This demonstrate the problem with applying the UIC 406 method on single track lines. To analyse the capacity consumption on double track lines, Jensen s new deterministic model has been applied. This shows that only 1 % of the investigated train sequences for Rail Net Denmark s operation plan for 2027, can be accomplished with a capacity consumption less than or equal to 100 %. By extending the Lille Syd line from single to double track, and relocating parts of the traffic from the main line passing Ringsted to the

71 CHAPTER 4 55 Lille Syd line, it is possible run 18.4 % of the train sequences on the main line, and 12.6 % on the Lille Syd line. By additionally increasing the line speed on the Lille Syd line up to 200 km/h, these numbers will decrease slightly, but never be lower than the 1 % found in the base scenario. Thus, the network capacity is in all cases improved, when relocating traffic to the Lille Syd line. To support these findings, effort has been put into applying Jensen s model to evaluate edge capacity. Based on large variance in the train orders derived from the results it is, however, not possible to interpret these results. The model is therefore in its current stage, mainly applicable for evaluation of the overall network capacity. To enable evaluation of edge capacity it will be necessary to withdraw same train order, which will require a timetable and thereby contradict the original purpose of Jensen s model. Finally the potential travel time savings are found when relocating the traffic to the shorter route passing Lille Syd. From here it is seen, that all freight trains obtain travel time savings when redirected to the shorter route passing the Lille Syd line. Regarding passenger trains, these obtain however only travel time savings when additionally upgrading the line speed. The findings of the analysis suggest therefore benefits in relation to capacity consumption and, to some extent, time savings when redirecting certain lines to Lille Syd. In the following chapters are therefore conducted to give a technical assessment of the possibility of extending the Lille Syd line from single to double track, while upgrading the line speed to evaluate the project costs.

72 56 CAPACITY ANALYSIS

73 Chapter 5 Norm Foundation Based on the assessment of gains in the network capacity, by integrating the Lille Syd line in the Copenhagen-Fehmarn corridor, and the gains in travel time savings for upgrading the line, the following chapter presents the norm foundation. Rail Net Denmark s norm requirements and Track Rules are compared to the European TSI requirements to establish and discuss the differences. In the further upgrade of the Lille Syd line, both the Danish and European requirements are taken into consideration, to determine the most cost and time saving efficient solution. As of January 1st, 2015, Rail Net Denmark is required to comply with the TSI INF on all conventional lines. The TSI INF requirements are dictated incorporated in the Danish Track Rules and requirements stated by the European Commission. The norm foundation solely include engineering disciplines in relation to track design and dissociate from disciplines soil, power supply, signalling systems etc. The chapter is divided into 8 subsections considering; standards, track geometry, ballast profiles, track center distances, structure gauges, platforms, level crossings, and a summary of the norms and requirements used for the Lille Syd line upgrade. The subsections includes an introduction considering the motivation for the section, and a summary of the most important parameters related to the upgrade of the line. The norm foundation is based on the regulations stated in Table 5.1. In general, only regulations concerning speeds up to 200 km/h are considered. 5.1 Standards Railway standards and norms are defined in order clarify the safety and technical requirements, a railway line must comply with. Seen from a European perspective, individual norms have existed within each country. However, since the agreement of implementing the Trans-European Transport Network, a common norm foundation for the member states to comply with has been formulated in the TSI requirements. Though, national norms still play a central role in the handling of the infrastructure. Current section presents the structure of the Danish norms, and the influence of the TSI requirements. The section is divided in three; the national railway norms, the TSI requirements and the application of the norms. A summary of the standards and norms relevant for the Lille Syd upgrade is provided in the end. 57

74 58 NORM FOUNDATION Table 5.1: Regulations used for establishing the norm foundation Regulations Danish Terminology Date Track Engineering Rules Sporregler 1987 May 2015 BN1-6-5 Tværprofiler for ballasteret spor March 2014 BN1-9-2 Sikkerheds- og opholdszoner påperroner June 2012 BN Indbyrdes placering af spor og perron October 2006 BN Sporafstand og frisporsmærker March 2008 Stucture gauges Fritrumsprofiler January 2014 TSI INF Infrastruktur November 2014 TSI PRM Tilgængelighed for handicappede November 2014 EN Jernbaneudstyr - spor - konstruktionsparametre August 2009 EN A1 Jernbaneudstyr - spor - konstruktionsparametre November 2009 EN /3 Jernbaneudstyr - Fritrumsprofiler January 2010/ May 2013 UIC fiche Rullende materiel fritrumsprofil November The Danish National Railway Norms Rail Net Denmark operates, maintains and develops the Danish state-owned infrastructure. The infrastructure is covered by the Danish railway norms, which function as technical standards. The norms ensure the development and compliance of safety and technical requirements when processes within the infrastructure are carried out. The overall goal of the norms is to ensure that the infrastructure is designed, maintained, planned and operated in accordance with Rail Net Denmark s requirements and standards. The key aspects in the standards are; safety, economy, comfort and demands of customers, and the environment (Rail Net Denmark, 2007). The railway norms are divided in three levels according to priority in case of deviation and approval procedures: BN1 norms are technical standards defined in order to comply with safety and functional requirements. The requirements are defined by Rail Net Denmark and authorities. BN1 norms shall be complied with and always be approved by authorities. BN2 norms are technical standards defined by the board of directors in Rail Net Denmark. BN2 must not contradict BN1 norms and must too be complied with and approved by Rail Net Denmark. Deviation from BN2 norms must only take place within the boundary of BN1 norms and demands dispensation from Rail Net Denmark s board of directors. BN3 norms are specific instructions on how to follow BN1 and BN2 norms. BN3 norms are allowed to be deviated from as long as BN1 and BN2 norms are complied with. In special cases where the norms do not fully cover an area or are too expensive to comply with, dispensations can be given by Rail Net Denmark and must be approved by an authority (Rail Net Denmark, 2007).

75 CHAPTER 5 59 Rail Net Denmark has described a rule hierarchy concerning the technical rules for the railway infrastructure (Nielsen, 2012), see Figure 5.1. As of January 2015, the TSIs are to be complied with within the entire Danish railway network. Working just below the TSIs are the Danish Legislation and declarations from the Danish Transport Authority. The Danish Railway norms, which earlier played the most central role within rule setting, are now being adapted towards the handling of the TSIs. In the bottom of the hierarchy are common EN standards and UIC leaflets as these are more or less integrated or else overruled by the Danish norms. The EN standards are developed by the European Committee of Standardisation (CEN). These are characterised as being national standards in every of the 30 European member state countries. The UIC leaflets are mostly developed in corporation between expert members within the railway networks of the UIC and experts from the industry and standardisation organisms. The UIC leaflets are used as references with the goal of obtaining a standardised construction and operation measure for the railway. They contain technical specifications and must therefore be complied with to allow cross-boarder movement. The UIC leaflets are most often integrated in the European and the national norms (UIC - International Union of Railways, 2015). TSI The Danish Legislation Declarations from the Danish Transport Auhority The Danish National Railway Norms Importance EN Standards UIC Figure 5.1: The rule hierarchy described by Rail Net Denmark (Nielsen, 2012) and illustrated by the Authors The Technical Specifications for Interoperability TSI is an abbreviation for Technical Specifications for Interoperability. The TSI requirements are formulated by the European Railway Agency and are acknowledge by the European Commission, which has approved them as guidelines for developing and ensuring an interoperable Trans-European Railway. The overall goal of implementing the TSI requirements in the Danish railway network is to ensure harmonisation with the remaining European railways. The Interoperability Directive, operating within the European Commission, works towards the objective of improving the performance of the rail transport to increase access, introduce interoperability across boarders, ensure a concatenated European network and finally secure a high safety level (European Railway

76 60 NORM FOUNDATION Agency, 2012). The TSI requirements are therefore mandatory demands for any European railway. The requirements concentrate on two systems; a structural system and a functional system. The structural system focuses on the materialistic aspect and engage with topics as infrastructure, interlocking, energy and the rolling stock, whereas the functional system focuses on operation and maintenance (Danish Transport Authority, 2010b). Since 2002, the decision of formulating and implementing the TSI requirements in the European railway sector, the norm and standard hierarchy has changed. In 2009, the European Railway Agency formulated the TSI requirements concentrating on the infrastructure for conventional Railways, and in 2011 the European Commission approved of them. In 2014, the TSI requirements for conventional and high speed railway lines were consolidated and were sat to function from January 1st, 2015 (European Railway Agency, 2015). The entire railway network in Denmark shall therefore comply with the TSI requirements, however, the sector of metros, trams, light rail systems, and separated local, suburban or urban passenger services are not included (European Railway Agency, 2012). The TSI relevant for presence project is the TSI INF engaging in requirements for tracks, turnouts, bridges, tunnels, and platforms etc. To comply with the TSI requirements a railway line is initially characterized according to the relevant type of traffic running on the line. In general, a line is characterised according to profile, axle load, speed, train length, and operational platform length (Den Europæiske Unions Tidende, 2014a), see Figure 5.2 valid for the Lille Syd line. The traffic categories are divided according to passenger and freight transport. The former two parameters characterises the type of rolling stock allowed on the line, whereas the latter three are guidelines for the traffic types (Liu and Balsby, 2015). The TSI requirements are not as specific as the national norms, and in some cases it is up to the individual country to determine what norms to comply with. In general, the TSI requirements shall always be complied with when constructing a new line, whereas upgrading and track renewal must be carried out based on the TSI requirements. Before a new line or a modified infrastructure can be put into operation, the infrastructure shall be verified by an assessor, and a notified body (NoBo). The assessor is a competent, independent person ensuring, that the system complies with the safety requirements. The NoBo is an authorized body ensuring, that the infrastructure complies with the TSI requirements (Liu and Balsby, 2015). In cases where it is assessed too costly to comply with the TSI requirements, dispensations can be given. Table 5.2: TSI parameters according to traffic type valid for the Lille Syd line (Den Europæiske Unions Tidende, 2014a) Traffic code Profile Axle load [tonnes] Speed[km/t] Platform length [m] Parameters for passenger transport P3 DE Parameters for freight transport F1 GC

77 CHAPTER Application of the Railway Norms The TSI requirements and the Danish railway norms are often reliant on the type of work carried out. In general, a deviation is made between construction of new lines, track renewal and upgrades (Liu and Balsby, 2015). New lines deal with the construction of an entire new line Track renewal is defined as an exchange of a subsystem, that will not change the performance of the system. A distinguish is made from maintenance work, as track renewal makes it possible to adapt the railway line to the TSI requirements An upgrade is defined as a change in a subsystem carried out to improve the performance of the system The construction of the second track along Lille Syd is considered as construction of new lines. The upgrade of the existing track is considered as an upgrade, whereas track renewal is omitted in this project Summary of Standards with Respect to the Lille Syd Line The Danish railway norms are in recent years being modified to encompass the European TSI requirements to facilitate a common Trans-European network. From January 2015, the TSI requirements are to be complied with in all upgrade and track renewal projects as well as construction of new lines. The upgrade of the Lille Syd line, and construction of a second track, shall therefore also comply with the TSI requirements. The TSI requirements specify, that the lines are characterised according to traffic parameters. The Lille Syd line is in current project characterised with the category P3 for passenger transport and F1 for freight transport. 5.2 Track Geometry Track geometry is one of the decisive factors for the permitted speed of a railway line. Present section therefore describes the requirements to the track geometry, when upgrading and constructing a new line. The track design is divided into a horizontal alignment, denominated the track alignment, and a vertical alignment, often referred to as the longitudinal profile. The alignments are composed by the following geometrical elements: Track Alignment (Horizontal Alignment) Straight elements, curves, cants, and transition curves (including super elevation ramps) Longitudinal Profile (Vertial Alignment) Straight elements, vertical curves, and gradients The physical appearance of the elements is deceive for the behaviour of the rolling stock. Infrastructure owners has therefore formulated a long list of instructions and requirements on the composition of the elements, to achieve a track geometry with a sufficient level of safety, comfort, and cost-effectiveness (Esveld, 2014). Rail Net Denmark has assembled their track geometry design rules in the documentation; Track Engineering Rules(TER). This rule set divides the requirements into three categories;

78 62 NORM FOUNDATION Requested Regulations Requirements which should be used for construction of new lines and track renewals to ensure a high level of comfort and reduced wear. Standard Regulations Requirements which shall be used for construction of new lines and track renewals to ensure an acceptable level of comfort and wear. Exceptional Regulations Requirements which can be used in cases, where it is either impossible or very expensive to comply with the standard regulations. When applying these requirements, special permission must be granted by the technical system administrator at Rail Net Denmark, since it reduces the level of comfort and wear, but never compromise safety (Rail Net Denmark, 2013c). In addition to the Danish TER, exists TSI requirements and European Norms regarding the track geometry. The TSI requirements are stated in the documentation; TSI INF, and contain exclusively minimum values (Den Europæiske Unions Tidende, 2014a). As described earlier became a new version of the TSI INF effective for the Danish conventional lines the 1th of January This new version contains a large revision of the track geometry requirements, which has induced an update of the Danish TER, published 1st of May, However, the only revision in relation to the track geometry includes rephrasing of the application field of the TSI requirements; from TEN-lines to the entire railway network (Rail Net Denmark, 2013c). Based on this, a reason to believe that a new edition is on the way exists. The European Norm documentation on track geometry is found in the standards; EN part 1; geometry normal track and 2; geometry turnouts. These standards contain recommendations for both normal and exceptional values (European Committee For Standardization, 2010), (European Committee For Standardization, 2009). In the section below, a short introduction to the most essential elements included in the horizontal and vertical track geometry design for railway lines with speeds up to 200 km/h is given. The description is based on Rail Net Denmark s requirements, and comparisons are made with the associated TSI requirements, to ensure that all valid requirements are met. Furthermore, a comparison with the EN13803 standard is made, to investigate the potential of relaxing certain design requirements, to reduce construction costs in upgrading projects Horizontal Curves Curves are placed in the horizontal alignment to enable the track to change direction. The permitted speed is depended on the size of radius. Furthermore, the minimum requirement for the horizontal radius applies as well as the length of elements. These are summarised in Table 5.3, (Rail Net Denmark, 2013c). When determining the curve radius, the general design rule is to strive towards the largest possible value to achieve a high design speed. Yet, the Danish TER establish an upper limit of 25,000 meters to ease the construction. The lower limit depends on the type of track. In the Danish TER, a distinction is made between main tracks, additional trains routes and sidings. For requirements concerning track along platforms reference is

79 CHAPTER 5 63 made to the TSI INF (Rail Net Denmark, 2013c). In the TSI INF and EN , distinction is only made between normal track and track along platforms. Looking at the values in Table 5.3, it is seen that the Danish TER are more strict regarding curve radii on main track, since the exceptional regulations capture the minimum requirements for both the TSI and EN standards. For track along platforms, it is, however, the TSI, which contains the most restrictive requirement for curve radii. This can result in large expenses for straightening out curves along platforms, when carrying out upgrade and track renewal projects. The larger radii along platforms is established to minimize the size of gap between vehicle and platform, and thereby ensure safe access and egress for passengers (European Committee For Standardization, 2010). Requirements regarding the length of elements are valid for both curves and straight track and serve the purpose of ride comfort. These are not incorporated in the TSI INF, while EN solely states a recommended minimum length of 20 meters. The Danish TER contain in addition to a minimum required length of 20 meters, required values depending on the line speed, see Table 5.3. Thus, TER are more strict than both the TSI INF and the EN In general, these requirements should be met when constructing new lines, or if conditions enable it (Rail Net Denmark, 2013c). Table 5.3: Design rules for circular curves (Rail Net Denmark, 2013c), (Den Europæiske Unions Tidende, 2014a), (European Committee For Standardization, 2010). Rail Net Denmark TSI INF EN Element Req. Std. Exp. Minimum Normal Exp. Horizontal Radius R h [m] Main track 700 (1 300 (1 150 ( Along platforms Length of Elements (2 L k, L s [m] V 200 km/h 0.40V 0.25V 0.20V ( Comments R h 25, 000 L k,s 20 meters 3) Special permission; L k,s = 0.10V can be granted for I < 40 mm in adjacent curve (See subsection 5.2.2) Cant When running in curves, the rolling stock is affected by the gravity and an outward centrifugal force C with a magnitude corresponding to (5.1). C = V2 R h (5.1) This force can result in several undesirable effects such as; passenger discomfort, displacement of wagon load, increased wear, risk of derailment, etc. These undesirable

80 64 NORM FOUNDATION effects can be limited or prevented by reducing the centrifugal force through larger radii, lower speed, or by applying a cant size. (Esveld, 2014) The cant size is defined as the difference in height between the two rails and is constructed by lifting the outer rail in curved track. Figure 5.2 illustrates how the gravity and centrifugal force affect the rolling stock with and without cant. In Figure 5.2a, no cant is applied, and the rolling stock is therefore exposed to a resultant force K, composed by the downward gravity and lateral centrifugal force, referred to as lateral acceleration. By lifting the outer rail, as indicated in Figure 5.2b, it is possible to make the resultant force K coincident with the median plan of the coach (Rail Net Denmark, 2013c). In this case, a total balancing of the centrifugal force entails the greatest possible level of passenger ride comfort, and reduction of wear on the infrastructure at the same time. The cant size achieving the total balance of the centrifugal force is in the terminology of track geometry referred to as the balanced cant h a. The balanced cant is calculated by formula (5.2). h a [mm] = 11.8V2 [km/h] R h [m] (5.2) (a) Without cant (b) With cant Figure 5.2: Gravitational and centrifugal force affecting rolling stock in curves (Nielsen, 2014) Cant Deficiency and Excess In most cases the applied cant h will differ from the balanced cant h a causing an either over- or under balancing of the lateral acceleration for the design speed. This difference is denominated as either cant deficiency I (5.3) or cant excess E (5.4) (Rail Net Denmark, 2013c). I = h a h f or h < h a (5.3) E = h h a f or h > h a (5.4)

81 CHAPTER 5 65 For uniform traffic the balanced cant is optimal to increase ride comfort and reduce wear on the infrastructure. A large mix of slow and fast traffic will, however, entails, that a balanced cant for the top speed will cause considerable cant excess for the slower traffic leading to excess wear on the low rail. Thus, balancing between a certain degree of cant deficiency for fast trains and cant excess for slower trains is needed to even the wear on the high and low rail (Esveld, 2014). Given that the Danish conventional lines are designed to account for both passenger and freight traffic, it is important to consider this balancing. The focus in a speed upgrade project will mainly lie on cant deficiency, considering the aim of increasing the design speed. Yet, requirements for cant excess must also be complied with. As stated in chapter 4, freight trains are assumed to travel with a speed of 100 km/h. This is later used to take cant excess into account, when upgrading the speed profile of the Lille Syd line. Both the Danish TER and EN define the limit of cant excess as 110 mm, see Table 5.6 (Rail Net Denmark, 2013c). The TSI INF does not consider cant excess, but instead it allows for slow freight trains by dividing the requirement for cant deficiency into two categories; operation of rolling stock complying with the TSI regarding locomotive and passenger trains, and regarding freight trains. Limit values for allowed cant deficiency are based on the requirement for safety and passenger comfort. In Table 5.4, the consequences of too high cant deficiency are listed. The comfort requirements are based on a Swedish investigation showing the limit value, where 20 % of the passengers feel discomfort. The additional phenomena are based on rough estimates (Rail Net Denmark, 2013c). Table 5.4: Effects of high cant deficiency, (Rail Net Denmark, 2013c) Cause Discomfort for walking passengers Discomfort for standing passengers Discomfort for seated passengers Displacement of track Derailment Overturning Cant Deficiency 106 mm 118 mm 165 mm > 200 mm > 300 mm > 500 mm The current requirement for cant deficiency is listed in Table 5.5. Compared to the values listed in Table 5.4, it is seen that all requirements are placed within the safety critical values. In general, the Danish TER contain the most restrictive requirements. Complying with the standard regulation entails cant deficiency to be kept below 100 mm, resulting in no discomfort for passengers, regardless of position in the train. Within the exceptional regulation special allowance can be granted for cant deficiency up to 150 mm for speeds above 140 km/h. This can cause discomfort for standing passengers, while seated passengers remain within the defined comfort zone, see Table 5.4 (Rail Net Denmark, 2013c). Looking at the values in EN , cant deficiency up to 183 mm can be accepted in exceptional cases, thus exceeding the comfort zone established in Table 5.4. The TSI INF match more or less the exceptional regulation in TER, yet, with a slightly higher permit-

82 66 NORM FOUNDATION Table 5.5: Design rules for cant deficiency and -excess in plain curved track (Rail Net Denmark, 2013c), (Den Europæiske Unions Tidende, 2014a), (European Committee For Standardization, 2010) Rail Net Denmark TSI INF EN Element Req. Std. Exp. Mixed Passenger Normal Exp. Cant Deficiency, I [mm] = 3.8V2 R (V 140) 130 for freight (2a) 130 (3) 183 (3) (V > 140) (1 153 for pas. (2b) - - Cant Excess, E [mm] - E Comments: and V 250 km/h 2a) valid for rolling stock complying with the TSI regarding freight trains 2b) valid for rolling stock complying with the TSI regarding locomotive and passenger trains 3) valid for for 80 < V 200 km/h ted cant deficiency of 153 mm, compared to 150 mm stated in the TER. The potential of easing the Danish TER to harmonise the TER with the TSI is currently investigated. This is described in a mail correspondence between Gang Liu (Grontmij), Jimi Okstoft (Rail Net Denmark) and Bo Nielsen (Rail Net Denmark), see Appendix E. The mail correspondence deals with questions in relation to the speed profile, when upgrading the railway section between Hobro and Aalborg from 120 to 200 km/h (a project similar to the upgrade of the Lille Syd line). Here, Nielsen states, that he (the Track System Administrator in Rail Net Denmark) works determinedly towards accomplishing this change. The mail is received January 21st, 2015, yet, the change is not incorporated in the latest edition of the Danish TER published 1st of May, An additional edition with a revised version of chapter 2 is therefore expected. Comparing the requirements for cant deficiency to the limit value of passenger discomfort in Table 5.4, it is seen that a cant deficiency of 153 mm still is below the limit, where seated passengers feel discomfort Requirement for Cant Size The discussion on relaxing requirements is continued when establishing the maximum cant size. The maximum cant size accounts for problems, which might arise in case the rolling stock is forced to run slow or stop in a curve. This includes; passenger discomfort, risk of freight trains shifting loads, or derailment, etc. (Esveld, 2014). The current requirements are divided into plain track and track along platforms, where the cant size is limited due to the certainty of the rolling stock standing still (European Committee For Standardization, 2010). A summary of the current requirements are listed in Table 5.6. For normal track, the TSI INF is split into requirements for mixed freight and passenger traffic and solely passenger traffic to account for the member states, which have designated passenger railway lines. This deviation is not incorporated in the Danish TER, since all conventional lines can be used for both freight and passenger transport. In general, it is seen that the Danish TER contain the most strict requirements. The possibility of relaxing the requirements for cant size is dealt with in the previous mentioned mail correspondence (Appendix E). Here, Bo Nielsen states, that due to an increase of permitted cant size from 170 mm to 180 mm in the latest edition of the TSI INF, it has been decided to increase the maximum cant size to 160 mm within the standard regula-

83 CHAPTER 5 67 Table 5.6: Design rules for cant size for plain curved track (Rail Net Denmark, 2013c), (Den Europæiske Unions Tidende, 2014a), (European Committee For Standardization, 2010) Rail Net Denmark TSI INF EN Element Req. Std. Exp. Mixed Passenger Normal Exp. Cant (1, h [mm] Plain Track = 8V2 R Track at platforms Comments: 1) h R applies for curved track with small radii to eliminate the risk of derailment tions. This change was undergoing a final assessment in the moment of the mail exchange ( ). Thereby, Nielsen grants full endorsement to the Hobro-Aalborg project on applying a cant size of 160 mm. In relation hereto, Nielsen states, that increased cant size is preferred above increased cant deficiency on the prerequisite, that the requirement concerning cant excess is met. This change is also not incorporated in the current edition of TER. Again, it is assumed that an additional edition of the TER is on the way, and the cant size can therefore be designed up to 160 mm in present project Requirements in Turnouts The same conditions to non-compensated lateral acceleration, as described for plain track curves, apply for turnouts placed in curves. However, the dynamic effect, and thus the overall lateral acceleration, is much greater due to a relatively poor track geometry in turnouts. This entails that the cant deficiency and the cant size are kept lower than in plain curves (Esveld, 2014). The Danish TER distinguish between cant deficiency in inside curved turnouts and contra flexure curved turnouts (turnouts bent in opposite direction), while EN describes different types of turnouts. The TSI INF contains no specification regarding turnouts placed in curves. The requirements are listed in Table 5.7. Here, it is seen that same requirements for the maximum cant size apply for the Danish TER and the EN standard. Regarding cant deficiency in turnouts, EN lists different values depending on the line speed and the type of turnout (e.g. fixed crossings, moveable parts, etc.). For limiting values, reference is made to the standard regarding plain track. By comparing the EN recommended values to the Danish TER standard regulation it is, however, clear, that the Danish TER once again contain the most strict requirements. Turnouts are preferably placed on straight track to avoid bending the standard components, leading to an increased cost during purchasing and maintenance work. If not possible, turnouts can be placed in curves with large radii and a maximum speed of 200 km/h according to the Danish TER (Rail Net Denmark, 2013c).

84 68 NORM FOUNDATION Table 5.7: Design rules for cant size and deficiency in turnouts (Rail Net Denmark, 2013c). Rail Net Denmark TSI INF EN Element Req. Std. Exp. Min. Normal Exp. Cant in turnouts, h [mm] Cant Deficiency I [mm] in turnouts Inside curved As main track Contra flexure curved (V 100 km/h) (1 (V 160) 120 (1 (V > 160 I 80 (V > 100km/h) Comments: 1) Recommended values for crossings with moveable parts. Limiting values depend on the specific design and is left to the Infrastructure Manager to decide. The radius of the branch equals the current curve radius, while the radius of the diverted branch is determined by (5.5). 1 R u = 1 R b ± 1 (5.5) R m Here, R m is the main line radius, and R b is the original radius of the branch. In an inside curved turnout the denominator is added, and in an contra flexure curved turnout the denominator is subtracted (Rail Net Denmark, 2013c). In a contra flexure curved turnout, the cant size shall additionally comply with (5.6) to account for the negative cant size, which appears when the diverted branch is bent in the opposite direction of the curve, according to the Danish TER. h I 11.8 V2 v R u (5.6) V v is the speed in the diverted branch, and R u is the resulting radius of the bended branch (Rail Net Denmark, 2013c) Transition Curves Transition curves are placed in the horizontal alignment to obtain a smooth variation of the curvature between straight elements and curves or between two curves. Most often they are constructed as clothoids with a constant super elevation ramp used for changing the cant size, and thereby causing a constant change of the lateral acceleration. However, discomfort can be caused by short transition curves, if the rate of change of cant abruptly terminates at the end of the curve. In such case, it can be preferred to construct the transition curve as a forth degree parabola with a s-shaped super elevation ramp. The rate of change of cant size is zero in both ends and twice as large in the center compared to the corresponding clothoid. Controlling the position of forth degree parabolas is, however, far more difficult than with clothoids. The majority of transition curves are therefore shaped as clothoids, while the forth degree parabolas are only applied in exceptional circumstances of short transition curves to achieve better ride comfort. Adjacent curves or curves and straight track shall always be combined with transition curves, in accordance to the Danish standard regulation in TER (Rail Net Denmark, 2013c). The permitted speed in transition curves is evaluated based on the three criteria listed below. In all cases, it is assumed, that the transition curve is constructed as a clothoid

85 CHAPTER 5 69 with coincident linear super elevation ramp. L is the length of the transition curve in meters, and V is the speed in km/h. Rate of change of cant deficiency as function of time di I[mm] V[km/h] [mm/s] = dt 3.6 L[m] (5.7) is a measurement for the change of the lateral acceleration in unit of time. The passengers experience this force as a jolt and it serves exclusively a purpose of passenger comfort (Rail Net Denmark, 2013c). I is the difference in cant deficiency between the connection points of the transition curve. Rate of change of cant as function of time dh h[mm] V[km/h] [mm/s] = dt 3.6 L[m] (5.8) is a measurement for the change of cant size in unit of time. h is the change in cant size between the connection points of the transition curve. Rate of change of cant as function of length dh h[mm] [mm/m] = dl L[m] (5.9) describes the gradient of the super elevation ramp. The rate of change of cant deficiency as function of time (5.7) serves exclusively a purpose of passenger comfort. In the aforementioned Swedish investigation of passenger comfort, it was found that 20 % of passengers feel discomfort at the values listed in Table 5.8 (Rail Net Denmark, 2013c). By comparing these values to the associated requirements listed in Table 5.9, it is seen that neither TER nor EN compromise the ride comfort for seated passengers. Walking passengers will, however, experience discomfort when passing transition curves within the standard regulations. The TSI INF contains no specifications on the composition of transition curves. Crosswise comparison of the requirements show a large correlation, with an exception of the EN standard permits vales of 100 mm/s for di dt within exceptional regulations, whereas the Danish TER maximum permits 90 mm/s. Table 5.8: Limit values for when 20 % of passengers feel discomfort at high values of rate of change of cant deficiency as function of time, (Rail Net Denmark, 2013c) Cause Discomfort for walking passengers Discomfort for standing passengers Discomfort for seated passengers di dt 41 mm/s 77 mm/s 112 mm/s The super elevation ramp is evaluated based on both the rate of change of cant as function of time (5.8) and the rate of change of cant as function of length (5.9). The associated requirements serve both the purpose of passenger comfort and safety. Therefore applies

86 70 NORM FOUNDATION Table 5.9: Design rules for transition curves(rail Net Denmark, 2013c). Rail Net Denmark TSI INF EN Element Req. Std. Exp. Design Normal Exp. Rate of change of cant deficiency as function of time, di dt [mm/s] V 200km/h (For I 168) 90 (For I > 168) Rate of change of cant as function of time, dh dt [mm/s] V 200km/h ( (For I 168) (2 ) Rate of change of cant as function of length, dh dl [mm/s] All speeds Comments: 1) For I 150 mm and di dt 2) For I 153 mm and di dt dh 70 mm/s can dt dh 70 mm/s can dt 85 mm/s 85 mm/s 50 (For I > 168) more toughen requirement for dh dt compared to di dt, which exclusively serves passenger comfort. Crosswise comparison of the requirements in Table 5.9, shows large correlation between EN and the Danish TER. This is a result of the EN standards being incorporated in the Danish TER. Yet, the TER contain slightly more restrictive requirements for rate of change of cant as function of length. The differentiation between note 2 and 3 is caused by the previously mentioned upward revision of permitted cant deficiency in the latest edition of the TSI INF Abrupt Change of Cant Deficiency As stated above, curves and straight track or curves and curves shall always be combined with a transition curve, to comply with the TER standard regulations. However, within exceptional regulations, the transition curves can be omitted (Rail Net Denmark, 2013c). This can be necessary at stations containing several minor curves or if the variation of radii is very limited in adjacent curves. (European Committee For Standardization, 2009) This special case results in an abrupt change of cant deficiency. In such cases, a fictive length of 20 meters is applied to determine the rate of change of cant deficiency. This virtual transition curve is based upon the assumption, that cant deficiency is changed over a length equal to the distance between the bogie centres of a characteristic train (European Committee For Standardization, 2009). The values are thereby not comparable to the requirements listed in Table 5.9. Instead, the Danish TER uses the requirement defined in (5.10). di dt = IV 135mm/s for V 200 km/h (5.10) In EN , the requirements for abrupt change of cant deficiency in plain track are based on the magnitude for change of cant deficiency without consideration of virtual transition. The requirements are listed in Table The requirements of the EN and the Danish TER are incomparable due to their different definitions.

87 CHAPTER 5 71 Table 5.10: Recommended values of abrupt change of cant deficiency in plain track, (European Committee For Standardization, 2009) EN Speed Interval I lim V 70 km/h 50 mm 70 < V 170 km/h 40 mms 170 < V 200 km/h 30 mm Straight Track Between Curves Two adjacent curves containing two separate transition curves is divided by a straight track element. For the straight track applies same requirements for the element length as stated for curved track in subsection If the straight track is very short or none-existing is the curves adjacent. Curves bend in same direction should be connected by shared transitions curve, while opposite curves should be adjacent to allow for the cant size is changed in different directions (Rail Net Denmark, 2013c) Longitudinal Profile Up until now, requirements regarding elements in the horizontal alignment have been covered. Focus will now shift to the vertical alignment, referred to as the longitudinal profile. This alignment contains elements of straight lines and circular curves. The requirements for designing the longitudinal profile includes; gradients of straight elements and radii in vertical curves Gradients Gradients are used to change the track level. This can not be steeper than permitted by the maximum available adhesion force between the driven wheels and the rails (Esveld, 2014). The requirements for the TER and the TSI INF are listed in Table EN contains no specification on gradients. The requirements stated in the TSI INF are only valid for newly constructed lines, because changing the gradient of existing lines is very costly. On sections where the rolling stock typically stands still (along platforms or stabling yards), the requirements are more restrictive than on open lines. This is caused by the fact, that large gradients require more energy to accelerate and decelerate due to adhesion. At the same time causes this an increase of the wear on brakes and rails, resulting in larger operation and maintenance costs (Esveld, 2014). The TSI requirements have earlier, more or less, been incorporated in the exceptional regulations of the Danish TER. In the latest revision of the TSI INF, the requirements for highspeed and conventional rail systems are combined. This entails, that reference to HS TSI INF and CR TSI INF respectively in the Danish TER becomes ineffective. This makes

88 72 NORM FOUNDATION it difficult to outline the exceptional regulations in the TER. Therefore, only the requirements valid for the track included in the former CR TSI regulations are listed. The requirement for gradient is listed in Table The largest difference in the Danish TER and the TSI requirement for gradient is, that TSI permit gradient of 35.0 o / oo, whereas the Danish TER maximum permits 25.0 o / oo on short sections. This is caused by the TSI distinguish between passenger lines and mixed traffic, while Rail Net Denmark only considers mixed traffic. The 35.0 o / oo is thus dedicated passenger lines. Furthermore shows the requirement that larger gradients can be permitted on short sections, hence 25.0 o / oo applies only for line sections up to 0.5 km, while 15.6 o / oo is permitted for line section up to 3.0 km. In relation to the construction cost, it is advantageous to allow larger gradients in order to limit the reshaping of existing terrain, when constructing the top of sub ballast. On the other hand, large gradients entail an increase in operation cost, due to the reasons described above. Permission to applying exceptional regulations require, therefore, a detailed analysis in relation to speed, train length and braking conditions for all types of rolling stock intended to run on the infrastructure. Table 5.11: Design rules of gradients in the longitudinal profile (Rail Net Denmark, 2013c), (Den Europæiske Unions Tidende, 2014a) Rail Net Denmark TSI INF Element Req. Std. Exp. Design New Lines Gradient p[ o / oo ] Stabling tracks Tracks along platforms Other main tracks, for l 3.0 km 1) 35.0 reliefs track or sidings 2) 25.0 for l 0.5 km Comments: 1) Only permitted where the rolling stock neither accelerates nor decelerates during normal operation. 2) For new P1 lines dedicated to passenger traffic, which furthermore complies with; (a) p 25 for more than 10 km in the average profile, (b) l 6 km Vertical Curves Vertical curves of suitable radii are inserted in the longitudinal profile when changing the gradients to ensure safety and comfort (Esveld, 2014). According to the Danish TER, vertical curves with a minimum length of 20 meters are required, when the gradient is changed more than 1 o / oo, while smaller changes can be constructed as direct breaks of straight line elements (Rail Net Denmark, 2013c). This is, again, more restrictive than the EN , where the change shall exceed 2 o / oo, before a vertical curve is required for railway lines with a design speed up to 230 km/h (European Committee For Standardization, 2010).

89 CHAPTER 5 73 The minimum requirements for the vertical curve radii are listed in Table The Danish TER requires, that the size of the vertical radius depends on the design speed and is dissociated at 200 km/h. Since current project only focuses on a speed upgrade up to 200 km/h, initiates on upgrading the vertical alignment are assumed to be obsolete. (Rail Net Denmark, 2013c) The requirements listed in the TSI INF and the exceptional regulations in EN distinguish between vertical curves placed in crest and hollow curves, where the most restrictive requirements are found for hollow curves. In general, the Danish TER contain the most restrictive requirements. This is specially valid when comparing with the TSI INF. A possible cause for this large difference could be, that the Danish curvature in general is relative plane compared to some of the other member European states. The need for small vertical curves are therefore found obsolete in the Danish railway sector. Table 5.12: Design rules of vertical curve radii (Rail Net Denmark, 2013c), (Den Europæiske Unions Tidende, 2014a), (European Committee For Standardization, 2010) Rail Net Denmark TSI INF (1 EN (1 Element Req. Std. Exp. Crest Hollows Normal Exp. Vertical Radius R L [m] V 200 km/h 1.0V V V a) 900 3a) 0.35V V 2 (crest) 10, 000-5, 000 (2 2, V 2 (hollow) 40, 000 Comments: 1) Without consideration of line speed 2) 2,000 for existing track Turnouts Vertical curves shall be placed outside turnouts and super elevation connecting ramps, to the possible extent. Otherwise, the vertical radius shall be increased as far as possible, (Rail Net Denmark, 2013c), (European Committee For Standardization, 2009). The Danish TER states, that the exceptional regulations in Table 5.12 does not apply for turnouts placed in ascending breaks, (Rail Net Denmark, 2013c). In the EN , it is stated that turnouts can be placed in both descending and ascending breaks, when complying with the following requirements (European Committee For Standardization, 2009); Normal; R v 5, 000 meter for hollow and R v 3, 000 meter for crest Exceptional; R v 3, 000 meter for hollow and R v 2, 000 meter for crest The TSI INF contains no specification regarding the placement of turnouts in the longitudinal profile. When analysing different solutions for the upgrade of the Lille Syd line, the focus will lie upon the horizontal alignment. Requirements in relation to the gradient and minimum vertical curve radii are therefore not specifically applied in the further analysis, given

90 74 NORM FOUNDATION that no longitudinal profiles will be conducted. However, the design rules are clarified to understand the connection between the projected alignment and the existing terrain, and the large amount of earth work caused by this. When constructing the longitudinal profile, the aim is to achieve balance between cut and fill to limit the handling of sub soil, which is a heavy post in the construction cost Summary of Track Geometry with Respect to the Lille Syd Line Preceding section deals with the Danish TER, the TSI INF, and the EN standards valid for the geometrical design of the track. A comparison of the different requirements are conducted, with the purpose of using different requirements in the further analysis of the upgrade of the Lille Syd line, when speed profiles are conducted. In general the TER are more restrictive than the remaining requirements, and a relaxation might result in construction cost savings. Solely, the horizontal alignment is considered in the further analysis, as no longitudinal profiles are designed. This conclusion was the same in (Jensen, 2012) 5.3 Ballast Profile Current section describes the Danish requirements for the ballast profile. The TSI INF does not contain any requirements for the structure of the ballast profile, hence no references are made according to European requirements. A deviation in the Danish requirements is made according to speed, and type of project; upgrade, renewal, or construction of new lines. The section contains one overall passage concerning the components within the ballast profile and a summary regarding the parameters relevant for the Lille Syd upgrade is provided in the end Components within the Ballast Profile The ballasted track structure is the most commonly used railway structure nowadays. The main purpose of the superstructure is to retain the track location and transfer energy from the rolling stock to the substructure. This requires it to be able to absorb and distribute forces, so the top of subsoil can handle the tension from the rolling stock. In addition, it must ensure appropriate drainage of the track bed to avoid settlement as a consequence of periods of rain or frost (Fongemie, 2013). Figure 5.3: Typical cross section of ballast profile, (Rail Net Denmark, 2014a)

91 CHAPTER 5 75 Danish railway lines are traditionally constructed as ballasted track. Here the superstructure is composed by a sub layer of compressed gravel and an upper layer of ballast stones. The shape of the ballast profile is decisive for, how the trackbed handles forces from the rolling stock. The Danish Railway Norm BN1-6-5 describes the requirement for the design of the ballast profile (Rail Net Denmark, 2014a). Here, it is stated, that all sections included in a TEN-corridor must comply with the relevant TSI-requirements. The TSI INF documentation contains no specification on the shape of the ballast profile. Instead, it defines requirements for the forces the track bed shall be able to handle (Den Europæiske Unions Tidende, 2014a). Rail Net Denmark assesses, that these are incorporated in BN1-6-5 norm, yet, a NoBo must confirm this statement to enforce a given project. Based on Rail Net Denmark s assessment, the TSI requirements will not be further addressed in relation to the ballast profile. Instead, focus will lie upon the Danish ballast profile illustrated in Figure 5.3. From Figure 5.3 it is seen, that the top of sub soil and sub ballast are constructed with an inclination x p, which ensures drainage of the track bed towards either ditch or trough. The thickness of ballast B t and sub ballast U t is determined by vertical loads and the increased dynamic loads, which occur at high speed. Another purpose of the sub ballast is to protect the ballast layer from soiling from the sub soil, which can impair the ballast layer s ability to absorb tensions or disable drainage of the track bed. The widths of ballast shoulders B sk, track bench P bb, and the slope of the ballast a are constructed to avoid lateral track displacement in curves, as large widths and small slopes result in larger lateral resistance. (Jacobsen and Meyer, 2011) Rail Net Denmark s requirements for the physical appearance of these elements are described in BN1-6-5; Ballast track (Rail Net Denmark, 2014a). Here, the BN-1 minimum requirements and additional stricter BN-2 requirements are used for track renewal, upgrade and construction of new lines. Deviation from BN-2 regulations require acceptance by the technical system administrator (TSA) at Rail Net Denmark. In addition, the requirements depend on the line speed, since a higher speed entails larger tensions in the track bed. Rail Net Denmark operates with three speed intervals; Design speed V 160 km/h Design speed 160 < V 200 km/h Design speed 200 < V 250 km/h Since no Danish railway lines are built with a design speed above 250 km/h, no requirements exist for larger speed intervals. Because this project maximum investigate speed upgrade up to 200 km/h, the latter category is left out of context. The requirement is summaries in Table In general, consistency is shown between BN-1 and BN-2 requirements for upgrade and renewal of lines, which only differ in relation to thickness of the sub ballast. The minimum BN-1 requirement thereby enables reduction of the width of the track bed. This could for instance be desirable to obtain proper clearance below bridges.

92 76 NORM FOUNDATION Table 5.13: Requirements for ballast profile for Danish railway lines with maximum axle load of 22.5 ton on concrete sleepers. Special condition can cause relaxed or toughen requirements, for these cases see (Rail Net Denmark, 2014a). BN-1 Renewal, Minimum Requirements BN-1 Existing Profile Upgrade and New Lines Element V < V 200 V < V 200 Width of Ballast Shoulders, B sk [m] Slope a Ballast Thickness B t [m] Sub Ballast Thickness U t [m] Track Bench P bb [m] depended on size of cant ( (1 Inclination of trackbed X p [ o / oo ] Normal Requirements BN-2 Renewal, Upgrade BN-2 New Lines Element V < V 200 V < V 200 Width of Ballast Shoulders, B sk [m] Slope a Ballast Thickness B t [m] Sub Ballast Thickness U t [m] Track Bench P bb [m] depended on size of cant ( ( ( Inclination of trackbed X p [ o / oo ] Comments: 1) The track bench shall additionally be extended 0.15 m and 0.30 m respectively, in curves with a cant size of 5-80 mm and mm 2) Sites with existing track bench of minimum 2.75 m can be retained. For renewal and upgrading projects, this is not valid in turnouts. 3) Applies for the speed interval 120 < V 160. For V 120 km/h is P bb = Upgrading When upgrading the line speed from 120 km/h to 160 km/h applies same requirements for the ballast profile and no large changes are required. According to foot note three in Table 5.13, will changes to the cant size from 0 to the interval 5-80 mm or trigger extension of the track bench. This ensures that the profile can resist the lateral track displacement. Awareness it put on this when upgrading the Lille Syd line. Furthermore will changes to the cant size entail minor modification to the ballast layer, to ensure the ballast shoulders are parallel with the sleeper. This will require ballast supplement. When upgrading the line speed to 200 km/h should the thickness of subballast be increased 100 mm, according to Table This is due to the before mentioned increased dynamic loads when increasing the speed.

93 CHAPTER New track along existing New track shall comply with regulations for new lines, hence the ballast thickness is 350 mm and the track bench is 3800 mm see Table Furthermore applies requirements for placements of new track along existing within BN1-6-5, so the construction of new track never compromise the level of drainage of the existing track. To ensure this states BN1-6-5 that the top of sub ballast and top of sub soil shall be placed a minimum of 0.10 meters below the corresponding level of existing track, see Figure 5.4. This applies when the type of sub ballast in both existing and new track is the same type. Figure 5.4: Cross section for extension of new track along existing in case of same sub ballast for both new and existing track (Rail Net Denmark, 2014a) Summary of Ballast Profile with Respect to the Lille Syd Line Preceding section presents the requirements for the structure of the ballast profile. The requirements are divided according to speed, and type of design; renewal, upgrade and construction of new line. When upgrading the Lille Syd line to 160 km/h and 200 km/h respectively, the ballast profile of the existing track must comply with the requirements for upgrades in Table This can possible lead to an acquirement of additional sub ballast or extension of the track bench at places, where the cant size is changed. When expanding the infrastructure from single to double track, the new track is categorised as new line, and the ballast profile shall therefore comply with the regulations for new lines. Furthermore, an additional requirement applies for the placement of the new track along existing, so the new track never compromise the level of drainage. This requirement affects the design of the longitudinal profile, which is not treated within this analysis. The associated requirements can therefore be found in (Rail Net Denmark, 2014a). 5.4 Track Center Distance Present section introduces the Danish requirements for the track center distances described in BN In relation to this, a comparison to the TSI requirements is made. The section also includes a description of the determination of fouling points, this is, however, not described in the TSI INF. The section is divided in two; requirements for the track center distance and determination of fouling points. A summary of the valid

94 78 NORM FOUNDATION track center distances for the Lille Syd line is provided in the end of the section. The fixation of the track center distance between two tracks ensures avoidance of collision between trains. The distance is established to take aerodynamic impacts of high speeds into consideration, as well as the location of signals, signs, and personnel present in case of shunting. On straight track, the distance is measured as the perpendicular distance between the centerlines f 0, see Figure 5.5a. In curves a supplement is added depending on the radius and whether both tracks are curved, Figure 5.5b, or one is straight, Figure 5.5a. These distances are, however, not valid if objects are placed between the tracks. For radii smaller than 1500 m a supplement [e] is added depending on the radii, the supplement can be found in BN (a) Two straight tracks (b) A straight and a curved track (c) Two curved tracks Figure 5.5: Determination of track center distances according to track geometry (Rail Net Denmark, 2008) The track center distances are differentiated according to type of project carried out; upgrade, renewal and construction of new lines, and the minimum distance for tracks in service. Furthermore, the distances are differentiated according to speed, where a larger speed requires a larger distance. The minimum center track distances used for operation is given in Table For upgrades and renewal projects, the minimum track center distances are in general 100 mm larger than the minimum distances for tracks in service. These minimum distances can only be used with permission from technical system administrator in Rail Net Denmark. The listed BN2 requirements are general requirements, which shall be complied with, in upgrading and track renewal projects as well as construction of new lines. The only difference between BN1 and BN2 norms are the larger difference distances for speeds below 120 km/h. In addition hereto, the requirements at stations state a distance of at least 4750 mm. (Rail Net Denmark, 2008) The TSI INF determines the track center distance according to the traffic categories, see Table 5.2, where the minimum nominal track distance is 3800 mm (Den Europæiske Unions Tidende, 2014a). Besides stating the minimum distance, the TSI INF refers to the EN for placement of adjacent tracks. The EN standard determines the limit distance between centres according to the widening (European committee for standardization, 2013). This is, however, not investigated further in depth, as compliance with the Danish regulations also ensures, that the TSI requirement is met Fouling points The Danish norm BN also considers the location of fouling points. Fouling points are markers placed at the exact limit, where two trains will not be clear of each other. The distance from the centerline of the tracks to the fouling point is 1 2 f 0. Depending on

95 CHAPTER 5 79 Table 5.14: Minimum track center distances for conventional railway lines (Rail Net Denmark, 2008), (Den Europæiske Unions Tidende, 2014a) V 120km/h V 160km/h 160 < V 200km/h Minimum in service (BN1) 4000 mm 4150 mm 4400 mm Upgrades and track renewal (BN1) Upgrades, track renewal and new lines (BN2) 4100 mm 4250 mm 4500 mm 4250 mm 4250 mm 4500 mm TSI INF mm whether the respective tracks are curved with a radius R < 1500 meters or not, a supplement [e] is added, see Figure 5.6. The track center distance used for placement of fouling points is the BN1 distances according to upgrades and track renewal, see Table A circular tolerance area of 100 mm within the placement of fouling points is allowed. (a) Two straight tracks (b) A straight and a curved track (c) Two curved tracks Figure 5.6: Placement of fouling points according to track layout Summary of Track Center Distance with Respect to the Lille Syd Line Preceding section treats the relevant requirements for the track center distances for lines in service, as well as for projects concerning upgrades, renewal and construction of new lines. Furthermore, the requirements are differentiated according to speed, where higher speeds require larger distances. By complying with the Danish requirements, the TSI requirements are met. Considering the line upgrade of the Lille Syd line, a track center distance of 4250 mm is valid for speeds up to 160 km/h, whereas for speeds up to 200 km/h, a distance of 4500 mm is required. Thus, construction of the second track requires a distance of 4500 mm. 5.5 Structure Gauge Current section presents the requirements for structure gauges valid for electrified lines in Denmark. Furthermore, a presentation of the TSI requirements for the structure loads is provided, as well as a comparison of structure loads valid for the German-Scandinavian

96 80 NORM FOUNDATION corridor. The section is divided in two; Danish requirements and European requirements, and a summary of the valid structure gauges and loads relevant for the Lille Syd line is provided in the end of the section Danish Requirements for Structure Gauge The structure gauge defines the limit from where outer objects such as bridges, signals, and platforms can be placed, without interfering with the rolling stock. The Danish requirements for the structure gauge are based on international regulations (UIC fiche 505), allowing approved international trains to run on Danish lines. The structure gauges are determined according to the kinematic movement of the wagon to ensure that it does not conflict with outer objects during movement. The track center distances, found in table 5.14 above, define the structure gauge on main lines, except S-lines. The reference for determining the structure gauge is defined according to the level through the top of the rails (SO) and the profile center, see Figure 5.7. In curves, either horizontal or vertical, the structure gauge must be increased according to table 5.15 below. For full table see Rail Net Denmark s regulations on Fritrumsprofiler. (Rail Net Denmark, 2014c) (a) Straight track (b) Curved track Figure 5.7: Structure gauge defined according to track geometry, (Rail Net Denmark, 2014c) Table 5.15: Supplement for structure gauge in horizontal and vertical curves, (Rail Net Denmark, 2014c) Horizontal Lateral Horizontal Radius Additional value Vertical Radius Additional values m 5 mm m 5 mm m 10 mm m 10 mm m 15 mm m 15 mm Considering the placement of objects, a supplement is added in order to consider future maintenance. 100 mm in height and 50 mm in width is added for track renewal and new lines. The structure gauge is, furthermore, dependent on whether the line is electrified or not. The requirements for non-electrified lines are not covered, as electrification of the Lille Syd line is currently carried out. For electrified lines, an overhead catenary system must be considered. A distinction is made between; existing electrified lines and new

97 CHAPTER 5 81 electrification of lines. For lines to be electrified, a distinction is made between; lines with existing bridges built before 2012, and lines with new bridges, see Figure 5.8. The EA profile is used on open lines, whereas the EBa profile is used at bridges and constructions. The difference between the two profiles is the layout of the EBa profile. For lines maintaining existing bridges, a smaller EBa profile is permitted for speeds between 80 and 160 km/h. In general, Figure 5.8 is valid for speeds between 80 and 200 km/h, whereas lines specially used for slower running trains must comply with a different profile. Specific regulations are valid for structure gauges at stations, see section 5.6. Figure 5.8 focus on the upper part of the structure gauge, the area adjacent to the tracks is designed differently, for further details on this see Rail Net Denmark s Fritrumsprofiler. (Rail Net Denmark, 2014c) (a) New electrification with existing bridges (b) New electrification with new bridges Figure 5.8: Structure Gauge for newly electrified lines, valid for 80 < V 200km/h, (Rail Net Denmark, 2014c) International Requirements for Structure Load In order to comply with the European Commission s goal of having and interoperable infrastructure, it is of great importance that the European structure gauges are harmonised. In Denmark the regulations focus mainly concern structure gauges, whereas the international regulations focus on loading gauges. The loading gauge is a measurement concerning the largest possible cross section, that can be loaded onto a wagon (Boysen, 2014). Therefore, the TSI requirements do not define a specific structure gauge, however, this can be determined by adding 100 mm to the loading gauge (Liu and Balsby, 2015). The European Railway Agency has developed a set of standard TSI gauges valid for the continental part of Europe to achieve interoperability (Den Europæiske Unions Tidende, 2014a). The TSI INF refers to the EN :2013 (E) norm, for determination of different international loading gauges. The general static profile width in continental Europe is 3.15 m, which also is valid for the GC profile according to the freight traffic category F1 in Table 5.2. The EN standard appoints the G1 profile (3.15 m x 4.28 m) as the general profile used for international transport, whereas

98 82 NORM FOUNDATION the GA, GB and GC profiles originally were applicable for container transport. The EN recommends the GC profile used on the core TEN-T lines (European committee for standardization, 2013). In Denmark, the Oresund link accepts the GC profile and is prepared for the Swedish profile SE-C, which also is valid for the Fehmarn Belt link. The new line Copenhagen-Ringsted accepts the loading gauge GC. (Boysen, 2014) Considering the location of Denmark, the Swedish and German requirements for structure gauges are relevant to examine. Figure 5.9 below illustrates the different static loading gauges relevant in the German-Scandinavian corridor. The German system allows loading gauge G2, and so does the Danish system. The Swedish SE-A loading gauge is similar is height, however, a larger width is allowed. The Swedish standard gauge is the SE-A gauge, yet, the southern part of Sweden and all new lines accept the SE-C profile. Table 5.16 below summarises the main loading gauges accepted in Denmark, Sweden, Germany and Norway as well as the European standards.the dimensions are based on maximum widths and heights however, the profiles do vary in shape, whereas most of them have pitched tops, see figure 5.9 (Boysen, 2014). Table 5.16: Static loading gauges accepted in Europe [m x m], (European committee for standardization, 2013), (Boysen, 2014) TSI/EN Denmark Germany Sweden Norway G1: 3.15 x 4.28 G2: 3.15 x 4.65 SE-A: 3.40 G2: 3.15 x 4.65 Oresund: GC: 3.15 x 4.65 GA: 3.15 x 4.32 SE-B: 3.40 x 4.30 U: 3.40 x 4.45 standard (rectangular) Prepared load units. for C: All 3.60 of these x 4.83gauges define the permissible static dimensions, i.e. not including dynamic movements. Copenhagen-Ringsted: Fehmarn Belt: GB: 3.15 x 4.32 SE-C: 3.60 x 4.83 EXISTING LOADING GC: GAUGES 3.15 x 4.65 C: 3.60 x 4.83 GC: The 3.15 height x 4.65 of the static DE3: gauge 3.29 x reference 4.68 profiles in - Continental Europe - is up to 4.65 m above top of rail for gauges G2 and GC, as well as for the Swedish gauge SE-A, and 4.83 m for gauge C. The width is In3.15 general, m for all the of the Swedish individual system static gauge SE-Areference accepts profiles rollinguic stock of (G1), sizesg2, G1, GA, GAGB, and GB1, GB, GB2, and GCGB-G6 in some and GC parts (European of Sweden. Commission In addition, 2006). the In Sweden Danishand andnorway, German the system static gauge do not reference accept profile width is 3.40 m for loading gauges SE-A, SE-B and NO-U, and 3.60 m for loading gauge C. the SE-A dimensions due to the width. The Norwegian system is compatible with the Existing and planned loading gauges and intermodal gauges in northern Europe are shown in Figure Swedish SE-A profile, hence accepts larger dimensions than Danish and German system. 4, together with the nominal height of electrical overhead lines. The cut-away top corners of the traditional gauges G2, GC, SE-A, NO-M and NO-U should be noticed. Figure 5.9: Loading Gauges in the German-Scandinavian corridor, based on (Boysen, 2014) Figure 4. Existing and planned loading gauges and intermodal gauges in northern Europe According to the traffic categories valid for the Lille Syd line listed in Table 5.2, the structure Representing gauge for the passenger many types transport of loads that isare the rectangular DE3 profile, in projection, and the including GC profile boxed and is valid palleted for goods as well as rolls of paper standing upright, the useful cross sectional area of a loading gauge is measured as the largest rectangular section that can be inscribed within the gauge and above the standard floor level of most wagons, 1.2 m above top of rail, see Figure Useful cross section (m2) 11, ,068

99 CHAPTER 5 83 freight transport. The profile DE3 is determined in accordance with the European gauges G2 and GB, and is assessed to be applicable in parts of the European network (European committee for standardization, 2013). In a project concerning the speed upgrade of the line Hobro-Aaalborg, Grontmij has sketched the differences in the G2 and DE3 profile in relation to the Danish EBa structure gauge (Liu and Balsby, 2015), see Figure The red lines illustrate the DE3 profile, while the green lines represent the G2 profile. Based on this, it can be concluded that the DE3 profile is a bit larger than the G2 profile, however, the general EBa structure gauge will comply with both. In the upgrade of the Lille Syd line, the EBa profile is assessed applicable at bridges and constructions, while the EA profile is applicable at open lines. Figure 5.10: The differences in structure clearance, (Liu and Balsby, 2015) Summary of Structure Gauge with Respect to the Lille Syd Line Preceding section deals with the Danish requirements for structure gauges in relation to electrified lines. The TSI requirements consider structure loads instead of structure gauges. A comparison of the structure loads valid in the German-Scandinavian corridor showed, that Sweden and Norway permit wider structure loads compared to Denmark and Germany. However, the Oresund link is constructed to accommodate these wider loads, and so is the Fehmarn Belt link. The upgrade of the Lille Syd line requires the GC and DE3 profile to be met. The implementation of the second track on the line will require reconstruction of all road bridges. By applying the general EA on profile open line, and the EBa profile at bridges and constructions, both the GC and DE3 profile is accommodated. This is valid for both the speed upgrade and the construction of the second track.

100 84 NORM FOUNDATION 5.6 Platforms Present section introduces the Danish and European requirements for location and dimension of platforms. The Danish requirements are consistent with the TSI requirements according to platform heights, however, platforms constructed before 1979 do not always comply with these. No Danish requirements are stated according to the platform lengths, but specific lengths are required in the TSI INF. Finally, requirements for safety zones and open spaces have minor deviation. The section is divided in two considering the Danish and TSI requirements for the platform location and height, followed by the safety zones and open spaces. A summary of the section is provided in the end with respect to the Lille Syd line Danish Requirements for Platforms The Danish track norm BN engages in the location of platforms. The norm is valid for platforms with heights below 1250 mm. The location of the platform according to the track is determined based on the nearest track. The front edge of the platform is placed in relation to the upper edge of the nearest track (SO). The nominal platform distance depends on the track geometry; straight or curved, and on the platform height. In curved track the outer rail is often raised, which effects the determination of the nominal distance depending on, whether the platform is place in the inner or outer curve. The distance as well as tolerances for the platform height and placement of platforms in operation are indicated in the norm. (Rail Net Denmark, 2006c). (a) Placement of platform along curved track (b) Placement of platform along straight track Figure 5.11: Nominal placement of platforms along track (Rail Net Denmark, 2006c) The standard platform height for intercity and regional lines is 550 mm. The standard height was adopted in 1979 due to large variations in heights for earlier constructed platforms. Therefore, the platform height in Denmark varies between 260 and 760 mm. In general, the expenses of reconstructing platforms according to the standards are high. However, Rail Net Denmark strives to comply with the standard height in case of larger modifications of the platform facility. (Rutter, 2012) TSI Requirements for Platforms According to the TSI requirements, the nominal platform height is required to be either 550 or 760 mm above the SO-plan. The distance between the track and the platform is determined to allow service of the G1 profile (3.15 m x 4.28 m), see section 5.5, with a tolerance of maximum 50 mm (Den Europæiske Unions Tidende, 2014a). In general, the

101 CHAPTER 5 85 distance between the platform and coach is to be minimised. The track along the platform is in general required to be straight and can without exception, be placed in a curve with radius below 300 m. However, no exact requirements are established for existing track along new, upgraded or renewed constructed platforms. (Den Europæiske Unions Tidende, 2014a) As mentioned in section 5.5, the TSI requirements are formulated for railway lines according to the traffic type. The line category for the Lille Syd line is determined to be P3 for passenger traffic and F1 for freight traffic. The category for passenger traffic states, that the operational platform length for P3 is meters. However, in case of large geographical, urban area or environmental challenges part of a line is allowed stray from the parameters, see Table 5.2. (Den Europæiske Unions Tidende, 2014a) In general, during an exchange of parts in a subsystem due to maintenance work, the TSIs are strived to be followed. It is here to be noted, that before a system can fully comply with the TSI requirements, all parts of the system must be TSI consistent. In addition hereto, deviation from the TSI requirements in relation to the nominal platform height is allowed during specific upgrade or renewal projects. This is especially valid, to avoid changes to a bearing construction.(den Europæiske Unions Tidende, 2014a) Safety Zones and Open Spaces The Danish railway norm BN1-9-2 describes the requirements for safety zones and open spaces at platforms in relation to platform widths, striping, signage, and placement of objects. The requirements are to be complied with in upgrading projects and for construction of new lines. Requirements for signage and striping are not covered in present section, as focus is on the layout of the platform. The norm, published in 2012, differentiate between TEN lines and lines with speeds above 160 km/h along the platform. From January 1st, 2015, the TSI requirements have become valid for the entire Danish network, resulting in the norm being valid for all platforms in the upgrade of the Lille Syd line. The relevant TSI in this case is the TSI PRM concerning accessibility for disabled persons. The TSI concerns the width of the platform in accordance to safety zones and open spaces. In addition hereto, the TSI PRM states, that the platform length must reflect the number of passengers at the platform in rush hours. A passenger is granted 1 m 2 in the open spaces. (Den Europæiske Unions Tidende, 2014b). In general a platform is divided in two areas; a safety zone, which only allows presence of passengers while a train is stationary at the platform, and an open space where presence is permitted at all times. Figure 5.12 illustrates these. Table 5.17 defines the minimum width of the safety zones in reference to the line speed. A safety zone with width of 0.85 m is valid for speeds up to 160 km/h, and for speeds up to 200 km/h a width of 1.35 m is required. Different requirements are valid to the minimum open space widths depending on whether or not objects are present at the platform. Table 5.19 defines the requirements for the different types of platforms; side and centre platforms, along with the determination of the minimum platform width according to size of objects. In general, the width is required to be 2 m, however, for small object less than 1 m in length, the open space width can be reduced to 1.6 m. Comparing with the TSI requirements, a larger width is required for larger objects.

102 86 NORM FOUNDATION Table 5.17: Minimum widths of safety zones, b sik, (Rail Net Denmark, 2012a) V [km/h] 0 < V m 160 < V m V > 200 Conventional lines presence not permitted Table 5.18: Minimum widths of open spaces, b oph, (Rail Net Denmark, 2012a), (Den Europæiske Unions Tidende, 2014b) Rail Net Denmark TSI PRM No object Little, g 1 < 1 m Large, g 2,3 > 1 m Little < 1 m Large > 1 m 2.0 m 1.6 m 2.0 m 1.6 m 2.4 m Figure 5.12: Sketch of the safety zones and open spaces at platforms (Rail Net Denmark, 2012a) Summary of Platforms with Respect to the Lille Syd line The Danish and European requirements for the location, height and dimension of platforms deviate only according to dimensions. The Danish requirements prescribe a platform height of 550 mm, whereas the TSI requirements state a platform height of either 550 mm or 760 mm. The width of the platforms are determined according to the line speed, and divided in areas of safety zones and open spaces. Open spaces are required to have a width of 2.0 m whereas safety zones differ between 0.85 m and 1.35 m. According to the TSI requirements is the dimension of the platform determined according to the amount of passengers. The Lille Syd line requires a platform length of m and the dimension of the open space zone is determined according to this. For the upgrade of the Lille Syd line, it is expected that a prolongation of the platforms are needed (Rail Net

103 CHAPTER 5 87 Table 5.19: Minimum requirements for platform widths, b per, (Rail Net Denmark, 2012a) Platform type Cross section of platform Minimum requirements for widths Side platforms Centre platforms Without objects b per = b sik + b g With objects b per = b sik + b oph + b g Without objects b per = b sik (platform side 1)+b sik (platform side 2)+b oph With objects b per =(b sik + b oph, side 1)+ (b sik + b oph, side 2)+b g Denmark, 2013d). 5.7 Level Crossings Present section describes the Danish and European requirements concerning level crossings. The issue of level crossings is widely discussed. In 2014, the Danish Transport Authority passed a law allowing a speed increase to 160 km/h through level crossings on the Lille Syd line Rail Net Denmark (2014f). Considering current thesis, an upgrade to 200 km/h of the line will counteract this. Rail Net Denmark defines the construction of full automatic level crossings in the rulebook SODB. The technical requirements for level crossings are not covered in present project. However, experiences from other European countries and projects concerning the impacts of speed upgrades in level crossings are presented, to determine the possibility of increasing the speed at the Lille Syd line while maintaining the level crossings. The section is divided in five subsections, considering requirements, issues and initiatives for improving the safety in level crossings. Initially an introduction to the Danish requirements and reasons for speeds through level crossings are given. A pilot project with connected analysis concerning different initiatives for improved safety level is then described. The allowed speed and implemented initiatives in level crossings in Germany and Sweden is followed by the general European requirements. Finally, another perspective of road blocking time is considered, and a summary of the section with respect to the Lille Syd line is provided in the end Danish Regulations for Level Crossings Level crossings pose a threat for both the railway and the road users. In the past ten years level crossings have caused 59 accidents, of which 22 are fatal. In general, most accidents arise from the road users or pedestrians behaving incorrectly according to the use of the level crossing. In relation hereto, Rail Net Denmark has since 2009 been working on eliminating all non-protective level crossings, and a major part have been closed (Elkjær, 2014). The Danish Road Traffic Act 5 state that road users shall demonstrate caution in level crossings, and that level crossings can not be crossed if a train can be heard or seen. On the contrary, Rail Net Denmark s safety regulations 11 prescribes, that a level crossing can only be passed when the signals indicate, that it is activated (Accident Investigation Board Denmark, 2014). Besides posing a threat, level crossings also makes up an obstacle for increasing the speed on certain regional railway lines. In recent years, analyses of the safety impact of increasing speed in level crossings have been conducted. The allowed speed through level

104 88 NORM FOUNDATION crossings has since 1991 been 140 km/h, based on international experiences and norms. Several analyses have been carried out since the 1990s with somewhat diverging conclusions, however, the allowed speed through level crossings has been maintained at 140 km/h. In June 2014, the Danish Transport Authority approved the speed through automatic protected level crossing to 160 km/h. The approval was based on accident data from Germany gathered by Rail Net Denmark (Rail Net Denmark, 2013d). The approval has been long in the making, and numerous of accident investigations have been conducted Risk Analysis Concerning Increased Speed in Level Crossings In 2012, COWI conducted an analysis dealing with speed upgrades of regional railway lines up to 160 km/h. In relation hereto, an analysis of the safety impact of increasing the speed in level crossings was considered. The analysis was based on initiatives of closing all non-protected level crossings, and ensuring that all level crossings used for a speed above 120 km/h were of the full barrier type. The analysis was carried out based on a declaration from the Danish Transport Authority stating, that increased speed in level crossings will be accepted as long as the current safety is not reduced (COWI, 2012). In a pilot project concerning the speed increase from 120 km/h to 160 km/h for the line Aalborg - Frederikshavn, a risk assessment of the safety in level crossings was conducted. All in all, the project concludes that increasing speed in level crossings will lead to the expectation of increased risk of accidents. The analysis concerns an upgrade of all protected level crossings to the full barriers type and a closing of all non-protected. Two types of estimations are carried out to indicate the increased risk of running 160 km/h, and to indicate what groups of persons are affected. Data for the past 6 1/2 years is used, and consequences of collisions with increased speed up to 160 km/h is estimated by extrapolation. Initially, the risk reduction of running with 140 km/h, while upgrading the level crossings are investigated. A risk reduction in this case will indicate the permitted increase in risk, an increase in speed to 160 km/h will allow. In general, the results indicate a risk reduction of 3 % of the safety goal. The road users experience a marginal reduction of 0.1 %, the passengers experience a reduction of 1,5 %, while the members of staff experience 14 %. In the second calculation, the speed is now increased to 160 km/h, and only marginal risk reductions are experienced. The road users experience more or less the same reduction in risk. However, the passengers experience an increase of 2 %, while the members of staff experience an increase of 33 %. Based on the calculations, it was concluded that the examined risk reduction initiatives were not effective in ensuring a unchanged level of risk. It was recommended, that initiatives on preventing collisions and reducing the impact of passengers and especially the locomotive driver were to be prioritised (COWI, 2011). In 2013, Rail Net Denmark has tested five different types of initiatives in level crossings. The initiatives are inspired by level crossings in Europe. An analysis concerning the behaviour of the road users in relation to illegal crossings of the railway, the speed and braking characteristics of the road users are carried out. In general, it is shown, that the amount of illegal crossings occur when the time span from the activation of the level crossing to the train passes is long (Jensen, 2014b). The five types of initiatives in level

105 CHAPTER 5 89 crossings are given below. Net below the barriers Flashing lights in the road lanes Variable road signs Extension of barriers to 3/4 barriers Acoustic bells faced towards pedestrians Based on the initiatives, the company Trafitec has conducted an analysis of the impacts, see Table The first initiative has the largest impact on pedestrians, whilst the behaviour of cyclists and car drivers is unchanged. In the initiative with flashing lights on the road lanes, the road restriction time is increased with 15 seconds causing more illegal crossings. In the contrary, the amount of catastrophic brakes are reduced with 85 %. The experiment concerning placement of a variable road sign 165 m in front of the level crossing caused the road users to reduce the speed with 8-10 km/h. The reduced speed will most likely affect the safety in the level crossing. The initiative with the extension of the barriers resulted in increased bell time and caused an increase in illegal crossings. However, when the level crossing is fully protected there is a decrease in illegal crossings. The final experiment considered bells turning towards the pedestrians ringing when the level crossing is turned on. The number of illegal crossings is decreased, especially in the time span just before the barriers are lowering. The effect is only valid for pedestrian, cyclists and motorcyclists. (Jensen, 2014b) Table 5.20: Results for initiatives on improving the safety in level crossings, (Jensen, 2014b) Initiative Decrease in illegal crossings Comments Net below Flashing Extension of barriers lights barrier Constant bells 80 % 20 % 37 % 34 % Primarily valid Only valid for Valid for pedestrians, for pedestrians pedestrians cyclists and motorcyclists In general, the results have been positive, however, the initiatives must be considered implemented based on the location and specific level crossing Allowed Speed in Level Crossings in Germany and Sweden In Germany, the speed limit through level crossings is 160 km/h, whereas the limit in Sweden is 200 km/h. In Sweden the allowed speed in level crossings without barriers is 140 km/h, whereas 160 km/h is allowed in half and full type barriers. If the level crossing is supplied with an obstacle detector, the speed is allowed increased to 200 km/h. The obstacle detector monitors if there are any obstacles in the free space of the level crossing, when it is turned on. However, these types of level crossings are not that common (Mornell, 2006). Sweden is, however, not the only country in Europe, where obstacle detection is implemented, though, the purpose varies. In Sweden, one of the main goals is to ensure safety while increasing the speed to 200 km/h, whereas the German goal is to

106 90 NORM FOUNDATION ensure safety by not increasing the speed above 160 km/h. In a comparison of accidents per level crossing, Sweden has twice as many accidents compared to Denmark, whereas Germany has a significantly low number, however, all three countries lie below the European average (Danish Transport Authority, 2010a) International rules for level crossings Internationally speaking the awareness concerning safety in level crossings has increased throughout the past decades. Based on the realisation of having too many accidents in level crossings, a forum known as ELCF (European Level Crossing Forum) has formed. ELCF hosts meetings twice a year, where European countries meet and exchange knowledge and experiences of how the issue is dealt with (I, 2013). In 2014 the invites included 43 countries spread across the globe, and the central focus was the distraction of young people using cell phones. The International Union of Railways presented in a leaflet from 2005, the additional safety measures that are to be taken into account for railways operated with a speed between 120 km/h and 200 km/h. Speeds above 200 km/h are not tolerated in level crossings, however, national values must be complied with in case of discrepancy. In general, the leaflet suggests that all level crossings must be considered closed if the speed increases 120 km/h, and all level crossings without technical protection are not allowed for speeds above 120 km/h. Besides the general suggestions, the leaflet puts forward a list for prioritizing the closure of level crossings. Level crossings should be considered closed in cases where one of the following points are considered true; In cases of heavy and/or slow moving road traffic occurring within a certain frequency In case of gaining any size of profit, when comparing maintenance and attendance with the investment cost of constructing a bridge or a bypassed road In case of private or rarely used level crossings, and level crossings solely used by pedestrians Besides in general suggesting the limitation of operating level crossings, the leaflet recommends what type of level crossings to be used at specific speeds. Level crossings must be equipped with barrier systems for speeds above 120 km/h. Half barriers are accepted for speeds up to 160 km/h, whereas full barriers are recommended for speeds above 160 km/h (International Union of Railways, 2005). The TSI safety norms concerning level crossings engage solely in restrictions for high speed railway lines. For lines constructed for high speed or lines where the speed is 250 km/h or above, level crossings are not allowed (Transport Minister: Kristensen, 2013) Road Restriction Time - Another Aspect Besides solely considering the safety aspect, another aspect to consider is the percent of time, where the level crossing will hinder traffic on the road. Depending on the location,

107 CHAPTER 5 91 the type of level crossing, and the speed of the train, the road restiction time consists of several minutes every time a train is passing. For a line operating 2 trains every hour in each direction, and with the assumption of a speed of 100 km/h, the variation in total road restriction time varies between 1-2 minutes (Rail Net Denmark, 2014d). This will result in a road restriction time of up to 13 % in a daily hour. In a report from 2011, COWI estimates the increased road restriction time to be eight seconds, if the speed in the level crossing is increased from 120 km/h to 160 km/h (COWI, 2011). Taking the North Western railway line as an example, an introduction of a bill suggested, that a major part of the level crossings should be closed and bypass roads built, as four trains every hour would result in blockage of the road 30 % of the time (Transport Minister: Barfoed, 2009). If traffic on a line is to be increased with either faster running trains or freight trains, the percentage of the road restriction time is increased in correlation hereto. In order to evaluate whether or not to maintain a level crossing, the number of road users must be considered, when determining whether a bypass road is a better alternative Summary of Level Crossings with Respect to the Lille Syd Line Preceding section treats the requirements for level crossings, and the issues related to a speed increase herein. In 2014, the allowed speed in level crossings were increase from 140 km/h to 160 km/h. The decision had numerous analyses and pilot projects supporting it, as well as experiences from Germany. An UIC leaflet from 2005 presents the safety measures to be taken into account for allowing level crossings up to 200 km/h; for speeds above 160 km/h full barriers are required. Rail Net Denmark has closed down all nonprotected level crossings in the past years along the Lille Syd line, and in accordance with the speed upgrade in 2018, 5 level crossings are maintained in the upgrade to 160 km/h (Grontmij, 2014). The basis for decision formulated by Rail Net Denmark states, that the signal exchange will make it technical possible to manage the higher speed. Focusing on the speed upgrade to 200 km/h of the line, level crossings are expected to be closed, considering German limitations of 160 km/h, and the fact that 200 km/h is seldom allowed in Sweden. 5.8 Norm Summary Current section summarises the norm foundation for the upgrade of the Lille Syd line. Two types of upgrade speeds are considered; 160 km/h at stations and 200 km/h, as well as construction of a second track for similar speeds. In general, Rail Net Denmark s requirements are followed, however, attention towards the European requirements is drawn, to determine the most cost and time saving efficient solution. The chapter considers solely disciplines related to the track design, and the overall findings in sections are summarised here. Track geometry Considering the track geometry design for the Lille Syd line, the Danish requirements are in general stricter, than the European requirements. In the further analysis of the track design, both the Danish and European requirements are considered, to determine the speed profile with the possible highest speed. However, the chosen speed profile for the Lille Syd line are designed to comply with the Danish regulations.

108 92 NORM FOUNDATION Ballast profile The Danish requirements for the ballast profile is provided, since no European requirements are stated. Deviation in speed and type of lines (existing lines, upgrades, renewals and construction of new lines) are considered. The characterisation of the ballast profile is applied when determining the construction cost of the upgrade. Track center distance The Danish and European requirements for the track center distances are considered, to determine the location of the second track for Lille Syd. The location is determined according to the line speed, and deviation between 160 km/h and 200 km/h is taken into account. The location of the second track plays a role in the construction cost, as it is closely relation to expropriation of land. Structure gauge The structure gauge is considered to ensure interoperability throughout Europe. No requirements for structure gauges are provided in the TSI INF, instead structure loads are presented. The two structure load profiles valid for the Lille Syd upgrade is the GC and the DE3 profile. By complying with the Danish requirements for structure gauges, both profiles are accepted. This is considered in the construction cost for the second track, as new bridges will become a necessity. Platforms In general, the Danish and European requirements for the platform layout are more or less compatible, however, the length of the platforms are not. Considering the upgrade of the line, prolongation of platforms, reconstruction according to the width, and a possible increase in platform height is expected. This will contribute to a rather large post in the construction budget. Level crossings Level crossings are considered, as issues regarding increased speed from 140 km/h to 160 km/h, has raised the question of safety levels. A speed of 160 km/h through level crossings was allowed in Denmark in The TSI requirements allow speeds up to 200 km/h, however, by comparing German and Swedish requlations, an upgrade to 200 km/h will possibly require out of level crossings. The upgrade of the Lille Syd line is based on the norm foundation described in preceding chapter. The track geometry described is used to determine the horizontal track design, whereas the remaining disciplines are considered in the project description and as posts in the construction estimate. The following chapter 6 considers the track optimisation model, proposed to facilitate the optimisation of cant size according to speed in the existing infrastructure. The Danish TER and the TSI requirements for the geometrical horizontal track design are implemented in the model.

109 Chapter 6 Optimisation Model In chapter 4 it was seen, that the network capacity of the corridor between Copenhagen and the future fixed link at Fehmarn Belt can be improved by integrating the Lille Syd line. This entails allocation of two freight train paths and two passenger train paths from the main line passing Ringsted to the Lille Syd line, in the most trafficked hour. A prerequisite for this is extension of the Lille Syd line from single to double track. Furthermore, it is necessary to upgrade the line speed to at least 160 km/h, to avoid sacrificing the travel time of the passenger trains. The norm foundation presented in chapter 5, forms the basis for realising the upgrade of the infrastructure. Present chapter focuses on the speed upgrade of the existing track, according to the horizontal geometric track requirements stated in section 5.2. In relation hereto, an optimisation model applicable for investigating the optimal cant size according to the geometric track design and goal speed is proposed. The sections below present the objective and capabilities of the proposed model, followed by a description of the development. Finally, the potential future development of the model is discussed. 6.1 Objective As described in section 5.2, a series of track geometrical requirements must be met when upgrading the line speed. In a screening of potential speed upgrades of a railway line, the track geometry is therefore often one of the first parameters to be considered. Today s procedure for the initial speed upgrade investigation, includes an examination of the existing curvature, with the purpose of increasing the line speed without violating the track geometrical requirements. Attempts are then made to optimise the cant size to obtain an increased line speed. This procedure is carried out manually and the line speed is assessed for each curve. Due to the long list of requirements, this can be quite time consuming. This is especially the case for areas where curves share one transition curve, thus, changing one cant size will influence the adjacent curve. This increases the level of complexity and makes it difficult to obtain the best solution. This manual process makes it furthermore difficult to conduct a fast screening of the impact of easing the TER requirements for a specific project. This can be valid in case of 93

110 94 OPTIMISATION MODEL easing the TER according to the TSI requirements as described in section 5.2. In present project, the potential gain of relaxing the TER by allowing larger cant sizes within standard regulations, and permitting a cant deficiency of 153 mm within exceptional regulations is described. Based on this, an optimisation model computing the cant size required for obtaining the highest speed profile is proposed. Changes to the requirements can therefore be typed in the model, to evaluate the effects on the speed profile. This model is applicable for investigating the optimal changes in the curvature on the Lille Syd line for a design speed of 160 or 200 km/h, respectively, according to the scenarios presented in chapter 4. Furthermore, it serves as a tool for realising the large number of future upgrading projects expected to rise from the political initiative of Togfonden DK introduced in chapter Model Framework The proposed model is composed by the Danish TER and considers both requested, standard and exceptional regulations. It enables optimisation of the design speed in all curves by variation of cant size with respect to the curve length. The mathematical model is presented in Appendix H, and is more or less a mathematical interpretation of the Danish TER. The model is explained in detail in Appendix H, while present section focuses on an overall description of the model framework Model Input The model consists of four sets including; curves, possible cant sizes, possible speeds and penalties to ensure a minimum use of exceptional regulations. A curve dataset contains the required information of each curve. The possible cant sizes vary from 0 to 160 mm in multiples of 5 mm, with exception of the interval 5-20 mm, because these are normally set to either 0 or 20 mm. The set of speeds includes the possible line speed in multiples of 10 km/h up to the target speed, which for present project is 160 km/h and 200 km/h, respectively. The set of penalties corresponds to each of the eight constraints, which makes up the TER standard requirements, in addition to cant sizes, see Table 6.1. A penalty is linked to each of these constraints, which entails a minimum use of exceptional regulations in curves not complying with the standard regulations. The size of the weight is assessed based on the findings in section 5.2. From here, it is decided that violating cant deficiency will trigger the largest penalty, whereas violation of cant size does not entail any penalty, due to violating of cant size is preferred over cant deficiency according to section 5.2. Violating the remaining standard requirements is assessed to have same significance. Based on this is violation of cant deficiency assigned penalty value of 2, while the remaining penalties is assigned a value of 1, see Table 6.1. Hence, cant deficiency is twice as important as the remaining. In a more detailed study on the penalty system it would be desirable to make a sensitivity analysis to clarify how changing the assigned penalties will influence the result.

111 CHAPTER 6 95 Table 6.1: Weight applied as penalty to ensure minimum use of exceptional regulations for curves not complying with standard regulations Constraint Index Assigned Penalty Cant deficiency > 100 mm 1 2 Length < 0.25V m 2 1 di dt dh dt > 55 mm/s 3, 6 1 > 50 mm/s 4, 7 1 dh dl > 2 mm/m 5, 8 1 The model is composed by three binary decision variables; 1 if curve c C applies cant size k K and speed s S x cks = 0 Otherwise 1 if curve c C applies cant size k K and speed s S y ckpsq = and curve c+1 C applies cant size p K and speed q S 0 Otherwise 1 if curve c C applies exceptional regulation i z ci = 0 Otherwise (6.1) (6.2) (6.3) Here the x-variables and y-variables are used to capture the model constraints. The y- variables are used for constraints regarding transition curves, while the x-variables are used for all additional constraints. This distinction is necessary, to determine the values h and I, which depend on the cant size and cant deficiency in two adjacent curves, when these share one transition curve. Additional constraints, such as cant deficiency less or equal to 100 mm, is on the contrary only determined in the specific curve. This can therefore be determined without consideration of the adjacent curves. The model ensures that x cks is set to 1 if curve c uses cant size k and speed s. Likewise y ckpsq is set to 1 if curve c uses cant size k and speed s and curve c + 1 uses cant size p and speed q. Otherwise, the variables are set to 0. The z-variables are used to determine whether a curve violate any of the eight standard regulations listed in Table 6.1. z ci is therefore set to 1 if curve c violate standard regulation i, otherwise 0. The required curve data is listed in Table 6.2. This includes geometrical information of the curve and transition curve in relation to curve radii, length, start stationing for the associated transition curves and their lengths. In addition hereto, binary variables are applied to express whether a curve contains a platform, turnout or contra flexure curved turnout, and whether the specific curve should comply with requested or normal regulations. If nothing is stated, the curve will comply with exceptional regulations. Based on these input, values associated with the requirements presented in the norm foundation

112 96 OPTIMISATION MODEL in section 5.2 is generated (e.g. cant deficiency, cant excess, etc.). The model computes values for all possible combinations of cant sizes and speeds. This computation requires thereby large memory capacity. In a large curve set where many possible speeds are investigated, can it be beneficial to divide the curve set into smaller sections. Furthermore, elimination of certain speeds will result in shorter computation time and less requirement of memory capacity. Table 6.2: Required data input Curve Data Curve radius Curve length Length of first transition curve Length of second transition curve Start stationing of first transition curve Start stationing of second transition curve Platform Turnout Contra flexure curved turnout Application of requested regulations Application of standard regulations Unit Meters Meters Meters Meters Meters Meters Binary Binary Binary Binary Binary Objective Function The objective function seeks to maximize the sum of curve speeds and minimize the amount of applied exceptional regulations. Both factors are multiplied with the curve length, to ensure that the longest curves are prioritised. The objective function is shown in (6.4). Maximize x cks sl c w i z ci L c (6.4) c C k K s S c C i E where C is the set of curves, K is the set of cant sizes, S is the set of speeds and E is the set of exceptions. L c is the curve length and w i is the applied penalties. A constraint ensures that x cks only is set to 1 in the chosen combination of cants and speeds constituting to the largest speed. Additional combinations of possible cants and speeds are therefore excluded. Likewise, constraints ensure that z c i only is 1 if curve c uses exceptional regulation i Constraints The constraints are the requirements stated in TER. To differentiate between requested, standard and exceptional regulations, the first two are multiplied with the binary variable for requested and standard regulations, respectively. In case none of these requirements are activated, the exceptional regulations are applied. Same procedure is used when dis-

113 CHAPTER 6 97 tinguishing requirements for platforms, turnouts, etc. (6.5) shows an example of how the constraints are included in the mathematical model in Appendix H. This constraint applies standard regulations for cant size along platforms, which should be less or equal to 60 mm according to TER. The binary variable is n c = 1 when standard regulation applies in curve c, and pl c = 1 when curve c is placed along a platform. If any of these are set to zero, the constraint will become ineffective. The x variable is set to zero in the combinations of cants and speeds, which do not comply with the constraint. k K s S n c pl c kx cks 60 c C (6.5) In order to compute solutions for requested regulations, it is, however, necessary to change to regulations for cant size and cant deficiency, given these are provided as exact numbers in TER, which most likely will be impossible for the model to solve. The upper bound of requested cant size of 115 mm is therefore applied instead of the mathematical expression in (6.6). The requirement for cant deficiency in (6.8) is changed from equal to less or equal to. In case of very large radii, the requested cant deficiency will move towards zero. This is problematic for theoretical large curve radii, which can occur if two straight elements with a small break are combined. To adjust for these cases, a value of +1 is added to the constraint according to (6.9). h requested TER = 8V2 R (6.6) h requested model 115 (6.7) I requested TER = 3.8V2 R (6.8) I requested model 3.8V2 R + 1 (6.9) The model technical constraints which control the x and y variables are listed below. (6.10) states that the sum of all x-variables equals 1. This ensures that one cant size and one speed are chosen for each curve. The y-variables are then set in accordance to the x-variables according to Equation 6.11 and Equation 6.12 k K s S x cks = 1 c C (6.10) y ckpsq x cks + x c+1,pq 1 c C, k K, p K, s S, q S (6.11) Forces y ckpsq = 1 when x cks = 1 and x c+1,pq = 1. 2y ckpsq x cks + x c+1,pq c C, k K, p K, s S, q S (6.12) Forces y ckpsq = 0 if either x cks = 0 or x c+1,pq = 0. Finally, constraints are used to ensure that the cant size is only changed if the curve is connected to a transition curve. If not, the cant size in the adjacent curve is chosen, or a cant size of zero, if the curve is connected to straight track.

114 98 OPTIMISATION MODEL Model Implementation The mathematical model is developed with guidance from Jesper Thorsen, a fellow master student at DTU. To apply the mathematical model to generate alignment solutions, it is necessary to implement the model in an optimisation programme. This part of the project has been conducted by Thorsen, due to his expertise within CPLEX. Thorsens ambitions are to advance the model further in a separate special course at DTU Transport. The script used to generate the alignment solution for present project is placed in Appendix I. The computation time varies depending on the number of curves and the investigated set of speeds. To reduce the computation time and the requirement for memory capacity, the curve set can be separated into minor sections, as long as the set is not split where two curves share a transition curve. Another way of reducing the complex computation is to reduce the number of investigated speeds. For instance, lower speeds can be taken out of a minor curve set, if the existing design speed is high. In current project, the set of curves were split into 6 minor sections of 10 curves with corresponding transition curves. The computational time for all possible cant sizes and speeds was approximately 10 minutes. 6.3 Field of Application In its current stage, the model is applicable for investing the optimisation of the line speed by adjusting the cant size. The model furthermore enables variation between allowance of the three types of regulations; requested, standard and exceptional. When allowing exceptional regulations, a weighting system ensures that a minimum of exceptional regulations are applied. The model output is given as a list of cant sizes and speeds chosen for each of the examined curves. It is possible to divide the curve set in smaller dataset, to obtain faster computation. In relation hereto, it is however important not to divide the curve set at places, where adjacent curves share one transition curve. By modifying the model constraints, it is possible to investigate different requirements. For instance, the EN standards could replace the TER requirements, to investigate whether a higher line speed can be obtained. The model evaluates only curves, hence the allowed speed at straight elements will have to be considered separately. This is considered a relatively fast check, given that straight elements only shall comply with the requirements for the element length in relation to track geometry. It is therefore independent of the applied cant size. Based on the allowed speeds at straight track and the speed output delivered by the model, it is possible to sketch up possible speed profiles. Next step is then to smoothen out the profile, to enable trains to reach the top speed. This can be done by decreasing the found speed or making additional changes to the geometrical alignment, to obtain the desired speed profile. With the current edition of the model, these steps are considered manually.

115 CHAPTER Model Visions In its current state, the model is already a tool to assist the initial investigation of speed upgrades without changing the infrastructure, except for cant sizes. The potential for further development is considered large, and some of the improvement suggestions are assembled below. Initially, it would be necessary to cut down on the computation time, to ensure that more parameters can be considered. Today s manual approach for upgrading railway lines consists of three steps; 1. Change in cant size (already implemented) 2. Prolongation of transition curves 3. Increase in curve radii Step one is already implemented in the model, while step two and three are still on the drawing board. Inclusion of these steps requires, at first, incorporation of the straight track elements. Subsequently, the model can be enhanced to allow the geometrical changes. This entails incorporation of the formulas listed below, where (6.13) determines a new curve length when prolonging the transition curve length and/or increasing the curve radii. Hereto, (6.14) determines the displacement of start and termination point of the curve (Jensen, 2012). L new = r new r org. (L org (TCL1 org. + TCL2 org. ) 1 2 (TCL1 new + TCL2 new )) (6.13) x disp. = (r new r org. )tan( L org (TCL1 org. + TCL2 org. ) ) + TCL 2r org. 2 (6.14) Here org. is the original measurements, while New indicates the changed measurements. L indicates the curve length, r is the radius and TCL1 and TCL2 are the lengths of the first and second transition curve. TCL is the difference between the new and old transition curve length, and should be determined in both ends of the curve. (Jensen, 2012) An optimised model would therefore enable assessment of the possibility of carrying through these changes. To prioritise the sequence of steps listed above, the penalty system is additionally extended. Larger penalties are therefore associated with increasing the curve radii compared to prolongation of the transition curves. This applies especially for large curve displacements, which can be estimated according to (6.15), (Jensen, 2012). y = (r new r org. ) cos( L org+ 1 2 (TCL1 org+tcl2 org. ) 2r org. ) r new r org. (6.15) The estimated curve displacement could then be summarised by the model. In relation hereto, estimations of ballast supplements, extension of track bench and additional subballast could be determined, by incorporating Rail Net Denmark s requirements for the ballast profile described in section 5.3 in the model. These are caused by changes in the cant size, and can thereby be evaluated by comparing the new and original cant size in each curve. The quantities estimated by the model could then be used in an initial evaluation of the construction cost.

116 100 OPTIMISATION MODEL Another idea would be to enable the model to compute speed profiles. The model could then be linked to Jensen s running time calculation programme presented in chapter 4, and enable an immediately interpretation of the achieved travel time savings caused by the upgrade. Instead of optimising the speed in each curve without consideration of the obtainable speed, the programme could assist targeting where upgrades would result in a smooth speed profile. This would improve the connection between timetable and infrastructure planning. In a very visionary perspective, this could be taken even further. The model could consider the cost for upgrading the specific curve. This will depend on identified parameters such as; placement of turnouts, road or railway bridges, stations, level crossing or potential exchange of elements in the track construction. Furthermore, the quantities for curve displacement, ballast supplement, etc. could be multiplied with unit costs in the model. The time savings could then be multiplied with time values, and the objective function would seek to optimise time benefits according to construction costs. This is a very long-term vision for the simple optimisation model proposed in present project. Such development requires large efforts to programming the model, but also for assembling the model input. Evaluation of the local conditions is required to estimate the cost for each curve. The vision is therefore not to have a model, which can compute a socioeconomic analysis in a second. Instead the model could be applicable when evaluating a project, with eyes on locating the cheapest minutes. In the light of the numerous upgrading projects carried out in the Danish railway network in present years, such model could be very beneficial when aiming at reducing construction costs. 6.5 Summary Present chapter suggest an optimisation model for upgrading railway lines. The model evaluates a curve dataset by changing the cant size, to optimise the line speed in accordance with the Danish TER presented in section 5.2. The model is applicable in the initial screening, for upgrading the line speed without changing the horizontal alignment, except the cant size. Furthermore, it shows the immediately impact of easing certain design requirements (e.g. permitting higher values for cant deficiency). The model has a large potential of development. This includes incorporating additional degrees of freedom to allow for prolongation of transition curves and straightening of radii. Furthermore, it would be beneficial to connect the model to Jensen s running time calculation model, to enable immediately assessment of the upgrading effect. In the longterm perspective, the model could help targeting where to upgrade a railway line, to obtain the cheapest minutes. In the next chapter, the model is used to compute two alignment solutions for the upgrade of the Lille Syd line to 160 km/h and 200 km/h, respectively.

117 Chapter 7 Track Geometry Solution Based on the optimisation model develop in chapter 6, the track geometry for the two alternative speed solutions of 160 and 200 km/h is now analysed. The optimisation model, is used to determined the optimised cant size according to the speed objective, and track adjustments are implemented to obtain an accepted profile. Comparisons in travel times between the two scenarios are made to determine the effect of the upgrade. The final speed profile for the two alternatives are investigated further, to clarify the minimum infrastructural changes required to obtain the speed profile, while at the same time allowing the use of exceptional regulations to a larger extent. The chapter is divided in four sections considering the general considerations and the strategy for the upgrade, the two solutions, and finally a comparison with the EN standards. The curvature diagram for both speed upgrades is enclosed in drawing TCE placed in Appendix P, and the datalists can be found in Appendix J. 7.1 General Considerations within Speed Upgrades Speed upgrade projects have widely been considered in the past couple of years. In the wake of the decision of implementing the new Signalling Programme, COWI conducted a screening of upgrades on Danish regional lines COWI (2012). The majority of the results came out positive, and some of the investigated lines are currently being upgraded. In an upgrading project, it is of great importance to consider the length of the section, which can be upgraded to a higher speed. Furthermore, acceleration and braking characteristics for the specific train types according to stopping patterns are relevant. It is experienced, that shorter distances of upgraded speed can not always be utilised, and a smooth speed profile is therefore always attempted obtained. In his master thesis, Jensen describes the findings of an analysis carried out in the 1990 s, concerning the driving behaviour of locomotive drivers on the main line in western Denmark between Fredericia and Aarhus. The findings show that sections of 3-4 kilometres with an average higher speed were not utilised, even if the train was delayed. In a complex profile, the speed profile will be smoothened by the locomotive drivers, and smaller upgrades will therefore not be utilised. Based on the analysis, Bo Nielsen, Technical System Responsible at Rail Net Denmark, recommends the following minimum lengths of speed upgrades; (Jensen, 2012); 6 kilometres for speeds between km/h 101

118 102 TRACK GEOMETRY SOLUTION 8 kilometres for speeds between km/h 10 kilometres for speeds between km/h Keeping the recommeded section lengths in mind, the existing infrastructure of the Lille Syd line is upgraded to 160 and 200 km/h respectively. Smooth speed profiles are desirable, however, smaller speed upgrades are also considered, if these result in a theoretical travel time saving. In general, a speed upgrade project will in most cases challenge the existing infrastructure in order to comply with the national regulations. As stated in chapter 5 requested requirements are always attempted to be complied with, to minimised wear and adjust to passenger comfort. A speed upgrade will most likely demand high usage of track adjustments and in some cases new track design to comply with requested regulations. In order to minimise the cost of the upgrade, standard regulations are basically used, and only a limit amount of exceptional regulations are allowed. Implementation of the second track is based on the existing track layout. In general, it will be beneficial to implement a new track by use of requested regulations. In a future speed upgrade of the newly implemented track, it will therefore demand less complicated track work to upgrade the track to a higher speed. However, current project assumes the same track design for the second track, as the upgraded track solutions found in this chapter. This will result in the same speed profile for the two tracks, and thereby more or less the same running times depending on the stopping pattern and speed profile. 7.2 Strategy for Upgrading the Speed Profiles The speed upgrades consider the use of Rail Net Denmark s standard and exceptional regulations. Comparisons of the two regulations are conducted to clarify the minimum adjustment in infrastructure to obtain the objective speed. EN exceptional regulations are furthermore considered, to determine the impact of the different regulations in a speed upgrade project. The following solutions are prepared; Effort to upgrade to 160 km/h and 200 km/h respectively, with use of Rail Net Denmark s standard regulations and a minimum use of exceptional regulations Minimum effort to upgrade the track to 160 km/h and 200 km/h respectively, with greater acceptance of Rail Net Denmark s exceptional regulations Minimum effort to upgrade to 160 km/h with acceptance of EN standards In an upgrading process, different initiatives are considered to adjust the cant size. The initiatives are prioritised with respect to the level of complexity, which also is reflected in the cost of execution. The upgrades are carried out according to the following prioritisation; Cant sizes are optimised Lengths of transition curves are adjusted Curve radii are adjusted

119 CHAPTER A change in cant size results in a change in the vertical location of the rails. Such change is considered a track adjustment and is implemented by use of a tamper machine. Adjustment in the length of transition curves are also considered track adjustment and can be carried out in accordance with the change in cant. To minimize the use of exceptional regulations, minor track displacements are a prerequisite. In curves, an adjustment in radius will result in a displacement of the track. The actual track displacement depends on the radius of the curve and the lengths of the transition curves, according to the formulas described in chapter 6. Track displacements result in different consequences according to the extent of the displacement. Displacements less than 50 mm are considered as minor track adjustments and can be carried out simultaneously with an adjustment in cant size. Displacements between 50 and 100 mm are considered larger track adjustments, and for displacements between 100 to 200 mm reconstruction of drainage are required. For displacements larger than 200 mm a new track bedding is required (Grontmij, 2015a). At stations the track geometry is retained as far as possible by only allowing minimum changes in the geometry. In urban areas, except at stations where the line already is double track, it can be argued that an implementation of a second track will already demand expropriation, therefore a smaller displacements of the existing track will be accepted. It is assumed, that an increase in cant size will not interfere with the structure gauge. This is assessed as a fair assumption, as the platforms are reconstructed, and new bridges are required on open line sections, due to the implementation of double track. In general, exceptional regulations are only allowed in situations, where it is assess very costly to optimize the geometry. This is especially valid in urban areas or where the track geometry is complicated. 7.3 The Existing Track Geometry The existing speed profile for the Lille Syd line, presented earlier in section 4.1, are now being investigated with inclusion of freight trains, and line variants 20, 21 and 35/36. The utilisation of the existing speed profile is illustrated in Figure 7.1. By examining the utilisation of the existing speed profile, it is seen that a speed upgrade to 160 km/h at stations will benefit line variant 20 and 21, whereas line 35/36 will only possibly benefit, and the freight trains will not be affected. It can furthermore be concluded, that a speed upgrade to more than 160 km/h and a retention of 120 km/h at stations, will possibly only result in little travel time savings, as the lines will have difficulties reaching the maximum speed. A speed upgrade to 160 km/h on the total line is therefore investigated, with the expectation of leading to travel times savings for line variants 20 and 21. In addition hereto, at speed upgrade to 200 km/h is investigated in a later section.

120 104 TRACK GEOMETRY SOLUTION Driving Characteristics for Exsiting Speed Profile of Lille Syd 160 KjN Ølb Kj Hf Th Hz Ol Næn Næ 140 Speed [km/h] Stationing [km] Stations G4 and G6 GX 35/ Existing Speed Profile Figure 7.1: Utilisation of the existing speed profile at Lille Syd in accordance with line types 7.4 Upgrade of Existing Track to 160 km/h The upgrade is carried out in several steps to prepare the three solutions presented above. The process of upgrading consists of the following steps; Model run first time: change in cant size according to Rail Net Denmark s standard regulations and objective speed of 160 km/h Identification of elements where the objective speed is not reached Model run second time: change in cant size with acceptance of Rail Net Denmark s exceptional regulations in identified elements with a speed below 160 km/h Change in infrastructure according to the prioritised process mentioned in section 7.2, to eliminate redundant use of exceptional regulations Smoothening of speed profile to ensure utilisation of all upgrades Initially the cant optimization model optimises the cant in the existing infrastructure by only allowing Rail Net Denmark s standard regulations. Locations where the object speed of 160 km/h cannot be reached are identified, and the optimisation model is run again with allowance of exceptional regulations in the respective locations. In reality, it is not necessary to run the model twice, as penalties are linked to the usage of exceptional regulations. However, to obtain a graphical overview of the differences in the two profiles, this is done. Figure 7.2 illustrates the possible speed upgrades in the infrastructure according to the use of Rail Net Denmark s standard and exceptional regulations compared with the existing speed profile. In general, the target speed of 160 km/h is obtainable with the use of standard regulations except in three subsections; the area between Køge Nord and Køge, between Haslev and Holme-Olstrup and finally between Næstved Nord and Næstved. The area between Køge Nord and Køge station is examined in detail according to Figure 7.3. It is noticed, that the model output for standard regulations apply a lower speed

121 CHAPTER Speed Profile Difference in Rail Net Denmark's Standard and Exceptional Regulations KjN Ølb Kj Hf Th Hz Ol Næn Næ Speed [km/h] Stationing [km] 160 km/h RND Standard Regulations Existing Speed Profile 160 km/h RND Exceptional Regulations Stations Figure 7.2: Difference in speed profile according to use of Rail Net Denmark s standard and exceptional regulations compared with the existing speed profile presented in Figure 4.4 in chapter 4. It can therefore be assumed, that exceptional regulations are accepted in the areas numbered 4 and 6, for the line variant 35 and 36. However, this is not valid for area 2, as the line variants continues on the northern part of the Lille Syd line around stationing Speed Profile Difference in Rail Net Denmark's Standard and Exceptional Regulations Area between Køge Nord and Køge KjN Ølb Kj Speed [km/h] Stationing [km] 160 km/h RND Standard Regulations 160 km/h RND Exceptional Regulations Stations Existing Speed Profile Figure 7.3: Difference in speed profile according to use of Rail Net Denmark s standard and exceptional regulations in the area between Køge Nord station and Køge station The track between Køge Nord and Køge Station consist of several curves with various length and radius and only shorter straight line sections. The section is location in an urban zone with relative high density housing along the major part of the line. This result in a rather complicated track geometry leading to an allowed speed of 120 km/h except

122 106 TRACK GEOMETRY SOLUTION at Køge station, where the speed is 100 km/h. In general, the speed profile in this area can be divided in 6 smaller sections. It will not be beneficial to maintain a speed profile according to either of the two regulations used, hence the profile is smoothened. All 6 sections are examined according to the type and amount of exceptional regulations used. A smooth profile is then obtained by changing the lengths of the transition curves to limit the use of exceptional regulations. The curve lengths and radii are attempted not to be changed, due to the geographical location of the track and the possibly conflict a track displacement will result in. Travel speed calculations are performed to ensure all upgrades are utilised, as a minimum in one direction. The smoothened speed profile is shown in Figure 7.4, with the profile (in green) and the locations of where exceptional regulations are used (in red). According to Figure 7.1, it was seen that only line variant 20 and 21 can utilise a speed upgrade in the area between Køge Nord and Køge, due to the remaining lines stopping pattern. These are shown in both driving directions. The largest speed reduction is seen immediately after Køge station, where the speed is reduced to 120 km/h. This reduction cannot be eliminated with smaller changes in the infrastructure and use of exceptional regulations. The track consists of a short curve with an associating small radius, and two relative short transition curves. Furthermore, an inside curved turnout is located in the curve. It is assessed, that a total new track layout is necessary for upgrading the speed to more than 120 km/h. Speed Profile Smoothened Profile between KjN and Kj with Acceptance of Exceptional Regulations KjN Ølb Kj Speed [km/h] Stationing [km] Stations 160 km/h Exceptional Regulations Smoothened LN20 Towards Næstved LN21 Towards Næstved LN21 Towards Copenhagen LN20 Towards Copenhagen Exceptional Regulations used Figure 7.4: Utilisation of different speed profiles according to use of Rail Net Denmark s standard and exceptional regulations in the area between Køge Nord station and Køge station A further examination of the speed profile indicates, that not all areas with applied exceptional regulations, are utilised by the line variants. The two demonstrated areas, are only fully utilised by line variant 20, due to the driving characteristics of line variant 21 and the stop at Køge Nord station. It is now assessed, whether the speed in these areas should be maintained, by evaluating the amount and type of exceptional regulations used contra the speed curve and gain in travel time. To maintain a speed of 160 km/h

123 CHAPTER at the area before Køge station, from to , 11 exceptional regulations are used, and a curve straightening is required. The regulations are mainly used due to short transition curves. By lowering the speed to 150 km/h, only one exceptional regulation is required due to cant deficiency. The speed reduction will lead to an increase in travel time of 0.45 seconds for line 21 and 0.94 seconds for line 20. These travel time reductions are very small in relation to the changes needed changes in the infrastructure and amount of exceptional regulations used. The speed has therefore been lowered to 150 km/h in this section. The remaining infrastructure is analysed with more or less the same procedure, and the final smoothened profile is obtained for the 160 km/h solution in Figure 7.5. Compared to the existing infrastructure from Figure 7.1, a travel time reduction of 88 seconds are obtained for line 20 and 89 seconds for line 21. The 160 km/h solution is shown in Appendix J and the curvature is sketched in the enclosed drawing TCE placed in Appendix P. To obtain the speed profile for the final solution, 27 exceptional regulations have been used as well as adjustments in the infrastructure. These are presented in the following section. Smoothened Profile between KjN and Kj with Changed Geometry and Acceptance of Exceptional Regulations KjN Ølb Kj Hf Th Hz Ol Næ Næn Speed [km/h] Stationing [km] 160 km/h Smoothened Profile Exceptional Regulations Stations Figure 7.5: Final smoothened speed profile with indication of exceptional regulations used Upgrade to 160 km/h with Minimum Effort In addition to the final solution for the speed upgrade to 160 km/h, it is investigated whether adjustments in the infrastructure could be avoided by allowing exceptional regulations. Figure 7.6 illustrates the obtained speed profile with minimum adjustments in the infrastructure and equivalent increase in use of exceptional regulations. Table 7.1 and Table 7.2 summarises the amount of exceptional regulations used in the two solutions and the adjustments in the infrastructure, both according to type. From these it can be seen, that it is possible to spare 1 adjustment in curve radii and 10 adjustments in transitions curves, if the number of exceptional regulations used is doubled. The solution for the minimum effort upgrade to 160 km/h can be seen in the enclosed CD.

124 108 TRACK GEOMETRY SOLUTION The solution with minimum required infrastructure changes will, however, probably not be approved, due to the large extent of exceptional regulations. The final solution presented in Figure 7.5 is therefore the one being priced in chapter Speed Profile for 160 km/h with Minimum Change in Infrastruture and Acceptance of Rail Net Denmark's Exeptional Regulations KjN Ølb Kj Hf Th Hz Ol Næn Næ Speed [km/h] Stationing [km] Stations 160 km/h Exceptional Regulations Smoothened Exceptional Regulations Figure 7.6: Smoothened speed profile with minimum change in infrastructure and acceptance of exceptional regulations Table 7.1: Amount of exceptional regulations used in the two 160 km/h solutions 160 km/h final solution 160 km/h minimum effort Cant 4 5 Transition cuves Cant deficiency 5 6 Element lengths 7 8 Total of exceptional regulations used Table 7.2: Infrastructural changes according to the two 160 km/h solutions 160 km/h final solution 160 km/h minimum effort Curves 3 2 Transition curves Upgrade of Existing Track to 200 km/h The speed upgrade scenario for 200 km/h is now considered. In relation to the final speed profile for 160 km/h in Figure 7.5, it was seen that a speed of 160 km/h could not

125 CHAPTER be obtained between Køge Nord and Køge, and between Næstved Nord and Næstved. The speed upgrade to 200 km/h therefore only considers the line section from Køge to Næstved Nord. The speed upgrade method used for the 200 km/h alternative differs from the earlier method, by upgrading from the 160 km/h alternative solution found in Figure 7.5, in stead of the existing infrastructure. From Figure 7.7, it is seen that a speed of 200 km/h can be obtained on the entire line, except at three sections, if exceptional regulations are applied. However, the amount of exceptional regulations used is high, and the infrastructure must be upgraded, to allow a smoother profile. The procedure follows more or less the procedure in the 160 km/h alternative, however, travel time savings are not considered to the same extent, as the infrastructure is adjusted in the major part of the curves, see Appendix J and the enclosed drawing TCE placed in Appendix P. Speed Profile Difference in Rail Net Denmark's Standard Regulations and Exceptional Regulations KjN Ølb Kj Hf Th Hz Ol Næn Næ Speed [km/h] Stationing [km] Stations 160 km/h næstved Standard Regulations Exceptional Regulations Figure 7.7: Differences between use of Rail Net Denmark s standard regulations and exceptional regulations The speed profile is adjusted by changing the geometry according to the prioritised method from section 7.2. To obtain a relatively smooth profile, the use of exceptional regulations are accepted to a greater extent, as in the previous alternative. The final smoothened profile is illustrated in Figure 7.8. Exceptional regulations are applied in the areas of Herfølge and between Haslev and Holme-Olstrup. The area of Næstved and between Køge Nord and Køge is not upgraded further. The amount of exceptional regulations used as well as the infrastructural changes is summarised in the following section, in relation to a comparison with a scenario of upgrading to 200 km/h with minimum effort. The travel time savings for the 200 km/h alternative is now considered. From Table 7.3 it is seen, that the largest travel time savings are, as expected, obtained for the 200 km/h solution, furthermore, the savings for the two line variants are more or less equal. For line variant 35/36 only small travel time savings are obtained as expected due to the stopping pattern.

126 110 TRACK GEOMETRY SOLUTION Smoothened Profile between Kj and Næ for the 200 km/h Solution with Uptimised Geometry and Acceptance of Exceptional Regulations KjN Ølb Kj Hf Th Hz Ol Næn Næ Speed [km/h] Stationing 200 km/h Smoothened Profile Exceptional Regulations Stations Figure 7.8: Final solution for the 200 km/h alternative with identification of the use of exceptional regulations Table 7.3: Comparison of travel time savings for line variants 20 and 21 considering the two track solutions 200 km/h solution compared with existing infrastructure compared 160 km/h solution compared with existing infrastructure compared Solution for 160 km/h compared with 200 km/h solution Line variant 20 Line variant 21 Line variant 35/ minutes 3.67 minutes 0.7 minutes 1.47 minutes 1.49 minutes 0.1 minutes 2.26 minutes 2.18 minutes 0.6 minutes In a further analysis, it would be interesting to investigate the different sections of possible speed upgrades according to the construction costs and time savings. A socioeconomic analysis considering the number of passengers on the line and the respective time values, would clarify whether the speed upgrade could be justified economically. However, such analysis is considered outside the scope of present report. Instead a simple analysis of cost and benefit is applied by estimating the overall construction cost in chapter 8 and chapter Upgrade to 200 km/h with Minimum Effort As for the 160 km/h, the 200 km/h final solution is also investigated to clarify the minimum required changes in the infrastructure with acceptance of exceptional regulations used. Figure 7.9 illustrates the speed profile and the locations of use of exceptional regulations. No distinction from the 200 km/h final solution is made in the areas between Køge Nord and Køge, and at Næstved station.

127 CHAPTER Table 7.4 and Table 7.5 summarise the amount of exceptional regulations used in the two solutions, and the type of infrastructural changes. It is seen, that the amount of exceptional regulations used is more than doubled, in the scenario with minimum effort. It is especially exceptional regulations used in transition curves, which contribute to this. Also, the violation in cant deficiency is doubled. However, comparing with the requirements of the infrastructural changes, it is seen, that the amount of changes needed less than half in the minimum effort solution compared with the final solution. The solution for 200 km/h upgrade with minimum effort is placed in the enclosed CD. Again, it is assessed that the profile with minimum infrastructural requirements will not be approved, due to the large extent of exceptional regulations used. Therefore, this speed profile is not considered in the remaining project, and it is therefore the final solution speed profile for 200 km/h, which is being priced in chapter 9. Speed Profile for 200 km/h with Minimum Change in Infrastructure and Acceptance of Rail Net Denmark's Exceptional Regulations KjN Ølb Kj Hf Th Hz Ol Næn Næ Speed [km/h] Stationing [km] Stations 200 Exceptional Regulations Smoothened Exceptional Regulations Figure 7.9: Smoothened speed profile with minimum change in infrastructure and acceptance of exceptional regulations Table 7.4: Amount of exceptional regulations used in the two 200 km/h solutions 200 km/h final solution 200 km/h minimum effort Cant 4 4 Transition cuves Cant deficiency 7 13 Element lengths Total of exceptional regulations used

128 112 TRACK GEOMETRY SOLUTION Table 7.5: Infrastructural changes according to the two 200 km/h solutions 200 km/h final solution 200 km/h minimum effort Curves 11 4 Transition curves Comparison of Danish and EN Exceptional Regulations To investigate the differences between the Danish and EN exceptional regulations presented in section 5.2, the optimisation model is run with inputs of the existing infrastructure. The target speed is set to 160 km/h and the model allows exceptional regulations used in all curves. The speed profile in Figure 7.10 shows the differences in allowed speed according to exceptional regulations. The existing speed profile is shown in grey, the speed profile according to Rail Net Denmark s exceptional regulations are shown in blue, and the EN profile is yellow. In general, the EN exceptional regulations are less strict, however, in few cases it is the other way around. Deviations between the two speed profiles can be seen in the areas near Køge Nord, Ølby and Køge, and also near Næstved. In one of four cases the Danish requirements allow a higher speed than the EN. The reasons to the four deviations are summarized in Table 7.6. Comparing the use of exceptional regulations, only little difference is obtained with the allowance of EN exceptional regulations. Primarily, the differences are caused by the EN regulations allowing a higher values for di dt. The Danish requirements allow a larger cant deficiency in turnout, which is effect at Ølby station Speed Profile for 160 km/h Comparing EN Exceptional and RND Exceptional KjN Ølb Kj Hf Th Hz Ol Næn Næ Speed [km/h] Stationing [km] Stations Existing Speed Profile 160 km/h RND Exceptional Regulations 160 km/h EN exceptional regulations Figure 7.10: Comparing EN exceptional regulations with Rail Net Denmark s exceptional regulations

129 CHAPTER Table 7.6: Deviations in speed profiles when accepting Danish and EN exceptional requirements Difference in speed profiles (EN and Danish exceptional regulations) Location/difference Rail Net Denmark EN Køge Nord, Køge, Næstved di/dt 90mm/s di/dt 100mm/s f ori 168mm Ølby, turnout I 150mm I 130mm f orv 160km/h 7.7 Summary of Track Geometry Solution Preceding chapter covers the track geometry solutions obtained in the optimisation model, followed by a further adjustment in the track design. Two alternative track solutions are designed for a speed of 160 and 200 km/h respectively. Rail Net Denmark s standard regulations are applied, and a minimum of exceptional regulations are used to obtain a smooth speed profile. The upgrading process considers a prioritisation for initiatives according to cant sizes, adjustments in transition curves and finally adjustments in curve radii. Comparisons of the travel time savings in the two speed solutions, result in savings in the interval from 3.72 to 1.47 minutes for the line variants 20 and 21. The freight trains and the line variants 35/36 do not gain travel time savings in the speed upgrade, due to lower speed and stopping pattern. The final solutions for 160 and 200 km/h are further investigated, to clarify the minimum infrastructural changes required to obtain the specific speed profile. In general, the infrastructural changes can be halved, however, the amount of exceptional regulations used are doubled equivalently. The solutions with minimum changes in the infrastructure are not investigated further in the remaining analysis. The subsequent chapter 8 will present the project description for implementation of the two final speed profile alternatives.

130 114 TRACK GEOMETRY SOLUTION

131 Chapter 8 Project Description Based on the geometrical design for the two upgraded speed speed profiles found in chapter 7, the following chapter presents the project description for the upgrade and the implementation of a second track with similar geometrical characteristic. The chapter forms the foundation for the construction estimate by describing the overall type of works to be carried out. Based on this, an evaluation of the two speed scenarios investigated in chapter 4 and in chapter 7 will be carried out in relation to construction cost. The chapter is divided in 12 sections in accordance with Rail Net Denmark s structure for New Budgeting. In the respective sections, parallels are drawn to the relevant points in the 29 point list. Initially, the financial setup for budgeting is described followed by the different assumptions made for the project itself and the project description. Then follows the project description divided according to the disciplines. 8.1 Financial Setup and Budgeting In 2006, the Ministry of Transport presented a new procedure, known as New budgeting, for assessing the construction cost of renewal and new built lines. The main purpose with New budgeting is to base current and future budgeting of projects on experiences from already implemented projects. This can be fulfilled by establishing and maintaining a cost database to be applied in future projects. By allowing transparency between estimated construction costs throughout the different phases of a project, larger cost certainty can be obtained. A prerequisite is the structure of projects, which must be standardised to allow for comparisons between projects. (Brandt, M., Morberg, 2010). In relation hereto, a project and the cost estimate must be structured according to 12 different disciplines. To structure the different phases a project undergoes, Rail Net Denmark follows the Phase model, see Figure 8.1. The phase model describes five phases a project is undergoing, from initial idea to evaluation of the execution. As the project is moving through the phases, the level of detail is intensified in accordance with the phasing out of uncertainties. The model is set up to handle hazards and uncertainties in the construction estimates already in the initial project phase. In phase 1, the budget is added a correlation supplement of 50 %, whereas a supplement of 30 % is added in phase 2. In addition hereto, a cost accounting supplement is added to the basis cost estimations, to compensate for the possible lack and uncertainties of identified quantities. The supplement varies according to the disciplines, and experience based supplements from earlier projects can be applied. The current project is defined as a screening project, and is therefore located before the definition phase. To compensate for the various uncertainties, it has been decided to add 50 % 115

132 116 PROJECT DESCRIPTION in correlation supplement. Cost accounting supplements are not added in the screening phase.(brandt, M., Morberg, 2010) Figure 8.1: Rail Net Denmark s phase model, based on (Rail Net Danmark and The Danish Transport Authority, 2010) 8.2 Assumptions for the Project Description The following section covers the different assumptions made for the project description. Traffic assumptions are made to clarify the use and requirements for the line, both for normal operation and contingency operation. Besides the traffic assumptions, general assumptions for the speed upgrade and construction of second track are listed. Delimitations and interface projects are furthermore given, to clarify the context of the project and the parameters not covered. To estimate the cost of the different disciplines, experiences from other similar projects are considered. All projects have been carried out by Grontmij Traffic Assumptions Before attending to the different types of work to be carried out, the traffic assumptions and the project interfaces are clarified. The traffic assumptions are defined to clarify the different requirements set for the line, and forms the background for the upgrade and the design of the line. These lead directly to the investigations of the capacity utilisation of the network performed in chapter 4, for upgrading the speed, extending the line to double track, and relocating traffic. In general, no distinction in assumptions is made between the scenario for 160 km/h and 200 km/h. The traffic assumptions are listed below. The existing stopping trains, line 35 and 36, run with 30 minutes frequency 2 additional passenger trains (line 20 and 21) run with 60 minutes frequency The line is prepared for 2 freight trains in each direction every hour 2 crossovers on each side of Haslev are implemented, to promote flexibility and allow for overtaking of freight trains Local trains coming from Hårlev will connect to the northern part of the Lille Syd line between Ølby and Køge Nord For contingency operation, a 30 minutes frequency is maintained for line 35 and 36, and the remaining lines are relocated via Ringsted

133 CHAPTER Delimitations, Interfaces and Assumptions The project is in a screening phase and clear delimitations are stated to account for interfaces and parameters not included. A number of project interfaces are present in current stage, and assumptions in relation to the execution of these are made. These and other assumptions in relation to chapter 5, forms the background for the upgrade of the existing track and construction of the second track on the line. The following list contains the project delimitations, the interface projects and the assumptions forming the foundation for the project description. The project considers the upgrade of the existing line from stationing to The second track is established from stationing to before Køge station and then in stationing to just before Næstved station The second track is coupled to the existing second track at the stations The project does not consider Køge and Næstved station The second track is coupled to existing tracks in both ends of Køge station and to a dead end track at Næstved station An additional siding track is constructed at Køge Nord station to allow for overtaking of freight trains running towards Copenhagen The siding track at Køge Nord is not covered in the project description or in the cost estimate The coupling between Lille Syd and the new line Copenhagen-Ringsted is not considered The new Signalling System is implemented on the Lille Syd line The line has been electrified and upgraded in accordance with the Programming Report from 2014, (Grontmij, 2014) The TSI requirements and the Danish national norms are complied with in the project Inspiration and Procedure for Upgrading Projects In accordance with the numerous number of speed upgrades in the Danish railway network, a list of 29 points to take into consideration has been formulated by Rail Net Denmark. The list is not formal, and is prepared as a checklist for upgrading projects. According to the 29 point list, not all points are considered in present project. Table 8.1 shows the relevant points and the respective norms. The project description forms the basis for the construction cost estimate of the two investigated scenarios. Inspiration and experiences are sought from different, more or less similar, projects conducted by the consultancy firm Grontmij. The construction cost estimations are based on experiences from these projects. In general, five projects have been considered;

134 118 PROJECT DESCRIPTION Table 8.1: The relevant points in present project from the 29 point list, based on (Grontmij, 2014) Point Title Norms (Danish termination) 1 Track geometry Sporregler Track design Sporregler Ballast profile BN , Structure gauge Fritrumsprofiler Platforms BN1-9-2 and BN Embankment BN Catenary System FKI 27 Interlocking - 28 Crossings - The electrification and speed upgrade to 160 km/h of the existing Lille Syd line from 2014 The speed upgrade to 200 km/h of the line Hobro-Aalborg from 2014 The construction of a second track on the line Vamdrup-Vojens from 2012 The speed upgrade to 200 km/h on the line Ringsted-Odense, 2015 The upgrade and reconstruction of Hillerød station, 2015 The cost estimation is based on unit prices given according to the different disciplines. An assessment of the different projects and available unit prices is conducted, to create the best possible frame of reference for the cost estimate. Comparisons with the different projects are evaluated continuously in the project description. 8.3 Overview of the Disciplines The remaining part of the section is presented according to the 12 disciplines described in New budgeting. Upgrading projects are normally composed according to Rail Net Denmark s 29 point list introduced in chapter 5. The list is used to ensure that all relevant parameters are considered within a speed upgrade project. However, due to the fact, that current project covers both a speed upgrade and the implementation of a second track, it has been decided to structure the project description according to New budgeting. In addition hereto, it is decided only to glance at the 29 point list, to ensure that the most relevant points are considered, due to the project being in the initial screening phase. Table 8.2 lists the 12 disciplines and the relevant points from the 29 point in relation hereto Track Realising the traffic foundation described above, requires the Lille Syd line to be upgraded from single to double track. Furthermore, the possibility of upgrading the line speed to 160 and 200 km/h, respectively, is investigated. Based on this, a project description of

135 CHAPTER Table 8.2: Disciplines covered in the project description Disciplines in New budgeting Relevant points in the 29 point list 1. Track Earth Work Bridges and Constructions 15, (28) 4. Electrification System Power Supply - 6. Interlocking and Remote Control 27 (28) 7. IT, Tele and Transmission Systems - 8. Buildings Areas Forestry Additional Considerations Cross-disciplinary Costs - the track technical initiatives is conducted, to realise the upgrade for both solutions presented in chapter 7. This includes initiatives required for constructing the new second track and upgrading the existing. The project description is formulated in accordance to Rail Net Denmark s current requirements presented in chapter 5. In general, the construction of the new track must comply with the requirements stated for new built line, while the existing track must comply with the requirements for upgrades. The sections below contain short descriptions of the track related discipline and the associated assumptions. The drawing TCE in Appendix P show an overview of the required track work in a schematic track layout of the future Lille Syd line. The different activities are illustrated on numbered bars below the schematic track layout, to provide a geographical overview and illustrate the correlation between different activities. Furthermore, the drawing serves as a tool for estimating the quantities of the total track work. Finally, it enables a direct comparison between the 160 and 200 km/h solution Track Layout The track layout for the double track Lille Syd line is sketched in the top of drawing TCE No distinction in layout is made for the 160 and 200 km/h solution. In general, the location of the new second is based on examinations of overview maps. At crossing stations, all turnouts are removed and the existing track 2 is displaced due to a relocation of the existing platforms, cf. section 8.6 below. Figure 8.2 is an extract of the schematic track layout at Haslev Station. Here, four new crossovers are constructed with a distance of 1,400 metres to enable the possibility of overtaking freight trains. These furthermore enable operation of two passenger stop trains per hour in each direction to be retained during contingency operation.

136 120 PROJECT DESCRIPTION Figure 8.2: Extract of Haslev Station from the schematic overview plan TCE placed in Appendix P showing the location of new crossovers. New infrastructure (green), displaced infrastructure >200 mm (dashed blue) and removed (red) Furthermore, a crossover east of Ølby Station is installed to ensure connection to Roskilde for the local line. Finally, two crossovers and one turnout at Køge Station are installed to couple the new track to the local line. The interconnection between the Lille Syd line and the local line is not treated. This must be investigated in a separate project. The total track work shown on the schematic overview plan in drawing TCE is summarised in Table 8.3. Table 8.3: Summary of overall track work in relation to extension of the Lille Syd line to double track Both solutions Comment New Track 37.5 km 1) 1) 33.5 km for 200 km/h and 4 km for max 160 km/h New Turnout 2 unit see schematic drawing for location New Crossover 7 unit see schematic drawing for location Removed Turnouts 8 unit see schematic drawing for location Track Adjustment 47.3 km see bar Speed Profile and Curvature Point 1 in the 29 point list treats the speed profile in relation to the horizontal track geometry investigated in chapter 7. The proposed speed profile and associated curvature for the 160 km/h and 200 km/h solutions are sketched in bar 1 and 2 in the drawing TCE The extent of changed curvature is illustrated in bar 3 and the total estimate is summarised in Table 8.4. The displacement of the existing track caused by straightening of curves is shown in bar 6 and summarised in Table 8.4. According to guidelines stated by Grontmij in the Hobro-Aalborg and Ringsted-Odense speed upgrade projects, it is assumed that track displacements cause the following activities (Grontmij, 2015b);

137 CHAPTER mm displacement: realised in regular track adjustments mm displacement: realised in extended track adjustments, still changes are carried out within the existing track bed mm displacement: reconstruction of drainage >200 mm displacement: reconstruction of drainage and new track bed The displacements caused by straightening of curves are in general located within the interval of mm, however, approximately 25 metres of track east of Køge is displaced 299 mm. Furthermore, it is presumed that the existing track 2 in the station areas is displaced due to reconstruction of the platforms. No exact geometrical solution is made for this displacement. Instead, it is simply assumed that the track is displaced more than 200 mm, hence new track bed is required. It will be necessary to clarify the exact track geometry in station areas in later phases. It is assumed that track adjustments are necessary on the entire existing line due to the amount of track work carried out, illustrated in bar 8 in drawing TCE In a later phase it is assessed, if this can be spared at sections where no track work is carried out. Table 8.4: Summary of extent of changed curvature and displacement of track sketched in TCE km/h 200 km/h Comment Changed Curvature 7 km 17 km See bar 3 Track Displacement See bar mm 45 m 300 m mm 35 m 200 m mm 0 m 0 m 200 mm 3,100 m 3,100 m The geometry of the new track is considered outside the scope of this screening phase. Instead it is assumed, that the new track is built with same curvature and speed profile as the existing. In reality, all new lines are projected according to Rail Net Denmark s requested regulations in TER chapter 2, presented in the norm foundation in section 5.2. At a later project stage, it will be necessary to project the alignment of the new track, to ensure that the required speed profile can be achieved. This applies especially for the cases, where the new track is placed on the inside of a curve in relation to the existing track Superstructure The superstructure has the purpose of retaining the track, transferring energy from the rolling stock to the subsoil, and ensuring proper drainage of the track bed, see section 5.3. Increased line speed results in larger tensions in the track bed and increases the demand

138 122 PROJECT DESCRIPTION for the superstructure. The 29 point list therefore contains several points for investigating whether or not the superstructure can handle the increased speed. In addition hereto, a new track must be constructed according to the design speed. The superstructure consists of the track construction (rails, sleepers, fastenings) and the ballast profile. In the norm foundation in section 5.3, is only described the ballast profile, however still activities for exchanging the track construction is included in the project description. Below are the steps required for the superstructure described, first in relation to the track construction and subsequently in relation to the ballast profile. These are described in relation to the upgrade of the existing track to 160 and 200 km/h respectively, while constructing a new second track Track Construction - Upgrade The points 2-9 of the 29 point list ensure that the rails, sleepers and fastenings can handle the desired speed profile. The track construction in the base scenario is shown in bar 5.1 in drawing TCE This is based on the project material provided by Grontmij in the Programming Report for the speed upgrade and electrification of the Lille Syd line prepared for year 2018 (Grontmij, 2014). The line speed is therefore already upgraded to 160 km/h on the open line. According to the Programming Report, approximately 14 km track was exchanged with the track construction UIC60 Dmp during the track renewal carried out in year This track construction is composed by UIC60 rails and S99 sleepers, which Rail Net Denmark in general uses for track renewal. It is furthermore stated that approximately 12 km track consists of the type DSB60 Db Steel 700, which is an old rail from the 1970s placed on duo block sleepers. This is not changed in relation to the 160 km/h speed upgrade investigated by Grontmij. Yet, Rail Net Denmark has decided to change the track construction in a later separate project according to the Programming Report (Grontmij, 2014). It is therefore assumed, that this track construction will be replaced with UIC60 Dmp before present project is carried out. Thus, replacement of this is not included in the project description. Given all turnouts on the section between Køge and Næstved are removed as a consequence of the new second track, the existing turnouts are not addressed in the upgrading project. According to the basis for decision of the Lille Syd line speed upgrade, additional calculations show, that it will be possible to allow an axle load up to 22,5 tonnes on the line, contrary to the proposed 18 tonnes in the Screening Report (Rail Net Denmark, 2013d). It is therefore assumed, that this is valid for the 160 km/h scenario. For the 200 km/h scenario, additional investigations in later phases will determine, whether an axle load of 22,5 tonnes will be approved at this speed. This concerns especially the load capacity at bridges and embankments. The axle load is not considered in current project.

139 CHAPTER Track Construction for 160 km/h For the 160 km/h solution, the track construction is only investigated where the line speed is increased, hence in all station areas and at Køge and Næstved Station. DSB45 rails with steel quality 700 are exchanged with the track construction UIC60 Dmp in accordance with the track renewal project in The changes to the track construction required for the 160 km/h solution is shown in bar 5.1 in drawing TCE and summarised in Table 8.5. Track Construction for 200 km/h The 200 km/h alternative contains many uncertainties, given that the Danish railway network currently only comprehend maximum speeds of 180 km/h. This is about to change with realisation of the One-Hour Model between Copenhagen and Odense. The Ringsted-Odense speed upgrade project is preparing an upgrade of the line to 200 km/h at minor line sections. This upgrade has evoked a debate in relation to approving the existing track construction to 200 km/h. In the project foundation for the Ringsted-Odense Programming Report, the following assumptions are outlined (Grontmij, 2015b); S89 and S99 sleepers are demonstrated for 200 km/h with an axle load of 22.5 ton. The existing rail pad is exchanged for S89 sleepers. The final approval process remains unsolved for the CSM process of the Detailed Design New S16 mono block sleepers are approved for 200 km/h All wooden and duo block sleepers are exchanged with new S16 mono block sleepers In current project, it is decided to apply same assumptions as stated in the Ringsted- Odense project. However, since the final approval process for S89 and S99 sleepers remains unsolved, this assumption will result in a large project risk. Track specialists at Grontmij have pointed to the uncertainty in the cost of S16 sleepers, which can result in an additional cost of 200 DKK/sleeper according to sources at Rail Net Denmark, opposed to regular S99 sleepers. This results in an increased unit cost of 350 DKK/metre track for the new track construction for 200 km/h. It is decided to include this additional cost in present project, to emphasize the difference for the 160 km/h and 200 km/h upgrade. Attention is, however, drawn to the large uncertainties in the estimated unit cost, which must be controlled in subsequent phases. The changes to the track construction in the 200 km/h solution is shown in bar 5.2 in drawing TCE and the quantities are summarised in Table Track Construction - New Track The new track is constructed with the track construction; UIC60 Dmp, in accordance with the track construction used in the track renewal project in 2013 (Grontmij, 2014). For the 200 km/h solution are the S99 sleepers replaced with new S16 sleepers, as described above. The unit cost is increased 350 DKK/meter track for 200 km/h, to account for the additional cost of purchasing S16 sleepers. The location of the new track is shown in bar 5.2 in drawing TCE

140 124 PROJECT DESCRIPTION Table 8.5: Summary of the amount of track construction exchanged in the existing track according to the solutions sketched in drawing TCE , see bar 5 Track Construction 160 km/h 200 km/h Comment Exchange All 6.5 km 6.5 km Normal track construction UIC60 Dmp for 160 km/h and new S16 mono block sleepers for 200 km/h Exchange Sleepers 0 km 0.65 km S16 mono block sleepers for 200 km/h Exchange Track Pad 0 km 2.4 km Valid for existing UIC60 Dm Ballast Profile Another decisive factor for whether or not the superstructure can handle the increased line speed is the ballast profile. This is treated as point 10 in the 29 point list. The ballast profile must comply with the BN1-6-5 requirements presented in section 5.3, which differs according to upgrades and new built lines. Furthermore, the requirements differ for speeds up to 160 km/h and from 160 to 200 km/h. As a basis for both solutions, it is assumed that the existing track complies with all valid requirements listed in BN1-6-5 for speeds up to 160 km/h, due to the large modernisations the Lille Syd line is undergoing in present years. No adjustments are therefore made to the existing ballast profile beyond the activities directly caused by the speed upgrade or implementation of the new second track. This assumption is investigated further in a later project phase. Ballast Profile for 160 km/h Figure 8.3 shows a principal cross section of the 160 km/h solution without cant size, according to the requirements for BN1-6-5 stated in section 5.3 and the minimum track center distance according to section 5.4. The entire drawing is referred to as TSPR 3 NXS 001 and placed in Appendix P. As shown in Figure 8.3, the cross section of the existing track is retained, due to the fact that the same requirements apply for both speeds of 120 km/h and 160 km/h. The inclination of the top of subsoil and subballast induce, that the track bed is drained to both sides, hence towards a ditch and through the new track. This is allowed when applying the same type of subballast in the new track as the existing, and when the new track is placed so the top of subsoil and subballast is 100 mm below the corresponding layer of the existing track according to BN This solution is chosen because it involves a minimum of changes to the existing track. This is found desirable, due to the current modernisation of the existing track in combination with the present speed upgrade and electrification. In a later phase, it will be demonstrated whether or not, this is the best technical solution.

141 CHAPTER New Track L C Existing Track C L 8.5 meters Fill 3800 Fill Estimated embankment height 3000 a= 2.5 a= meters a= Ex. ballast 300 mm Ex.subballast 200 mm Subsoil Estimated ditch height 3000 Existing boundary a= meters Existing boundary Existing terrain New ballast 350 mm Newsubballast 200 mm Figure 8.3: Principal cross section for the 160 km/h solution. For entire drawing see TSPR 3 NXS 001 in Appendix P Principal shape of existing ditch Ballast Profile for 200 km/h 5.1 meter Existing boundary C L C Existing terrain til 550 mm, according to the project foundation of the Ringsted-Odense L speed upgrade, Cut a= 1.5 When increasing the line speed to 200 km/h, different requirements apply for the ballast profile. The thickness of subballast is increased from 200 mm to 300 mm according to New Track Existing Track BN1-6-5 described in section 5.3. Furthermore, the ballast shoulder is increased from b). A principal cross section for the 200 km/h solution is shown in Figure Cut 8.4. The 3800 entire drawing is placed as SPR 3 NXS 002 placed 400 in Appendix 400 P meters where it is stated, that this will be required in a future edition of the BN1-6-5, (Grontmij, a= 1.5 Fill a= a= meters New Track 3000 L C a= 1.5 New ballast 350 mm Newsubballast 200 mm Ex. ballast 300 mm Ex.subballast 200 mm Subsoil Existing Track C L Fill Estimated ditch height 3000 Principal shape of existing ditch When increasing the thickness of the subballast, the top of slope is changed from the center line to the left track so the entire track bench is drained to right ditch instead of through the new track Existing boundary a= 1.5 Estimated ditch height Fill Area Cut Area Approximately 7.5 meters a= 2.5 Existing boundary a= Existing terrain Scale 150 Revision 40 PRINT DATO: a= 1.5 Drawing number 1:100 A TSPR_3_NXS_ Ballast 300 mm Subballast 300 mm Subsoil New ballast 350 mm New subballast 300 mm Subballast prolonged to existing ditch Figure 8.4: Cross section showing ballast profile for the 200 km/h solution. For entire drawing see TSPR 3 NXS 002 in Appendix P Estimated ditch height 3000 CADFIL: R:\Projects\GLO\22\ \Misc\MAI\Speciale\Tegn\Tværsnit.dgn Existing boundary Page/Pages 1/1 Principal shape of existing ditch ly 8.5 meters terrain xisting boundary Cut New Track L C Existing Track C L Cut When increasing the thickness of the subballast, the top of slope is changed from the center line to the left track so the entire track isting boundary

142 126 PROJECT DESCRIPTION To increase the thickness of the subballast, it is necessary to remove the entire superstructure. In combination hereto, it is decided to surface the top of subsoil so the existing track bed is drained to a ditch, instead of being drained through the new track. The new track can then be placed in the same level as the existing. In a later phase it is investigated, whether this is the best technical solution. Changing the ballast profile according to the regulations for 200 km/h requires a supplement of both ballast and subballast, as indicated in Figure 8.4. The amount of additional ballast and subballast is estimated from the cross section area along the changed sections of the railway line. Optimising Cant The cross sections for the 160 and 200 km/h solution in Figure 8.3 and Figure 8.4 show the situation without cant size. In case of cant, the ballast shoulder shall be parallel to the sleeper according to BN Obtaining this profile requires additional ballast, hence the optimisation of cant size causes a ballast supplement. This is only valid for the existing track, given the unit cost for the new track is provided in cost per metre track. An example of this case is shown in Figure 8.5, where the cant size is increased from 0 to 80 mm (for entire drawing see TSPR 3 NXS 003) in Appendix P. A rough estimate of the required ballast supplement is found by multiplying the cross section area of the additional ballast shown in Figure 8.5 with the amount of changed curvature according to bar 5 in drawing TCE It is expected that this will lead to an overestimate, given that the cant size in most cases not are changed 80 mm. However, it is assessed that this overestimate compensate for the fact, that some of the ballast might be damaged in the construction phase. In a later project phase it will be necessary to make an earth model, to estimate the requirement for ballast supplement more precise. Extending track bench (Sketched for 160 km/h) from 3.15 meters to 3.30 meters 150 Existing boundary New boundary Existing Track C C L L Fill 3300 Cant Ballast supplement Subballast supplement Ex. ballast 350 mm Ex.subballast 200 mm Subsoil Figure 8.5: Cross section showing required ballast supplement and additional subballast when increasing the track bench 150 mm as a consequence of increased cant size. For entire drawing see TSPR 3 NXS 003 in Appendix P Existing boundary 150 New boundary Railway line placed at platform 200 km/h Furthermore, the change in cant size can also result in an extension of the track bench, Existing Track Existing Track because a supplement moved of either (> mm) mm or 300 moved mm is (<200 required mm) for cant sizes in the interval of 5-80 mm and C mm, respectively. Figure L C 8.5 likewise shows an example of L this. By comparing the existing and changed curvature in the two solutions sketched in Distance Distance Open space Safety zone 0 tform ight atform eight 0 Safety zone Open space

143 CHAPTER bar 1 in TCE , it is possible to identify the locations where the track bench is extended. It is found, that the track bench only is extended 150 mm. Places where this occur are sketched in bar 4 in TCE The changes to the ballast profile is summarised in Table 8.6 according to the assumptions stated above. Table 8.6: Summary of changes to the ballast profile based on enclosed cross sections and the schematic overview plan placed in Appendix P New track 160 km/h 200 km/h Comment Construction of new subballast 41,250 m 3 69,725 m 3 Cross section area; 1.10 m 2 (160 km/h) 1.95 m 2 (200 km/h) Upgrade Track Bench Extended 150 mm 375 m 630 m See bar 4 Subballast supplement 19 m m 3 Based on cross section area 0.05 m 2 Change of cant size See bar 3 Ballast supplement 875 m 3 2,125 m 2 Based on cross section area m 2 Extending ballast shoulders 0 m 3 8,375 m 3 Based on cross section area 0.25 m 2 Subballast Supplement 0 m 2 23,450 m 3 Based on cross section area 0.7 m Earth Work Extending the Lille Syd line to double track and upgrading the line speed require a great amount of construction work, in relation to constructing the substructure, drainage via ditches or pipes, and shaping the embankments. This is, from now on, referred to as earth work. In general, this has not been the main focus of present project and the norm BN for drainage has therefore not been introduced in the norm foundation in chapter 5 (Rail Net Denmark, 2006b). Still, it is necessary to outline a general work description to estimate the cost. The estimation of earth work is therefore based on a series of rough assumptions. These assumptions constitute to one of the largest project risks, because the amount of earth work makes up a large proportion of the total cost estimate. Experiences from other projects have shown, that unknown soil conditions, such as soft and polluted soil, have had a massive impact on the final cost estimate Soil Handling - New Track Evaluation of soil handling and earth balance require information of the level of the new track, the existing terrain, and the soil conditions. In present project, the earth balance is evaluated based on rough estimates of average embankment heights and slopes. These are estimated according to guidelines provided by earth work and railway specialists at Grontmij, and from own observations during the inspection of the Lille Syd line carried

144 128 PROJECT DESCRIPTION out the 1st of June, The inspection concluded, that the Lille Syd line in general is constructed in fill, yet, in a relatively plain terrain. Only at Ølby station, is a large railway embankment observed. Based on this, a general estimate is made; 80 % of the line is constructed in fill 20 % of the line is constructed in cut The line is drained to ditches on both sides, where the new track is constructed. Experiences from the Copenhagen-Ringsted project provided by specialists at Grontmij states, that an average embankment height varies from 3-4 meters. It is furthermore recommended, that the slope of the railway embankment is set to 1:2 or 1:3, when the earth conditions are unknown. Based on these guidelines and own observations, the following assumptions are drawn; Average embankment height: 3 metres Slope of railway embankment: 1:2.5 Based on these assumptions and the requirements for new built ditches according to BN1-11-1, the embankment and ditch profile can be drawn for both the new and existing line. It is assumed, that the railway embankment of the existing line is constructed with a slope of 1:1,5 and the average height of 3 metres. The principal cross section for the 160 km/h and 200 km/h solution constructed in either cut or fill, are shown in drawing TSPR 3 NXS 001 and TSPR 3 NXS 002 placed in Appendix P. An extract of the embankment profile for the new track constructed in fill is shown in Figure 8.6. Based on the cross section area of the fill (green hatched) and cut (red hatched), the amount of cut and fill is estimated respectively. No adjustments are made in relation to the compression factor. Furthermore, it is assumed that 50 % of the excavated subsoil can be built into the railway embankment according to experience provided by specialists at Grontmij. A layer of 30 cm top soil is removed, when changing the layout of the top of subsoil. This is estimated according to the length of the new built track and the approximate distance measured from the end of the existing track bench to the new boundary, sketched in the enclosed cross sections. Furthermore, the top soil is removed when constructing temporary gravel access roads and working spaces. This applies for all temporary expropriated areas, which is treated in section Calculation of the earth balance is placed in Appendix K. Table 8.7 summarises the earth balance for both speed scenarios. Finally, the width of the new track bed is taken into account, to estimate the amount of subsoil to be surfaced in the 160 km/h solution. For the 200 km/h solution is the width of both new and existing track bed likewise applied. The calculations are shown in Appendix K and summarised in Table Soil Handling - Upgrade In addition to constructing the new track, earth work related activities can be induced by an upgrade of the line speed. Point in the 29 point list therefore concern the drainage condition, embankment stability, etc. Activities to repair the existing condition

145 ch for 200 km/h Fill CHAPTER New Track Existing Track Aproximately 13.1 meters L C C L ne placed in fill New boundary a= 1.5 a= 1.5 Estimated ditch height a= 2.5 Existing boundary a= Existing terrain Ballast 300 mm Subballast 300 mm Subsoil New ballast 350 mm New subballast 300 mm a= Approximately 7.3 meters ne placed in cut Figure 8.6: Principal cross section showing the ditch when the new line is constructed in fill. For entire drawing see TSPR 3 NXS 002 in Appendix P. Cut (red hatched) Fill (green hatched) New boundary Estimated ditch height nt are given in mm unless otherwise specified New Existing retained Approximately 8.5 meters Table 8.7: Summary of the earth balance for both Cut speed scenarios, see Appendix K Existing terrain a= 2.5 New Ballast Existing boundary a= 1.5 a= Fill Area 160 km/h 200 km/h Excavated top soil 222,690 m 3 183,887 m 3 a= Fill volume 310,125 m 3 372,100 m 3 Reclaimable cut subsoil 108,375 m 3 95,478 m 3 Purchased subsoil 201,750 m 3 276,623 m 3 Surface subsoil 424,500 m 2 918,500 m New Track can be required as well as activities caused through changes related to the upgrade, when displacing the track or extending track bench. In present project it is assumed, that the Lille Syd line complies with all current requirements according to BN No activities are therefore listed in relation to repair of existing conditions L C New ballast 350 mm New subballast 300 mm Existing Track C L Ballast 300 mm Subballast 300 mm Subsoil a= Existing removedthe embankment New Subballast and ditch profile is changed Cut Area in locations where the existing Scale track is moved more than 100 mm or where the track bench is extended, in accordance 1:100 with the Hobro-Aalborg project (Grontmij, 2015d). This is only valid on open line, given that drainage pipes are installed in station areas. Locations where this is valid are sketched in bar 7 in drawing TCE and summarised in Table 8.8. Revision A PRINT DATO: Figure 8.7 shows the theoretical change of the profile when extending the track bench in cut and fill respectively. However, this is assessed rather complicated to accomplish, and is therefore omitted. Instead, changes in the embankment require demonstration of the embankment stability, which can possibly lead to a total reconstruction of the profile. This has been found to have a critical influence on the final cost estimate, leading to a heavy debate of the current norm foundation and the procedure for determining the embankment stability, according to specialists at Grontmij. This discussion is, however, considered outside the scope of current project. Yet, it is acknowledged that the

146 r special cases r 160 km/h) 130 PROJECT DESCRIPTION Extending track bench (Sketched for 160 km/h) from 3.15 meters to 3.30 meters Table 8.8: Summary of drainage requirements according to drawing TCE km/h 200 km/h Comment Extending track bench (Sketched for 160 km/h) from 3.15 meters to 3.30 meters New ditch 36,5 km 36,5 km Next to new track Reestablished ditch 400 m 650 m see bar 7 New drain pipes 2,750 m 2,750 m see bar 7 unresolved 150 geotechnical condition constitutes to a high project risk, which should be addressed already in an early project Existing Track phase. Existing boundary New boundary C C L L Fill lement lement Cant Existing Track Existing Track 3150 Cut 400 Ballast supplement Fill C C L C L C L L Subballast supplement Existing boundary 150 New boundary 40 Cant Ballast supplement Subballast supplement Ex. ballast 350 mm Ex.subballast 200 mm Subsoil 40 Cant Existing boundary New boundary 40 Ballas Subballa 40 mm) 40 Ex. ballast 350 mm Railway line placed at platform 200 km/h Ex.subballast 200 mm Subsoil (a) Extension of track bench placed in cut Existing Track moved (>200 mm) Ex. ballast 350 mm Ex.subballast 200 mm Subsoil (b) Extension of track becnch placed in fill Existing Track moved (<200 mm) istance 945 Height Safety zone Safety zone 400 a= 1.5 Existing Siding Track moved (>200 mm) C L a= 1.5 Existing Track moved (<200 mm) C L Distance Distance Distance Front edge of new platform Front edge of new 945platform Front edge of new platform each side of the line Safety as shown Open space in the cross sections Open in drawing space Safety TSPR 3 NXS 001 and zone zone 550 Height allast 350 mm ubballast 200 mm ubsoil Area Open space s otherwise specified Area Ballast 350 mm Subballast 200 mm Subsoil New Ballast New Subballast Height 550 Fill Area Bridges and Constructions Cut Area Distance height Platform Drainage Ballast 350 mm Front edge of new platform Subballast 300 mm Subsoil 550 Existing Track moved (>200 mm) Open space Safety Safety Open space zone zone In the km/h solution, 400 the existing track is drained through the new 550 track as described height Platform C L Figure 8.7: Consequences for embankment and ditch profile when extending the track bench. For entire Railway line placed at platform 200 km/h drawing see TSPR 3 NXS 003 in Appendix P Drainage Revision PRINT DATO: and the respective roads, as well as passenger crossings at stations. This refers to point C L Scale height Platform Distance in ballast profile above. For the 200 km/h solution, both tracks are drained to ditches on TSPR 3 NXS 002 placed in Appendix P. Ballast 350 mm At station areas, it is assumed Front edge that of new theplatform sidings will be displaced more than 200 mm, which Subballast 300 mm Subsoil result in construction of new track beds. In relation hereto, a new drain pipe is installed in the center of the two tracks as shown in Figure 8.8. It is assumed, that the drainage conditions for the Drainage Ballast main 350 mm track can be retained. The location of new drainballast pipes 350 mm is shown Subballast 200 mm Subballast 300 mm in bar 7 in drawing Subsoil TCE A summary of the new drain pipes Subsoil is stated in Table 8.8. Scale Revision PRINT DATO: Drawing number :100 A TSPR_3_NXS_003 C L CADFIL: R:\Projects\GLO\22\ \Misc\MAI\Speciale\Tegn\Tværsnit.dgn Page/Pages The point bridges and constructions covers the out-of level crossings between the railway 1/1 Existing Track moved (<200 m Drainage Drawing number C L height Platform Dist Bal Sub Su BDK_Titl 1:100 A TSPR_3_ 15; structure gauges, points 23-25; bridges and structure gauges and point 28; crossings, in the 29 point list. In upgrading projects, much attention is drawn to the structure gauge CADFIL: R:\Projects\GLO\2

147 40 40 Ex. ballast 350 mm Ex.subballast 200 mm Subsoil CHAPTER Railway line placed at platform 160 km/h Railway line placed at platform 200 km/h Existing Siding Track moved (>200 mm) C L Existing Track moved (<200 mm) C L E m Distance Distance Dist Open space Safety zone Safety zone Open space Open space Safety zone Front edge of new platform 550 a= 1.5 Height 40 a= Height 550 Front edge of new platform Front edge of new platform Ballast 350 mm Subballast 200 mm Subsoil Drainage Ballast 350 mm Subballast 200 mm Subsoil Note: All measurement are given in mm unless otherwise specified Figure 8.8: Principal cross section at platforms for the 160 km/h solution. For entire drawing including New the 200 km/h scenario, see TSPR 3 NXS 002 in Appendix P Existing retained New Ballast Fill Area Existing removed New Subballast Cut Area of existing road bridges and pathways crossing the railway. In current project, however, additional attention is brought to the fact, that the line is extended to double track, hence reconstruction of bridges, in any form, is expected. Furthermore, by upgrading the speed at stations to 160 and 200 km/h, respectively, passenger crossings in-level is prohibited. The national and the TSI requirements described in section 5.5 is considered for both the upgrade of the existing track to 160 and 200 km/h, and out-of level platform crossings are established in both scenarios. Scale 1:100 Rev A Bridges In the speed upgrade and electrification project of the Lille Syd line from 2014, major construction work on the existing bridges is carried out. Due to the electrification of the existing line, larger structure gauges are demanded to compensate for the implementation of the overhead catenary system. The project from 2014 investigated all road bridges and pathways, to determine the ones which did not comply with the enlarged structure gauge, EBa, described in section 5.5. In these cases, the bridges were either closed, raised or replaced with new bridges. Alternatively, the track was lowered. In total, the number of bridges along the line was scaled down from 19 to 18. It was furthermore assumed, that all 29 railway bridges was maintained and no additional work was carried out here. (Rail Net Denmark, 2013d) In current project, it is assessed that the speed upgrade to 160 km/h on stations will not have an impact on the bridges. However, the speed upgrade to 200 km/h demands a structure gauge height of 5,780 mm, hence 300 mm higher than the EBa profile for 160 km/h, according to section 5.5. Given the fact, that both upgrading scenarios contain the implementation of a second track, the existing road bridges and pathways shall therefore be extended to handle two tracks. Extension of these is assumed too costly, hence new bridges are constructed in both scenarios. The following assumptions are taken in relation this;

148 132 PROJECT DESCRIPTION (a) Road arc bridge south of (b) Railway bridge leading across Køge river south of Køge Holme-Olstrup station, photo station, photo taken taken Figure 8.9: All road bridges and pathways are exchanged and railway bridges are extended The existing road bridges and pathways are demolished New road bridges and pathways are constructed in the same location but having twice the size of the existing ones The existing railway bridges are maintained New railway bridges, identical in size, for the second track are constructed In the 200 km/h scenario, two level crossings at Køge is maintained, due to the speed not being upgraded. Two level crossings are reconstructed to out-of level crossings, and finally one pathway is closed The existing bridges are expected to be investigated in a further analysis to clarify, whether they are able to handle a speed of 200 km/h. The sizes of the respective bridges are listed in the Programming Report and summarised in Table 8.9. All new road bridges and pathways are constructed in accordance with the EBa profile. The cost for the two new out-of-level crossings are estimated to 50 million, based on a unit cost provided in a special course at DTU in summer 2014 (Jensen, 2014a). Table 8.9: Size of respective bridges relevant for the construction of the second track, see Appendix L Amount Existing size [m 2 ] New size [m 2 ] Removal [m 2 ] Road bridges and pathways 18 5,343 10,686 5,343 Railway bridges 29 4,005 4,005 - Additional LC bridges Platform Crossings Platform crossings are coved by point 28 in the 29 point list and so are level crossings in general. Currently, in-level platform crossings are established at all intermediate stations

149 CHAPTER between Køge and Næstved station. In-level crossings are, however, not allowed for line speeds above 140 km/h. Out-of-level platform crossings are therefore established on all intermediate stations in both speed upgrade scenarios. Simple platform bridges in steal with associated elevators are installed. According to the TSI PRM, barrier-free paths are required between platforms and other facilities at the respective stations such information centres, waiting areas, etc (Den Europæiske Unions Tidende, 2014b). At Ølby station, the existing platform access is via a tunnel. In this case, the tunnel is prolonged to account for the second installed track, and the access ramp is replaced with an elevator. The amount of platform bridges and elevators are summarised in Table (a) In-level passenger crossing (b) Platform bridge with associated elevator, photo taken at at Tureby station, photo taken Skørping Station (Bevensee, 2015) Figure 8.10: Exchange of in-level passenger crossings with platform bridges and elevators Table 8.10: Installed platform bridges and elevators at the intermediate stations Amount Stations New platform bridges 5 Hf, Th, Hz, Ol and Næ Extension of tunnel [m] 5 Ølb New elevators 11 Hf, Th, Hz, Ol, Næ and Ølb Electrification System The electrification system considers the installation of an overhead catenary system and covers point 26 in the 29 point list. The type of electrification system relies on the expected line speed, and different systems and placements of masts are therefore expected for the two upgrade scenarios. The electrification of the line in relation to the speed upgrade, and the construction of the second track is handled separately.

150 134 PROJECT DESCRIPTION Electrification of the Existing Track in Relation to the Speed Upgrade The existing Lille Syd line from Køge Nord to Næstved is currently being prepared for electrification, and the process will be completed in According to the Programming Report, the type of electrification system is not yet determined, however, the system is being implemented for speeds of 160 km/h. The final placement of masts will be determined by the Electrification Programme in the Detailing Phase. However, the Programming Report suggests a shift in placement of the masts, to distance them from buildings as much as possible according to Figure 8.12 (Grontmij, 2014). Considering the construction of the second track, the placement of masts will most likely result in a significant adjustment of the newly implemented system, due to the interference with the second track. In addition, removal of existing turnouts and implementation of new, will also result in adjustments. Further investigations will reveal the type of adjustments, the implementation of the second track and a speed upgrade to 200 km/h will induce. These are, however, not included in the further cost estimate. Køge Tureby Haslev Næstved Figure 8.11: Placement of electrification masts according to the Programming Report from 2014 (Rail Net Denmark, 2013d), red lines indicate masts placed east from the line, while blue lines indicate a western position Electrification of the New Track The new second track is being electrified according to national standards, and is sought to comply with the international TSI Energy requirements. Specification of the requirements are considered out of scope in current project. The assumptions for the electrification are listed below; The new line is being electrified in accordance with national and TSI requirements The length of the new line being electrified is 37,5 km 10 % is added to the cost for the 200 km/h scenario Cost for electrification of crossovers is estimated as 5 % of the total cost for electrifying the line

151 CHAPTER Smaller adjustments of the existing electrification are expected, however, these are not included in the cost estimate Table 8.11: Length of electrified line and estimation of crossovers Length of line [km] New line being electrified fo 160 km/h 37,5 New line being electrified for 200 km/h 37,5 Electrification of crossovers (5 %) 1, Power Supply Power supply covers the electrical installations, mainly at stations, and consists of the entities listed below, (Rail Net Denmark, 2014e). No exact link can be drawn to the 29 point list. Platform lighting Emergency power supply system Heating systems for turnouts Additional systems; electrical information boards, cooling and ventilation systems, elevators, fire extinguishing systems, etc. In the renewal of the station areas, new platform bridges with associating elevators are installed. Furthermore removal of turnouts are considered. The following assumptions are taken; New turnouts are equipped with heating systems The emergency power supply is to be reviewed in a further stage Station categorised items, such as platform lighting and the additional systems mentioned above, are assumed to be renewed in accordance with the national regulations and the TSI PRM. Current project does not handle any entities within the power supply category in further detail. Based on the assumptions mentioned above, the cost is estimated as a kilometre price, based on experiences from the Vamdrup-Vojens Project concerning the implementation of second track in the Southern part of Jutland (Grontmij, 2012). No distinction in cost is made between the upgrade of existing track and implementation of the second track to 160 km/h and 200 km/h respectively.

152 136 PROJECT DESCRIPTION Table 8.12: Power Supply estimate for the total Lille Syd line Cost estimate for power supply for Vm-Oj (20,4 km line) Cost estimate for power supply for the Lille Syd line (43,1 km) Price for entire line [DKK] Price per km [DKK/km] 7,116, ,833 15,034, , Interlocking and Remote Control In general, the term Interlocking and Remote Control covers the items listed below, and refers to point 27 in the 29 point list. Remote control Interlocking systems, block sections, relay groups Signals Train detection and controls; track insulators, axle counters, balises and ATC Level crossings As mention in chapter 2, the new Signalling System is expected to be fully deployed on the Lille Syd line in The new system will replace the majority of entities within the existing system. The system, known as; ERTMS level 2, is a radio-based system. The most apparent difference from today s system is the exclusion of the physical signals, used for granting movement authority. In the future, movement authority is provided to trains via a GSM-R radio connection from a Radio Block Center (RBC). The movement authority is grounded from the RBC, which receives information from the interlocking systems. The interlocking systems consist of train detection equipment and turnout settings. To grant movement authority the RBC receives information from the trains regarding position and direction. The function of the Remote Control Centres will not change, the traffic will still be controlled from here, however, the number of Remote Control Centres will decrease from 14 to 2 (Jensen, R., Rasmussen, 2014). The new Signalling System plays a major role in both the speed upgrade of the existing line and in relation to the installation of the second track. Regarding level crossings, currently five level crossings are maintained in the speed upgrade and electrification project from 2014 (Grontmij, 2014). In June 2014, the Danish Transport Authority approved the speed through automatic protected level crossings to 160 km/h. In relation to both the Signalling System and the level crossings, the following assumptions are made. In the scenario of 160 km/h and double track the five level crossings are maintained In the scenario of 200 km/h and double track, two level crossings are maintained at Køge, two are reconstructed to out-of-level crossings and the last one is closed The new system shall be decoded to handle the speed upgrade and implementation of a second track in both speed scenarios

153 Overordnet system forståelse Kørsel på førerrumssignal (ERTMS løsning) CHAPTER GSM-R radio connection Remote control Local interlocking system Radio Block Controller Figure 8.12: Illustration of the operation of the new Signalling System ERTMS, (Møller, Jens Holst and Jensen, 2015) No distinction in cost is made between the two scenarios 7 The cost is rather difficult to estimate and is associated with much uncertainty due to the fact, that the system is still not deployed and no adaptation has earlier been carried out in Denmark. The cost in current project is based on the cost estimated in the Vamdrup- Vojens project (Grontmij, 2012) and the Hobro-Aalborg project (Grontmij, 2015a) according to Table The cost in current project is estimated to DKK 50 million due to the fact, that current project is assessed more comparable with the Hobro-Aalborg project. Table 8.13: Cost estimate of interlocking and remote control for the Lille Syd line Price [DKK] Cost estimate for Vm-Oj 70,000,000 Cost estimate for Hobro-Aalborg 3,000,000 Cost estimate for the Lille Syd line 50,000, IT, Tele and Transmission Systems The term IT, Tele and Transmission Systems covers the entities mentioned below, and no direct link can be drawn to the 29 point list. Telephone systems Radio systems Transmission and data network systems Traffic information systems; screens, clocks and loudspeakers

154 138 PROJECT DESCRIPTION In general, old telephone systems are phased out and replaced by more modern systems on all railway lines. Transmission systems are in charge of the communication between the remote control centre, the interlocking systems, and the signals (Rail Net Denmark, 2014e). In current project, it is assumed that the Signalling Programme, will cover an upgrade of the existing telephone, radio and transmission systems. Furthermore, it is assumed that later project phases will cover the necessity of upgrading the traffic information systems on all stations, except at Køge and Næstved station. No estimated cost is therefore earmarked for this Constructions Constructions cover the installation of platforms and refers to point 17 in the 29 point list. In current speed upgrade project, the track geometry is changed to improve the performance of the line. In relation hereto, the TSI and the Danish requirements shall therefore be complied with. All stations along the Lille Syd is equipped with two tracks, except at Næstved Nord station. The speed upgrade of the existing line to either 160 or 200 km/h therefore requires an investigation of the existing platforms. In current screening phase, the design of the platforms in accordance to width of safety zones and open spaces is considered, as well as the overall location and type. The design is considered for both speed upgrade scenarios Platform Design for Speed Upgrades The current platforms at all stations, except Køge and Næstved station, do not comply with either the requirements for platform lengths stated in TSI INF, the requirements for platform heights stated in TSI INF and BN1-49-1, and the requirements for safety zones stated in BN1-9-2 and TSI PRM, covered in section 5.6. The platforms have various lengths and heights, and especially the widths of the safety zones and open spaces are critical for most centre platforms. Construction of new platforms is therefore required at all intermediate stations, except at Køge and Næstved station. As mentioned in section 5.6, the TSI INF requires a minimum platform length of 200 m (Den Europæiske Unions Tidende, 2014a). Minimum widths are determined according to the two platform zones; the safety zone and the open space. The widths of the zones differ according to the maximum speed, and type of platform (Rail Net Denmark, 2012a), see Table 5.17 and Table 5.18 in chapter 5. Furthermore, the TSI PRM states that the open space area should reflect the maximum instant number of passengers exchanging from a line in a rush hour, and each passenger is granted 1 m 2 (Den Europæiske Unions Tidende, 2014b). Referring to the number of daily passenger arrivals and departures at each station mentioned in Table 2.1 in chapter 2, Haslev Station has the largest passenger exchange, leaving Ølby out of account due to the influence of the suburban train. It is assumed that 25 % of all travels are performed during the morning rush hours from 07:00 to 09:00 AM (Lohmann and Landex, 2009). Furthermore, the frequency of stopping trains in daily hours is two trains per hour and direction, resulting in passenger exchange from 8 trains during the two morning rush hours. This contributes to a maximum passenger exchange

155 CHAPTER (a) Existing side platforms comply with requirements for platform widths, Tureby station (b) Existing center platforms do not comply with requirements for platform width, Herfølge station (c) Platforms do not comply with requirements for platform lengths, Holme-Olstrup station Figure 8.13: In general platforms on the Lille Syd line does not comply with either the national or the TSI requirements, photos taken of 85 passengers per train according to Equation 8.1. The open space on the platform at Haslev station can therefore not be smaller than 85 m 2. Passengers Train = passengers 25% in rush hours 8 trains in rush hours = 85 passengers per train (8.1) From section 5.6 Table 5.18 it can be concluded, that platforms constructed according to the minimum required widths and length, will respond to the maximum passenger exchange on all stations. In general, two side platforms are constructed on every station, except at Ølby station, where the existing side platform is replaced with a centre platform. At Haslev and Næstved Nord station, the existing side platforms comply with the width and height requirements, however, the length is too short. These two platforms are prolonged 55 m and 42 m respectively, see Figure All new platforms are constructed according to the minimum requirements listed in section 5.6. It is assumed that no constructions or objects with a size larger than 1 m are present at the platform. Existing station buildings are preserved, however, these are assumed located in a distance further than 3 m from the nearest track. The overall location of the platforms at all stations is illustrated in TCE In total 1 new centre platform and 8 new side platforms are constructed, and 2 existing platforms are prolonged in accordance with Table Areas Current section describes the demand for temporary and permanent expropriations of land, in relation to the speed upgrade of the existing line and the construction of the second track. Considering both permanent and temporary expropriations, the cost of expropriated area depends on the type of area zone and the geographic location. In both scenarios, the implementation of the second track to either 160 or 200 km/h will result in

156 140 PROJECT DESCRIPTION Prolonged platform Removed platform Displaced track 0.85 m 2.00 m New platform 200 m Figure 8.14: Illustration of the location and prolongation of platforms in accordance with the track relocation, existing platform removed (red), existing track displaced (blue) and construction of new platform (green hatched) Table 8.14: Size and location of newly implemented platforms Total size of platfoms Total size of platforms 160 km/h 200 km/h [m2] Location of platforms Side platforms [m2] 4,560 5,360 2 x Hf, Th and Ol 1 x Hz and Næn Centre platforms [m2] x Ølb Extension of platforms [m2] Hz extended 55 m Næ extended 42 m Removal of platforms [m2] 3,370 3,370 1 x Ølb, Hz 2 x Hf, Th and Ol New edge length [m] 2,097 2,097 1 x Ølb, Hz and Næ 2 x Hf, Th and Ølb permanent expropriation along the existing line according to the track layout illustrated in TCE In relation to temporary expropriations, these cover more or less the working areas needed to carry out the construction work. In general, no consideration of the exact location of temporary working areas and access roads are taken. The following assumptions are valid for both scenarios; Permanent expropriation in urban zones is limited, due to the fact that two tracks are already present near stations 5 % of the total permanent expropriated rural area is added to cover expropriated buildings in urban zones 1 % of the total permanent expropriated rural area is added to cover expropriated buildings in rural zones No permanent expropriation is needed to reconstruct the station areas

157 CHAPTER Upgrade of Existing and Construction of Second Track to 160 km/h Temporary Expropriation Considering the scenario of upgrading the existing track to 160 km/h, temporary working areas are only relevant at stations. These are established at all intermediate stations between Køge and Næstved, and at Ølby station. No work is required at Køge and Næstved station. The extent of the working areas is estimated based on experiences from the project of expanding the North Western railway line (Rail Net Denmark, 2012c). In total 6 working areas with the extent of 20m x 50m are established in urban zones. Regarding the construction of the second track to 160 km/h, a temporary working road along the 37,5 kilometres of new track is established. The width of the temporary working road is estimated to 10 m. An additional 10 % of temporary expropriation area is added to take the following into consideration; earth depots, working areas along the line, and construction of access roads, in cases where the existing roads are not sufficient. Table 8.15: Temporary expropriated areas for the 160 km/h scenario Temporary expropriations in urban zones Temporary expropriations in rural zones Working area near stations [m2] 6,000 Working road of 10 % additional 10 m [m2] working area [m2] 375,000 37,500 Permanent Expropriation The construction of the second track requires permanent expropriation along the 37,5 kilometres new track. It is assumed that all 37,5 kilometres are situated in rural zones. The expropriated area depends on whether the track is constructed in fill or cut. The width of the expropriated areas are shown in drawing TSPR 3 NXS 001, for a line with maximum speed of 160 km/h. The stations at Ølby and Næstved Nord currently only consist of a single track with a side platform. At Ølby station the location of the existing platform is maintained and exchanged to a centre platform, and the second track is constructed on the outer side of the platform, cf. drawing TCE At Næstved Nord station, a second side platform is constructed on the outer side of the new track, cf. drawing TCE Additional land is permanently expropriated in accordance with the extent of the platform. The expropriation is estimated to 300m x 5m. In relation to this, the following assumptions are made; The new track (37,5 km) is implemented in rural zones 20 % of the line is constructed in cut, resulting in an expropriated width of 5,1 metres 80 % of the line is constructed in fill, resulting in an expropriated width of 7,1 metres

158 142 PROJECT DESCRIPTION No permanent expropriation is needed for the platform at Ølby station Additional land is expropriated for the platform at Næstved Nord Table 8.16: Permanent expropriations according to area zones for scenario of 160 km/h Permanent expropriations in urban zones Permanent expropriations in rural zones Second track for 160 km/h [m2] Expropriation of buildings (5%) [m2] Expropriation of buildings (1%) [m2] 12, ,750 2, Upgrade and Construction of Second Track to 200 km/h Temporary Expropriation For the speed upgrade of the existing track to 200 km/h and the construction of the second track, temporary working roads are assumed constructed on both sides of the line, except in station areas. In the urban areas adjacent to the stations, it is assumed that the necessary track work can be carried out by situating the machinery in either of the two tracks respectively. Hence no temporary expropriations are assumed in these areas. The following assumptions are taken; A temporary working road is constructed on both sides of the tracks in rural zones 5 % is added to account for earth depots and working areas along the line The same size of working areas at stations are installed as for the 160 km/h scenario Table 8.17: Temporary expropriations according to area zones for scenario of 200 km/h Temporary expropriations in urban zones Temporary expropriations in rural zones Working area near stations [m2] 6,000 Working road of 10 m 5 % additional on both sides of track [m2] working area [m2] 750,000 37,500 Permanent Expropriation Permanent expropriations for the construction of the second track to 200 km/h are more or less similar to the expropriated area for the 160 km/h scenario. The total estimated permanent and temporary expropriated areas are summarised in Table The following assumptions are made; The new track (37,5 km) is implemented in rural zones

159 CHAPTER % of the line is constructed in cut, resulting in an expropriated width of 5,5 metres 80 % of the line is constructed in fill, resulting in an expropriated width of 7,3 metres No permanent expropriation is needed for the platform at Ølby station Additional land is expropriated for the platform at Næstved Nord Table 8.18: Permanent expropriations according to area zones for scenario of 200 km/h Permanent expropriations in urban zones Permanent expropriations in rural zones Second track for 200 km/h [m2] Expropriation of buildings (5%) [m2] Expropriation of buildings (1%) [m2] 13, ,750 2,618 All in all, it can be concluded that more or less the same amount of permanent expropriated area is expected in both scenarios. However, almost twice as much temporary expropriated area is expected for the 200 km/h scenario, due to the extra temporary working road Electrical Restrictive Covenants An electrical restrictive covenant is enjoined on buildings and land located in certain distances from the electrified line. The restrictive covenant limits the owner s right of disposal of the land, due to safety reasons. Distinction is made between rural and urban zones, and the type of object located on the land. Table 8.19 summarises the limits of the location of objects in relation to the zone. Table 8.19: Location of objects considering expropriation due to electrification of the line, (Rail Net Denmark, 2006a). Location of objects in urban areas Location of objects in rural areas x > 3m Trees x > 3m Trees x > 6m Buildings, machinary x > 10m Buildings, machinary 6m < x < 10m Flagpoles, wells 10m < x < 14m Flagpoles, wells 10m < x < 15m Electric fence higher than 2 m, Electric fence higher than 2 m, 14m < x < 19m special types of trees special types of trees The Programming Report from 2014, has determined the total area imposed by an electrical restrictive covenant (Grontmij, 2014). The area is measured from the track center line. By electrifying the second track, it is assumed that many of the influenced objects are already covered by the existing covenant. In relation to the Programming Report, the following assumptions are made; 10 % of the neighbour area found within the limit of 19 m in the Programming Report is added in current case

160 144 PROJECT DESCRIPTION 20 % of the buildings found within the 10 m limit is added This result in 8 hectare of land for the 19 m limit and 8 buildings. Later project phases will determine the allowed usage of these lots Forestry Forestry covers the environmental aspect of the construction and upgrade of a railway line. Maintenance of the greener areas adjacent to the railway is considered. This includes wild fences, plating of reserved forest, installation of compensatory greener areas, and compensatory noise screens. In relation to the construction of the second track and the associated electrification, a large part of the reserved forest is cut down. To compensate for this, double the area of forest is planted. The cost is estimated based on experiences from the Vamdrup-Vojens project (Grontmij, 2012). A cost per kilometre line is determined to estimate the cost for the Lille Syd line. No distinction is made between the two speed scenarios. Table 8.20: Forestry estimate for the Lille Syd line Cost estimate for forestry Vm-Oj (20,4 km line) Cost estimate for forestry for the Lille Syd line (43,1) Price for entire line [DKK] Price per km [DKK/km] 7,514, ,368 16,988, , Additional Considerations Additional Considerations cover supplementing technical surveys to clarify the condition and characterisation of the influenced area. This includes; geotechnical surveys, environmental and archaeological investigations, as well as measurements and borings. The cost is, similar to above, estimated based on experiences from the Vamdrup-Vojens project. Table 8.21: Estimate of additional considerations for the Lille Syd line Cost estimate for additional considerations Vm-Oj (20,4 km line) Cost estimate for additional considerations for the Lille Syd line (43,1) Price for entire line [DKK] Price per km [DKK/km] 50,945,119 2,497, ,168,434 2,497, Cross-disciplinary Costs In relation to the 11 earlier described disciplines, the cross-disciplinary item covers the different processes a project is undergoing. The cross-disciplinary costs are more or less scattered across the project phase. The estimated cost is given as a percentage of the total cost. The following items are covered in the discipline;

161 CHAPTER EIA investigations Projecting of alignment Construction management and technical inspection CSM and NoBo processes Safety approval from the Danish Transport Authority Project follow-up General owner costs Based on experiences from the screening project of Hillerød, 20 % of the estimated construction cost is added, to cover the cross-disciplinary costs (Grontmij, 2015c) Summary of Project Description A summary of the project description for the two scenarios of upgrading the existing track, and constructing a second track for 160 km/h and 200 km/h respectively is provided below. The project description forms the background for the estimated cost by determining the different types of work and initiatives to be carried out. The section is presented according to the disciplines in New budgeting. Track The track work includes; 37.5 km of new track, 16 new turnouts and removal of 8 turnouts in both solutions. Upgrading activities include; track adjustment, track displacement and improvement of the track construction. In the 200 km/h solution, the ballast profile of the existing track is changed to comply with requirements for 200 km/h. Earth Work Earth Work includes; construction of substructure, drainage and embankment. In the 200 km/h, the thickness of subballast is increased to 300 mm which causes changes to the top of subsoil. Large uncertainty is associated with the project description due to unknown geotechnical conditions. Bridges and Constructions All existing 18 road bridges and pathways are demolished and reconstructed to account for the double track. The 28 railway bridges are maintained, and 28 new bridges are constructed similar and adjacent to the existing ones. Two out-of-level crossings are constructed in the 200 km/h scenario, to compensate for the closing of two level crossings. Simple platform bridges with associated elevators are constructed at all stations except Køge and Næstved. Electrification System The installation of the second track requires electrification of 37,5 kilometre track. 10% is added to the cost in the 200 km/h scenario. 5% is added to account for electrification of crossovers. Power Supply New turnouts are equipped with heating systems. Emergency power supply and station categorised items are not covered. The cost is estimated based on a kilometre cost for the Vamdrup-Vojens project.

162 146 PROJECT DESCRIPTION Interlocking and Remote Control The new Signalling System is installed and shall be decoded to the constructed second track and speed upgrade. In the 160 km/h scenario, the 5 level crossings are maintained. In the 200 km/h scenario, 3 level crossings are closed, and 2 are maintain near Køge station. IT, Tele and Transmission System This item is not considered, as later project phases will determined the need for renewal of information systems at the stations. Constructions New platforms are constructed at all stations, except at Køge and Næstved. 8 side platforms and 1 centre platform is constructed, and the associating 9 platforms are removed. Two existing platforms are retained and prolonged. Areas Permanent expropriation is required for the construction of the second track. No permanent expropriation is needed for reconstructing stations, except at Næstved Nord station. Temporary expropriated area is required a long the new track, and also along the existing track in the speed upgrade to 200 km/h. Temporary expropriation considers temporary access roads, working areas and depots. 5% and 1% is added to the permanent expropriated area in rural and urban zones, respectively, to account for expropriation of buildings. Forestry Forestry covers the cost for installing and maintaining green areas along the line. Reserved forest is cut down along the new track, and is compensated for by planting twice the area. The cost is estimated on a kilometre cost from the Vamdrup- Vojens project. Additional Considerations Technical surveys and environmental investigations are considered to clarify the condition and characteristics of the line. The cost is estimated on a kilometre cost from the Vamdrup-Vojens project. Cross-disciplinary Costs 20% is added to the estimated cost to take cross-disciplinary items into consideration. These include EIA investigations, projecting of alignment, approval processes and general owner costs.

163 Chapter 9 Construction Cost and Project Evaluation 9.1 Construction Cost Present project is a screening of extending the Lille Syd line from single to double track, while upgrading the line speed to 160 km/h and 200 km/h, respectively. The construction cost is evaluated in accordance to level 1 in New Budgeting (Brandt, M., Morberg, 2010), corresponding to the subsequent definition phase. This is caused by lack of guidelines for conducting the construction cost estimate in an early screening phase. As described in previous chapter, New Budgeting is a concept of standardising the procedure for estimating construction costs, to enable comparisons between projects. The methods is based on theoretical quantities multiplied with experience-based unit costs. The cost estimate is structured according to the template for entry disciplines proposed by Rail Net Denmark and the Danish Transport Authority (Rail Net Danmark and The Danish Transport Authority, 2010) Unit Cost The unit costs are derived from previous projects and converted to the price level of the calculation year according to a price index. It can be necessary to scale the unit cost according to the project size, because large quantities often entail in lower unit costs. The geographical placement can likewise influence the size of the unit cost, due to an increased cost for transporting construction loads in urban areas. Hence, unit costs in the capital area are higher than in peripheral regions. Current project uses 2014 prices, derived from the cost estimate of the Hobro-Aalborg speed upgrade project, (Grontmij, 2015a). This project is found applicable in relation to the type of land zones and the speed upgrading activities. Furthermore, the Hobro- Aalborg project material contains the most recent update of Grontmij s unit cost database, due to its recent termination. Unit costs which are not covered in the Hobro-Aalborg project, are found in additional projects presented in chapter 8 and, in few cases, own estimates. Disciplines which are not assessed through the description of required quantities, such 147

164 148 CONSTRUCTION COST AND PROJECT EVALUATION as; power supply, forestry and additional considerations, are estimated according to the Vamdrup-Vojens project of implementing a second track. Based on the total cost for each discipline, a unit cost per kilometre track is determined and multiplied with the total extent of the Lille Syd line. The Vamdrup-Vojens project does too include upgrade of the existing line and construction of a new double track, and it is therefore assessed reasonable to make a direct comparison Correction Supplement A correction supplement of 50% is used for evaluation of the construction cost in phase 1, according to Figure 8.1. This supplement contains all project uncertainties in relation to estimation of quantities and unresolved project conditions exposing in large project risk. A correction supplement of 30% is used for evaluating project in phase 2. In addition hereto, a cost accounting supplement is added to each main discipline, to account for the uncertainties and high project risks. This enables a higher degree of targeting the uncertainties. The scope of current project is mainly related to track issues. Focus has therefore been on the track discipline and the construction of platforms, (placed under construction). Still, the remaining disciplines have been addressed to capture the overall cost estimate. It can therefore be argued, that these disciplines contain a larger project uncertainty, hence cost accounting could be added. It is however assessed, that these are included in the cost supplement of 50 %, why additional cost accounting will result in a too high estimate Cost Estimates Cost estimates are conducted for the 160 and 200 km/h solution and enclosed in Appendix M. No distinction is made in relation to the cost for the line upgrade and the cost for implementation of the new track, given that the project proposals include both. Evaluating the cost exclusively for the upgrade or the new track, respectively, requires separate submitted project proposals. The estimates are structured according to the 12 main disciplines in chapter 8, and the cost is derived from the estimated quantities. The distribution of the estimated costs is shown in Figure 9.1. In general it is seen, that the 200 km/h solution has the largest construction cost. This was expected, due to the more comprehensive construction work in the solution. Bridges and Constructions constitute to the largest project cost. This cost covers removal of existing and reconstruction of new wider road bridges. Furthermore, new railway bridges are built next to the existing ones, to support the new track. The cost is estimated based on the dimension of the existing bridges multiplied with a unit cost per square metre. In a later phase, it will be necessary to make a thorough investigation of each bridge, to estimate a more accurate cost. To reduce the construction cost, a further investigation of the road traffic will clarify whether certain road bridges can be closed. The additional cost in the 200 km/h solution is caused by the reconstruction of the two level crossings to two out-of-level crossings. This cost is estimated to DKK 50 mio. per level crossing, in accordance with information provided by Rail Net Denmark during a special course

165 CHAPTER Construction Cost Cost in DKK Millions TRACK 02 - EARTH WORKS 03 - BRIDGES AND CONSTRUCTIONS 04 - ELECTRIFICATION SYSTEM 05 - POWER SUPPLY 06 - INTERLOCKING AND REMOTE CONTROL 07 - IT, TELE AND TRANSMISSION SYSTEMS 08 - CONSTRUCTIONS 09 - AREAS 10 - FORESTRY 11 - ADDITIONAL CONSIDERATIONS 12 - CROSS-DISCIPLINARY COSTS Estimate 160 km/h solution Estimate 200 km/h solution Figure 9.1: Overview of the construction cost estimate for the 160 km/h and 200 km/h solution according to the 12 main disciplines in June, 2015 (Jensen, 2014a). The track sections upgraded to 200 km/h include two level crossings, hence DKK 100 mio. is reserved for reconstructing these. Sparing this cost will require closure of the level crossings, which most likely will be met with great opposition from the local population, and will therefore be difficult to carry through. The cost for Track considers the construction of the new track and the upgrade of the existing. The increased cost for the 200 km/h alternative is caused by the higher expenses for constructing the new track approved for 200 km/h, and the additional required changes to the existing track. This is also valid for the cost reserved for Earth Work. The difference in cost for Electrification is caused by the additional 10 % added to the construction cost in the 200 km/h solution. At the same time, the additional cost for Constructions is caused by the requirements for wider platforms for a line speed of 200 km/h. In relation to Areas, additional cost for access roads along the existing track and a slightly larger area of permanent expropriation is required for the 200 km/h solution. Cross-Disciplinary Costs are estimated as 20% of the total cost estimate, thus, a higher cost is reserved for the 200 km/h solution. The remaining disciplines are assessed to have same construction costs in the two project proposals. By summarising the cost estimate for the 12 disciplines and adding the correction supplement of 50%, the construction cost is estimated to DKK 1.95 billion for the 160 km/h solution and DKK 2.64 billion for the 200 km/h solution, see Table 9.1. An additional investment of DKK 690 million is therefore required, to increase the line speed from 160 to 200 km/h. Table 9.1: Total construction cost estimate, see Appendix M Solution Estimated Cost Cost including 50% supplement 160 km/h DKK 1.30 billion DKK 1.76 billion 200 km/h DKK 1.95 billion DKK 2.64 billion

166 150 CONSTRUCTION COST AND PROJECT EVALUATION Risk Plan Normally a risk plan is conducted to handle the uncertainties associated with the cost estimate. In recent years, this method has received much attention in the construction industry. Conducting large infrastructure projects therefore include risk workshops, for developing and advancing risk plans throughout the entire project realisation. This is organised in accordance to the guidelines in New Budgeting. In present project, an actual risk plan has not been conducted. Instead are all the uncertainties in relation to the cost estimate treated throughout the project description. From this, a list of hazards are identified for all disciplines, see Appendix N. It is assumed, that the hazards related to Earth Work, will constitute to the largest project risks. This is caused by large uncertainties, both in relation to the rough estimated quantities, and due to the lack of information of the geotechnical conditions. The project description is based on the general assumption, that the existing line complies with all valid requirements, and no cost is set aside for handling of soft or contaminated soil. The cost for the earth work is therefore only 8% of the total construction cost. Comparing with the Vamdrup-Vojens project, the earth work constituted 30% of the total cost (Grontmij, 2012). Based on this, it is assumed that the hazards associated with unknown geothechnical conditions have a high likelihood of occurring, and will have large impacts on the cost estimate. These hazards therefore result in a large project risk. Similar to this, the remaining hazards should be treated in a risk analysis, where the impact and likelihood is assessed. The risks are then imported to a risk matrix, to identify the largest project risks. Initiatives to manage the risks are then incorporated in the project risk plan. The risk associated with the unknown geothechnical conditions, can for instance be handled by investing in geotechnical surveys in an early project phase. Based on the risk analysis, it can be decided to add an additional supplement to the cost estimate. In current project, this is considered recommendable for the risk identified above, however, this has not been addressed. 9.2 Project Evaluation The cost estimate is now used to evaluate the project. This is done by weighting the cost and benefits presented throughout the report. The purpose with present project has been to investigate the possibility of including the Lille Syd line in the TEN corridor connecting Scandinavia and Central Europe. This has been proposed to investigate possible gains in flexibility, shorter travel times, and relieve in the capacity consumption, by introducing an alternative, shorter route. Initially, the capacity consumption was examined on the main route Fehmarn-Ringsted-Copenhagen, by using the network capacity programme developed by Jensen. The result of the analysis showed, that only 1.3 percent of the generated train sequences achieved a capacity consumption of less than 100%, meaning very few train sequences can actually be generated, according to chapter 4. The capacity consumption was then investigated, with the introduction of relocating traf-

167 CHAPTER fic to the Lille Syd line. As a prerequisite for the analysis, the single track line was extended to double track. Three different speed profiles were investigated including; the existing speed profile, an upgrade to 160 km/h and an upgrade to 200 km/h. The capacity analysis showed improvements in the network capacity in all three alternatives compared to the base scenario, where all traffic was run on the main line passing Ringsted. Though, the improvement decreased in connection to the speed upgrades, which at the same time caused an associated increase in heterogeneity. Hence, the analysis showed the largest improvements in the network capacity, when the line speed on the Lille Syd line was not upgraded. Based on the three alternatives, a screening of the potential travel time savings for relocating the lines was conducted in chapter 4. In all three speed scenarios, the result was positive for the freight trains. The shorter route along the Lille Syd line, resulted in travel time savings of up to 5.5 minutes per train. Regardless of any of the two alternative speed upgrades, the freight trains experienced no further travel time savings, due to their relatively low maximum speed. For passenger train, the result was somewhat different. The Lille Syd line would have to be upgraded to at least 160 km/h, to avoid compromising the travel time. However, an upgrade to 160 km/h would also only just result in travel time savings for line 21 and 35/36, and the line should therefore be further upgraded, to result in travel time savings for line 20, see Figure 9.4. To ensure travel time savings for passenger trains, two alignment solutions for 160 and 200 km/h were proposed, to consider the upgrade of the Lille Syd line. These were conducted in accordance with Rail Net Denmark s norm foundation presented in chapter 5. The obtained speed profiles for the two alternative solutions both resulted in travel time savings for all three line variants, when compared to the existing speed profile of the Lille syd line. The largest time savings were found in the 200 km/h solution, however, both alternatives came out positive, see Table 7.3. To gain an overview of the utilisation of the two upgraded speed profiles in relation to the route from Copenhagen to Nykøbing Falster, Figure 9.2 and Figure 9.3 illustrates the speed curves for the different line variants. As expected from the travel time savings, line 20 and 21 fully utilises both speed profile alternatives. Line 35/36 solely uses the 160 km/h alternative, and an attempt to reach the design speed of 200 km/h will result in continuous acceleration and deceleration, which will lead to high operational costs. An evaluation of the obtained upgraded speed profiles for the two Lille Syd alternatives, can be conducted by comparing the travel times for all the line variants according to the two routes; passing Ringsted or the Lille Syd route. The travel time savings differ from the ones found in the initial analysis in chapter 4, as the expected fully upgraded speed profile for 160 and 200 km/h could not be obtained. Figure 9.4 illustrates the difference in travel time savings. The potential travel time savings estimated in chapter 4 are shown in blue, and the realised travel time savings for the two project alternatives are shown in green. The travel times in the top of the figure illustrate the savings already found in the initial analysis. As expected, the potential travel time savings are larger than the realised savings. This is true for all passenger lines, however, freight trains experience the same travel time

168 152 CONSTRUCTION COST AND PROJECT EVALUATION Speed Profile -160 km/h Kh Nel 300 KjN Kj Hf Th Hz Ol Næ Lu Vo Oh Nv Ek Nf Ølb Næn 250 Speed [km/h] Stationing [km] GX G6 LN35/36 LN21 LN20 Speed Profile Stations Figure 9.2: Proposed speed profile for the 160 km/h solution Speed Profile -200 km/h Kh Nel 300 KjN Kj Hf Th Hz Ol Næ Lu Vo Oh Nv Ek Nf Ølb Næn 250 Speed [km/h] Stationing [km] GX G6 LN35/36 LN21 LN20 Speed Profile Stations Figure 9.3: Proposed speed profile for the 200 km/h solution savings regardless of the line being upgraded. Considering the 160 km/h solution, line 20 obtains an increased travel time of 1.9 minutes when being relocated via the Lille Syd line. Line 21 gains 0.7 minutes, which is a little lower than expected, and line 35/36 does not gain any significant savings. In the 200 km/h solution, all passenger lines experience travel time savings, though the existing line 35/36 only gain 0.7 minutes. Considering the braking curves shown in Figure 9.3, it will most likely be more beneficial for this line variant, to run at a lower speed. Line 20 obtains too, only a small time saving of 0.4 minutes, while line 21 obtains 2.9 minutes. Based on the results of the travel time savings, it is assessed that the only significant gain is obtained for line 21 in the 200 km/h solution. Because line 20 only saves 0.4 minutes,

169 CHAPTER Travel Time Savings When Relocating Lines to Lille Syd Existing -4.1 G6 GX LN35/ LN21 LN km/h G6 GX LN35/ LN LN Realised Potential 200 km/h G6 GX LN35/36 LN21 LN Travel time savings in minutes Figure 9.4: Travel time savings when relocating trains to the Lille Syd line from the main line passing Ringsted it will not be beneficial to relocated the line to the Lille Syd line, seen from a time saving perspective. Comparing the travel time savings with the construction cost estimate, it can be concluded that a speed upgrade to 200 km/h will not be cost effective. As stated above, the 200 km/h solution is DKK 690 million more expensive than the 160 km/h solution, and only one line variant will gain significant travel time savings. Looking at the time savings for the 160 km/h solution, more or less the same conclusion can be drawn. Little time savings are gained, and only for line 21, hence an upgrade to 160 km/h will not be cost effective either. Based on these findings, it can be concluded, that the existing Lille Syd line should not be further speed upgraded, as no significant travel time savings can be obtained despite the shorter route. A socio-economic analysis is expected to support this conclusion, as time savings are significantly small compared to the construction estimate. To assess the final network capacity according to the two obtained speed profiles in chapter 7, the network should be re-examined, though this has not been carried out. However, assumptions of the expected network capacity will indicate, that more or less the same result will be found as in the initial capacity analysis in chapter 4. It can be assumed, that the reduction in travel time savings for the realised profiles compared to the potential profiles according to Figure 9.4, will not worsen the network capacity, as the heterogeneity level will be decreased. The exact impact is however difficult to predict. The speed upgrade of the Lille Syd line did not result in any significant travel time savings, and the network capacity consumption did only increase in connection to the speed upgrade. However, implementing the second track on Lille Syd and avoiding a speed upgrade, did result in significant time savings for the freight trains and at the same time, the network capacity was improved. An alternative solution would therefore propose

170 154 CONSTRUCTION COST AND PROJECT EVALUATION an extension of the Lille Syd line to double track without upgrading the line speed. To avoid compromising any travel times, this proposal does not consider any relocation of passenger trains. This will result in time savings in the range minutes for freight trains, which is considered to be rather high. A new project proposal will change the prerequisites in the initial capacity analysis, and a new analysis should thereby clarify if the network capacity still is improved. From the initial study it was seen, that the network capacity was worsen, when the speed was increased due to the level of heterogeneity. It can therefore be expected, that the network capacity will be improved as long as the slow freight trains are separated from the faster passenger lines. Redirecting them via Lille Syd is expected to relieve the capacity consumption of especially the Copenhagen-Ringsted high speed line. The new proposed solution will furthermore entail in a reassessment of the project description. The proposed upgrading initiatives can thereby be omitted. These include reconstruction of the existing platforms, if assuming that the new second track can be coupled to the existing track at crossings station. However, the large cost estimates for bridges and the earth work can not be omitted. Still, this will result in a new and cheaper construction cost estimate, to be weighted against the traffic related benefits. 9.3 Recommendation Based on the project evaluation and the construction estimate, it is not recommended to continue the investigation of upgrading the Lille Syd line to either 160 km/h or 200 km/h, while implementing a second track. Instead, it is recommended to make a screening of extending the line to double track, possible only on some line sections, without upgrading the line speed. Implementation of a second track to complement the freight trains will result in time savings, improved network capacity and allow for flexibility in the network, by introduction of an alternative route. This will furthermore complement the vision of increasing the rail freight transport, especially the transit transport, in Denmark. No cost estimate is conducted for the implementation of the second track, without considering the speed upgrade, however, the cost is estimated lower than DKK 1.76 billion. A socio-economic analysis of the travel savings for the freight trains related to the cost estimate, will conclude whether or not the solution is feasible. In addition hereto, operational savings in relation to the shorter route, will contribute to the benefits. Though, it shall be mentioned, that public transport initiatives rarely contribute to a positive costbenefit result.

171 Chapter 10 Overall Process Flow Based on the different processes completed in present project, an overall process flow chart is proposed. This reflects more or less the processes a project in general goes through in a screening phase. However, new elements have been applied in relation to the network analysis, conducted in chapter 4 and the comparison with the TSI requirements in the chapter 7. The flow chart differentiate between processes for New Built Line and for Upgrades, and considers only initiatives for improving and changing the performance of the infrastructure. Renewal projects therefore require a different process flow. The chapter is divided in two sections treating the process flow considered in present project and an optimised process flow. In the second section of the chapter, the proposed process flow is reconsidered. Based on the experiences from the present project, an optimised flow with focus on the interaction between the traffic planning and the infrastructure planning process, is proposed. The optimised flow involves an optimisation model to improve the interaction between the planning sections. The idea and purpose of the model is presented. However, no effort is put into developing the model. Hence, the focus is on the optimisation of the process and not on the technical development Process Flow in Present Project The proposed process flow for present project is sketched in Appendix O. The flow is divided in three, considering; the preliminary investigations, the alternative solutions, and the evaluation of the solutions. The flow for both new built lines and upgrade projects is considered, and the entry of the optimisation process is illustrated. An assumption for the flow is, that the traffic assumptions and passenger prognoses cannot be sacrificed, hence alternative solutions must cover these. An alternative flow could include a process of reconsideration of the requirements for the infrastructure, however, the focus has been to meet the overall visions. The flow consists of 15 processes, seeappendix O, which are all covered in the section below. The processes are described in general terms, yet lines are drawn to the present project. The first four processes concern the preliminary investigations of the network. 155

172 156 OVERALL PROCESS FLOW 1. Investigate Visions The overall future Danish and European visions for the relevant railway network are identified. In current analysis, the visions reflect the goal of a Trans-European Transport Network presented by the European Commission, and the Danish visions originate from the Green Transport Policy and Togfonden DK, all described in chapter Clarify Requirements Based on the overall visions, the requirements for the network are clarified to define the demand for the infrastructure. In present project, the requirements are based on the traffic plan presented by the Danish Transport Authority and reflects the expected development in passenger and freight trains, to create a common planning foundation for the public transport, cf. chapter Network Analysis The infrastructure and expected number of trains are evaluated based on a capacity analysis of the network. The capacity analysis is conducted by use of Jensen s capacity assessment model, referred to in chapter 4. The infrastructure considers both ongoing, planned and proposed projects, and the expected number of trains are based on a future timetable presented by Rail Net Denmark for passenger and freight trains. Besides the capacity analysis, parameters such as travel time and network flexibility can also be addressed in current process. 4. Evaluation Based on the network analysis conducted in 3., an evaluation of whether the network can handle the proposed traffic at an acceptable level or not is made. Based on the preliminary investigations, alternative solutions are presented to improve the network or traffic handling. The following three processes (5, 6 and 7) are defined as being iterative. Different alternative solutions are investigated according to the requirements for the network. 5. Alternative Solutions and 6. Network Analysis Alternative solutions are presented to improve the traffic situation of the network. Input for the investigation is a new alternative to either the infrastructure, the operation plan or both. These are investigated in relation to the network capacity, and possibly also to the flexibility and travel time savings. In case no alternative for the infrastructure is demanded, the project will not initiate step 8. In this project, the alternative solutions consist of implementing a second track along the Lille Syd line and at the same time increasing the speed to 160 and 200 km/h respectively. The results of the capacity utilisation and time savings are described in chapter Evaluation An evaluation of whether or not the initiatives have lead to an acceptable improvement is made, and step 8. is initiated if the results are positive, else step 5. is considered. The remaining part of the processes consider the evaluation of the alternative solutions. The flow chart is divided in two according to the type of project. The optimised process flow is also initiated at this point, and will be described further in section 10.2.

173 CHAPTER Current Traffic Assumptions The object of this step is to cover the requirements for the infrastructure. Present traffic assumptions cover the expected traffic on the Lille Syd line, the integration of the local line and contingency operation, according to chapter Define type of project At this point, the project is defined to determine the parameters considered in the following steps. Two types of projects can be considered; New Built Line and Line Upgrades. 10a. Define Prerequisites for New Built Line Prerequisites such as delimitations within the project, interfaces with other projects and overall assumptions are clarified at this point. These determine what is covered in the project, and what is left out. In relation to present project, these are covered in chapter 8. Furthermore, the norm foundation for the project is clarified, to determine the regulations to be complied with, cf. chapter 5. 11a. Project Description according to New Built Line The project description outlines the type of work required to carry out the project. It also forms the background for the cost estimate, by considering the different disciplines involved in the project. chapter 8 covers present project description. 10b. Define Prerequisites for Upgrade of Existing Line Same procedure as in step 10a. above. 11b. Project Description according to Upgrade of Existing Line As opposed to the project description for New Built Line, this project description involves the 29 point list. This is solely relevant for upgrading projects, as it functions as a checklist, when the existing infrastructure is changed. As in point 11a., the different project work disciplines are described, to clarify the design and form a background for the cost estimate. 12. Construction Cost Estimate The construction cost is estimated based on the design and initiatives clarified in the project description. Unit costs are associated to the different quantities and types of work described, and experiences from earlier projects are applied to validate the obtained cost. 13. Risk Plan A risk plan is conducted to handle the different types of risks involved in the project. 14. Evaluate Solution In the last step of the analysis, the proposed solution is evaluated based on the construction cost and the gains of travel time savings and capacity utilisation, possibly flexibility. Present gains are reflected by the results obtained in chapter 9. In some cases, as for this project, it will not be possible to obtain the exact proposed alternative solution considered in the network analysis, step 6. In present project, it was not possible to obtain the goal speed of 160 and 200 km/h respectively on the entire line. The findings in the network analysis are therefore reconsidered and evaluated in relation to the estimated construction cost. Based on this, it will be determined whether the project should proceed to the next phase, cf. Figure 8.1.

174 158 OVERALL PROCESS FLOW The flow chart expresses the different processes present screening project has gone through. Compared with general screening projects, the major deviations are in the execution of the network analysis in step 3. and 6. Based on an evaluation of present project, the process flow has been reconsidered, and initiatives for improvement within the planning and design phase are presented in the following section Optimised Process Flow The process flow presented above can more or less be defined as a linear process, except for the iterative process in the network analysis (point 6.). In present project, after choosing the alternative solution, much effort was put into deriving the speed profile with the largest possible speed, without consideration of the construction cost. This was followed by an estimation of the construction cost, and an evaluation of whether the findings were acceptable to continue to the next phase. The process flow is now reconsidered. In a screening process it seems beneficial to consider many different types of solutions, both infrastructural and traffic-related. At the same time, a fast evaluation of these in relation to weighing the gains against the investments would benefit the process. An improved connection between the design phase of the infrastructure and the traffic planning phase is desirable. A proposed improvement to the process flow consists of a rather optimistic extension of the optimisation model developed in present project, cf. chapter 6. The proposed extension will lead to a better interaction between traffic planning and infrastructure design in upgrading projects. The model is expected to generate socio-economic considerations in relation to investments in the infrastructure and gains of time travel savings. The technical part of the model is not considered at current stage, as focus has been on improvements of the process flow. The proposed process flow is enclosed in Appendix O. 10c. Define Prerequisites These are similar to point 10b. in relation to defining the norm foundation, the project interfaces, the delimitations and further assumptions to be included in the model Define Goal The goal for the optimisation is defined according to the different objective function alternatives. 11c. Project Description according to Upgrade of Existing Line This point is similar to point 11b. in the previous processes. The project description estimates the infrastructural changes needed to comply with the requirements (point 2.), and describes the design and condition of the existing infrastructure Optimisation Model The optimisation model is set up according to the norm foundation and the objective function. Inputs of the existing infrastructure such as track geometry, ballast profiles, location of bridges, level crossings, platforms, etc are read as data files. Input of unit prices related to the change of the infrastructure elements are included, to estimate the cost for the different infrastructural changes explained in the project

175 CHAPTER description. Time values according to the specific line and number of passengers are included as the last input. The model is optimising the infrastructure and the corresponding travel times according to the objective function, and based on the unit prices and time values. Constraints can possibly be implemented in the model, to clarify what elements can be changed and what can not. An example, in relation to present project, could be the exclusion of changing the alignment in the areas near bridges or stations to ensure compliance with structure gauges. Alternatively, large costs could be added in the unit prices, if the alternative still should be considered Evaluate Solution According to Goal The solutions are evaluated based on the cost estimate and the travel time gains. Capacity gains and flexibility in the network also play a role. These parameters are difficult to implement in the model, however, the Network Analysis model can alternatively be attached to investigate the solution. It is assessed whether or not the solution fulfils the defined goal at an acceptable level. If not, the goal is reconsidered in accordance with the different objective functions. Alternatively, the requirements in point 2. could be reconsidered, in relation to relocation or changing in traffic Risk Analysis A risk plan is conducted for the accepted solution. The largest project risks are assessed, and an evaluation of these will determine, whether the project can continue to the next phase; the definition phase, or should be reconsidered in the objective function. The optimised process flow seeks to introduce a closer connection between the traffic planners and the infrastructure planners. By introducing a model that optimises the construction cost of infrastructure changes over the travel time savings, a strong background for decision support of infrastructural projects can be developed. The model is rather complex, and much effort and programming is required for the development. It is furthermore questionable, if the model can even be developed in the proposed state. However, the general idea of a iterative process flow within the optimisation of infrastructure initiatives and the travel time gains are assessed beneficial.

176 160 OVERALL PROCESS FLOW

177 Chapter 11 Discussion Based on the analysis and findings in current project, a discussion of the most central assumptions and methods used, is structured. In general, the assumptions and findings are evaluated in discussed throughout the different phases in the project. Present chapter therefore mainly focuses on the capacity analysis, as a total new method for determining the network capacity was used. The execution of the proposed solution is discussed in the final part of the chapter, to gain a perspective of current infrastructure initiatives carried out. The potential for including the Lille Syd line in the Copenhagen-Fehmarn corridor has been investigated from a traffic and construction-wise perspective. Initially, a capacity analysis was used to clarify the traffic related benefits, when relocating traffic from the main line passing Ringsted to the Lille Syd line. The solutions proposed in the capacity analysis were then advanced into a project description, with the purpose of estimating the construction costs. This was done to enable evaluation of the project costs and benefits. The capacity analysis is conducted by using Lars Wittrup Jensen s proposed model for assessing network capacity. The model is developed to assess the the strategic planning phase, and determines network capacity by generating all permutations of train sequences in a network, and estimating the cumulative capacity consumption based on compression of train routes. The percentage of train sequences with capacity consumption below 100%, thereby indicate the amount of possible timetables, which are used to determine the network capacity. The model requires no input of timetables and is therefore found applicable to evaluate the impact on the network capacity consumption, when relocating traffic from the main line passing Ringsted to the Lille Syd line. In this analysis, it was seen that the percentage of possible train sequences on the main line increased from 1.0% to 18.4%, when redirecting and thereby relieving the line. By analysing the associated route passing Lille Syd, the network capacity was 12.6 % for a non-upgraded Lille Syd line, and 7.6% or 4.2% when upgrading the line to 160 km/h or 200 km/h respectively. Interpretation of these results has led to the conclusion, that the network capacity is improved when including Lille Syd in the Copenhagen-Fehmarn corridor. However, a missing frame of reference makes it difficult to conclude anything in relation to the magnitude of these percentages. To evaluate the result, it will be necessary to define the acceptable 161

178 162 DISCUSSION percentage for running traffic on the line. Effort was therefore put into translating the model outputs into edge capacity results, which are more comparable with the UIC406 method. Because the method evaluates the capacity consumption by compressing train routes, the results are however not directly comparable with UIC406, still, a better indication of how to interpret the results was expected. The analysis showed however, that the variance in the permutation of train sequences was too large to interpret the results. To enable evaluation of the network capacity, a fixed train sequence is required, which contradicts the model s original purpose of assessing network capacity without timetables. In a further study, it would be interesting to apply the model on a existing network and assess the model output in relation to the capacity consumption found by applying the UIC406 method. Such analysis could maybe contribute to defining a level of reference to validate the results. An issue with using the model is, that it does not allow for overtakings in its current edition. The model is therefore more inflexible than reality, which possibly results in a higher capacity consumptions. It has therefore been necessary to predefine the locations for overtakings. This has been done based on future expectations of overtakings in the section between Ringsted and Nykøbing Falster, with the Fehmarn Belt link implemented, while the remaining part of the line is based on a guestimate. Throughout this project, focus has been on addressing the direction from Copenhagen to Nykøbing Falster in the most trafficked hour. However, results of the reverse direction has also been obtained. The found network capacity from the reverse direction is placed in Appendix G. These results show, that the percentage of possible train runs is almost twice as low in the direction from Nykøbing Falster to Copenhagen compared to the reverse direction. This seems odd, given that the speed profile is mirrored and the number of trains run in each direction is equal. The only explanation is therefore found in relation to the position of overtakings. This indicates, that the model is very sensitive when changing the overtaking pattern. In a further analysis, it would therefore be interesting to make a sensitivity analysis of the impact of overtakings on the network capacity. One of the basic assumptions was to retain Rail Net Denmark s operation plan and not change the stop pattern. Trains moved to the Lille Syd line included therefore only lines with no stop between Køge Nord and Næstved. In reality, it is expected that the inclusion of the Lille Syd line in the corridor between Copenhagen and Fehmarn, will result in a new operation plan. It is however expected, that the number of trains will be kept in the same level. Because Jensen s model solely considers the number of trains, it is expected that a changed operation plan will not change the overall results. Another uncertainty of the model input is found in relation to the type of rolling stock. In this analysis, it is assumed that direct trains running between Copenhagen and Hamburg or Copenhagen and Jutland will be able to exploit the 250 km/h design speed of the Copenhagen-Ringsted high speed line. The travel times of these trains are therefore estimated according to the train characteristics of a Velaro train. If this assumption does not hold true, the running times for the direct trains should then be prolonged. This will most likely improve the overall network s capacity consumption, because the fast trains will not that easily catch up with the slow freight trains. Furthermore, it will entail that the travel time related advantage of running on the Copenhagen-Ringsted line will be

179 CHAPTER reduced. The difference in travel time for the direct line 20 when running on the main line compared the Lille Syd line is thereby smaller. Finally, uncertainties are associated with the expected track layout and speed profile. The Copenhagen-Ringsted-Fehmarn corridor is expected to undergo a series of chances before the opening of the fixed Fehmarn Belt link. Alterations in these projects can therefore change the foundation for current analysis. The largest uncertainty identified, is the future Ringsted Station. In present analysis it has been assumed, that a flyover is constructed towards Fehmarn. If this is not the case, conflicts at Ringsted will occur and increase the capacity consumption. To limit these conflicts, it will be more desirable to relocate traffic to the Lille Syd. Based on the findings in the capacity analysis, the realisation of the project was investigated in relation to construction. This was done according to the norm foundation. The project proposal consists of both construction of new track and upgrade of the existing. The alignment of the new track has not been considered. Instead, the proposed alignment solution is based on the potential for upgrading the existing track. In reality, it will be necessary to project the new track, however, this level of detail is considered too large for this initial screening phase. The upgrade solutions are composed by a three step method including; optimisation of cant size, prolongation of transition curves and straightening of curve radii, where first step is conducted by the optimisation model presented in chapter 6. This method considers each curve individually, and it is therefore not possible to make large changes to the overall track geometry (e.g. removing elements and merging curves). This shortage has led to the result, that the speed profile at Køge and Næstved could not reach the target speed of 160 km/h and 200 km/h respectively. It would be interesting to project the alignment in Inrail, with the possibility of making more radical changes to obtain an smoother speed profile. This could include investigation of shunts around Køge and Næstved, as proposed by Nielsen in chapter 2. The project description is based on a series of assumptions, which have been highlighted throughout the report. One of the assumptions is the project delimitation at Køge Nord Station, where the Copenhagen-Ringsted line currently is constructing the interchange to Lille Syd as a single track tunnel. This raises the question of whether extending the Lille Syd line to double track is a realistic solution in the light of the ongoing surrounding projects. For instance, the Copenhagen-Ringsted line prepares overtaking tracks immediately after the coupling to Lille Syd, while the Ringsted-Fehmarn project prepares overtaking tracks between Ringsted and Næstved. Investments are therefore already made in relation to running freight trains passing Ringsted. Furthermore, the Electrification Programme prepares only single line bridges on the Lille Syd line. These projects suggest that no short-term vision exists for extending the Lille Syd line to double track, due to the investments earmarked for the main line passing Ringsted. In the long-term planning, the Lille Syd line could possibly play a role if a future need for increase in freight train paths occurs. The line has showed large potential in relation hereto, especially if a wish of separating slow and fast running trains should occur.

180 164 DISCUSSION

181 Chapter 12 Conclusion The current project has treated the possibility of introducing the Lille Syd line as an alternative route in the Danish TEN corridor from Fehmarn to Copenhagen. Based on a timetable for 2027, a network capacity analysis of the upgraded line Copenhagen-Ringsted- Fehmarn was conducted. The analysis was carried out using a recently developed capacity model created by Jensen. The results showed a rather high network capacity utilisation. With the goal of relieving the capacity utilisation, increasing the flexibility, and enabling possible travel time savings, an additional network capacity analysis, including the Lille Syd line in the corridor, was conducted. An initial capacity analysis of the single track line concluded that a double track was a prerequisite in order to allow for any additional traffic. Besides the implementation of a double track, three different speed alternatives for the line were proposed. The network capacity of the corridor was then investigated with the inclusion of a double track Lille Syd line and upgraded to line speeds of 160 and 200 km/h. Two passenger train line variants and two freight train paths were relocated to the line to relieve the main line via Ringsted. Investigations of the network capacity concluded that relocating traffic would relieve the capacity consumption. However, the network capacity is aggravated due to increased the speed on the line. This is caused by the higher heterogeneity level since fast passenger trains will catch up with slow running freight trains. Despite positive results when considering the entire network, it was not possible to investigate the edge capacity. For the alternative route via Lille Syd it can be concluded that passenger trains only obtain time savings if the line is upgraded to at least 160 km/h. Freight trains obtain significant time savings regardless of the line upgrades, due to a maximum speed of 100 km/h. To evaluate the option of upgrading the existing line and implementing a second track, a project description of an overall line design was presented. The proposed project solutions were all evaluated based on an overall project design and the related estimation of construction cost. The costs of the two project proposals are estimated to 1.76 billion for the 160 km/h solution and 2.64 billion for the 200 km/h solution. The largest costs are related to the expansion of the existing bridges. This is, however, a prerequisite for implementing the double track and the cost can only be reduced if bridges are closed. The travel time savings for the design of the proposed solutions show that passenger trains obtain no notable time savings despite the shorter route, however, freight trains gain up to 5.5 minutes per train regardless of any speed upgrade solutions. Based on the general lack of travel time savings for passenger trains an the rather high construction costs, it has been concluded that a further analysis of constructing a double track for a 165

182 166 CONCLUSION line speed higher than the existing is not recommended. However, it has been proposed that additional investigations on implementing a second track along the Lille Syd line to relocate freight transport are initialised. The relocation of freight trains from a large part of the new high speed line Copenhagen-Ringsted will result in capacity and flexibility improvements and furthermore entail significant travel time savings for freight trains. An important goal in the present thesis was to investigate how to relieve and optimise the process of the track design. An optimisation model for determining the optimised cant size according to curve characteristics was composed for that purpose. In its current state, the model optimises the chosen cant solution based on curvatures and line characteristics of placements of turnouts and platforms. The optimisation model furthermore allows for an easy investigation of the impacts if allowing international track norms. To develop and improve the model, implementation of calculations for the impact of curve straightening and related displacements are proposed. A direct link to the running time programme developed by Jensen, in his master thesis, will allow a fast evaluation of the possible obtained speed according to cant optimisation and possible change in infrastructure. The model is proposed to be developed further. A flow chart presents the different processes the project has completed. It illustrates the project work flow and clarifies the processes a project concerning speed upgrades and construction of new built line is undergoing. It is suggested, that the flow can be improved by including greater integration between the traffic planning process and the infrastructure design. The optimised process flow proposes the implementation of an early socio-economic evaluation of alternative infrastructural initiatives combined with the related cost estimation. This will contribute to an iterative process which is crucial for finding the best possible project solution. The project proposes a further development of the process flow to be included in the screening phase of upgrading projects.

183 Bibliography Accident Investigation Board Denmark. Personbil ramt af tog i overkørsel Stokholmvej ved Vinderup Underretning Fakta, K. P. Bevensee. Skørping Borgerforening, URL skoerpingborgerforening.dk/m/side.asp?id= H. E. Boysen. Øresund and Fehmarnbelt high-capacity rail corridor standards updated. Journal of Rail Transport Planning & Management, 4(3):44 58, ISSN doi: /j.jrtpm URL S D. M. Brandt, M., Morberg. Ny anlægsbudgettering påbanen URL trafikdage.dk/papers_2010/praesentationer/410_diddemohrmorberg.pdf. European Commission. Infrastructure - TEN-T - Connecting Europe, URL corridors/index_en.htm. COWI. Analyse af muligheden for at øge toghastigheden gennem overkørsler fra 140 til 160 km/t. Teknisk Notat, pages 1 15, COWI. Screening af regionalbaner til 160 km/t. Technical Report December, COWI A/S, Kongens Lyngby, Danish Transport Authority. Stationsstrukturen i Danmark - en screening af det statslige jernbanenet udenfor Hovedstadsområdet. Technical Report December, Danish Transport Authority. Ulykker og nærved hændelser på jernbanen. In Trafikstyrelsens Sikkerhedskonference, pages 1 21, 2010a. Danish Transport Authority. TSI - implementering og anvendelse. In Fra tanke til handling, pages 1 21, 2010b. URL FraTankeTilHandling-ClausRoser.pdf. Danish Transport Authority. Trafikplan for den statslige jernbane : Høringsudgave. Technical report, Danish Transport Authority. Gods på banen, 2014a. URL dk/da/kollektiv-trafik/trafikale-analyser/gods-paa-banen.aspx. Danish Transport Authority. Optimering af stationsstrukturen. Technical report, Danish Transport Authority, 2014b. COWI Danish Transport Authority. Kombiterminal ved Køge: Markedsanalyse. Technical report, 2010c. 167

184 168 BIBLIOGRAPHY Den Europæiske Unions Tidende. Kommissionens Forordning (EU) Nr. 1299/2014, 2014a. URL _356_R_0001&from=EN. Den Europæiske Unions Tidende. Kommissionens forordning (EU) Nr. 1300/ (1300): , 2014b. DSB. Østtælling 2008 DSB og DSB S-tog. Technical report, DSB. Fremtidens tog - Beslutningsoplæg DSB Kommunikation og Banedanmark Kommunikation. Stort sporarbejde påsydbanen, URL stort-sporarbejde-pa-sydbanen/. B. Elkjær. Jernbaneoverkørsler: Hver femte ulykke sker i Vestjylland, URL http: // C. Esveld. Modern Railway Track. MRT-Productions, Delft, 3.1 edition, European Commission. White paper: European transport policy for 2010, URL change/l24007_en.htm. European Commission. White Paper: Roadmap to a Single European Transport Area, URL DC0144. European Commission. The European Rail Network for Competitive Freight, URL freight_oriented_network_en.htm. European Committee For Standardization. EN : Railway application - Track - Track alignment design parameters - Track gauges 1435 mm and wider - Part 2: Switches and crossings and comparable alignment design situations with abrupt changes of curvature, European Committee For Standardization. EN : Railway applications - Track - Track alignment design parameters - Track gauges 1435 mm and wider - part 1: Plain line, European committee for standardization. EN : Railway Applications - Gauges - Part 3: Structure Gauges, European Railway Agency. Guide for the application of Technical Specifications for Interoperability (TSIs). Technical report, URL Document-Register/Documents/IU-TSI-Guide.pdf. European Railway Agency. TSIs Development, URL Core-Activities/Interoperability/Pages/TSIsDevelopment.aspx. Folketingets Trafikudvalg. TRU Alm.del. spørgsmå l 140, URL samling/20061/almdel/tru/spm/140/svar/335547/375800/index.htm. R. Fongemie. Introduction to the Permanent Way. In With emphasis on Danish infrastructure, pages 1 96, Copenhagen, Rail Net Denmark.

185 CHAPTER Grontmij. Programfaserapport - granskningsudgave: Anlæg af dobbeltspor i Sønderjylland (Vamdrup-Vojens). Technical Report December, Grontmij A/S, Glostrup, Grontmij. Programfaserapport Køge Nord - Næstved. Elektrificering og hastighedsopgradering. Technical report, Grontmij A/S, Glostrup, Grontmij. Additional Project Proposal - Speed Upgrade Hobro-Aalborg. Technical report, Grontmij A/S, Glostrup, 2015a. Grontmij. Hastighedsopgradering Ringsted-Odense - Projekteringsgrundlag. Technical report, Grontmij A/S, Glostrup, 2015b. Grontmij. Opgradering af Hillerød Station. Technical report, Grontmij A/S, Glostrup, 2015c. Grontmij. Anlægsbeskrivelse - Fagnotat: Hastighedsopgradering Hobro Aalborg (supplerende VVM). Technical report, Grontmij A/S, Glostrup, 2015d. Fonverne. I. ILCAD: Act Safely at level crossings, URL ILCAD-2013.html. International Union of Railways. Capacity. UIC Code 406, Original(1st edition):1 26, International Union of Railways. Safety measures to be taken at level crossings on lines operated from 120 to 200 km/h. UIC Code 762, Original(2nd edition), K. S. Jacobsen and J. B. Meyer. Optimering af sporkonstruktion ved hastigheder over 200 km/h. Technical report, Technical University of Denmark, Lyngby, L. W. Jensen. Opgradering af jernbanen mellem Odense og Aarhus. Technical report, Technical University of Denmark, Lyngby, L. W. Jensen. Jernbaneinfrastruktur, antal tog og hastighedsprofiler Rasmussen M.B Jensen, R. Hastighedsopgradering af Vejle-Herning. Technical report, Technical University of Denmark, Lyngby, 2014a. S. U. Jensen. Evaluering af forsøg med tiltag ved jernbaneoverkørsler. Technical report, Trafitec, Scion-DTU, Lyngby, 2014b. O. A. Jensen, L. W., Landex, A. and Nielsen. Assessment of Stochastic Capacity Consumption in Railway Networks. Technical report, Department of Transport, Technical University of Denmark, Ramboll Denmark A/S, Lyngby, M.B Jensen, R., Rasmussen. Course : Signaling Systems and Technology for Railways. Technical report, Technical University of Denmark, Lyngby, O. Kien. Hvorfor Er Den Indenlandske Godstrafik På Jernbane Stort Set Forsvundet? In Banebranchen, pages 1 17, Copenhagen, Rambøll. A. Landex. Capacity Statement for Railways. Technical report, Aalborg, Denmark, URL file_ /content.

186 170 BIBLIOGRAPHY A. Landex. Methods to estimate railway capacity and passenger delays. Technical report, DTU Transport: Department of Transport, Lyngby, A. Landex. Evaluation of railway networks with single track operation using the UIC 406 capacity method. Network and Spacial Economics, 9(1):7 23, 2009a. ISSN X. doi: /s A. Landex. Banegodstrafikkens konflikter - Sverige-Tyskland. Trafikdage påaalborg Universitet, (ISSN ):1 26, 2009b. A. Landex. Rail Traffic Engineering. (0.1):1 216, S. Larsen. Ingen køreledninger på den tyske side af Femerntunnelen før i 2024, URL ingen-koereledninger-paa-den-tyske-side-af-femern-tunnelen-foer-i G. Liu and F. Balsby. Teknisk notat vedrørende betydningen af reviderede TSIer på strækningen Hb - Ab J. Lohmann and A. Landex. Planning Public Transport. Technical report, Department of Transport, Technical University of Denmark, Lyngby, B. Marfelt. Nye knudepunkter for jernbanen i København: Glostrup og Ny Ellebjerg, URL nye-knudepunkter-jernbanen-i-koebenhavn-glostrup-og-ny-ellebjerg Ministry of Transport. Green Transport Policy. Technical report, Copenhagen, Ministry of Transport. En moderne jernbane: udmøntning af Togfonden DK. Technical report, Ministry of Transport. Undersøgelse af samdrift i togtrafikken på Sjælland Tommy Gade Møller, Jens Holst and Jensen. ERTMS introduction. pages 1 17, Copenhagen, Rail Net Denmark and Atkins. Sigrid Møller. Elektrificeret Køge-Næstved-bane skal aflaste strækning til Femern, URL elektrificeret-koge-naestved-bane-skal-aflaste-straekning-til-femern O. Mornell. Kartläggning av plankorsningar Trængselskommissionen. Reduktion af trængsel og luftforurening samt modernisering af infrastrukturen i hovedstadsområdet. Technical report, B. Nielsen. Track Alignment Design B. M. Nielsen. Hvad er en banes kapacitet. Trafikdage påaalborg Universitet 2004, pages 1 10, N.F. Nielsen. Banenormer i Banedanmark. In DGF-arrangement Baner og Geoteknik, pages 1 13, Odense, Rail Net Denmark. URL http: // Geoteknisk_Forening_foredrag_ pdf.

187 CHAPTER O. A. Nielsen. Fremtidens behov for Transportkorridorer i Hovedstadsområdet. Arbejdsrapport, Draft(3):1 61, N. Østergaard. Banedanmark: Nye terminaler skal genoplive godstrafik, URL banedanmark-nye-terminaler-skal-genoplive-godstrafik N. Østergaard. Danmarks jernbaner er europamestre i køreplansluft, URL danmarks-jernbaner-er-europamestre-i-koereplansluft Rail Net Danmark and The Danish Transport Authority. Banenotat, ny anlægsbudgettering på baneområ det. (april):1 24, Rail Net Denmark. BN : Fjernbanens Kørestrømsinstruks: Sikkerhedsinstruks for baner elektrificeret med 25kV, 50 Hz. Technical report, Rail Net Denmak, Copenhagen, 2006a. Rail Net Denmark. BN1-11-1: Afvanding af Sporarealer. Technical report, Rail Net Denmark, Copenhagen, 2006b. Rail Net Denmark. BN1-49-1: Indbyrdes placering af spor og perron. Technical Report , Rail Net Denmark, Copenhagen, 2006c. URL filarkiv/5039/bn pdf. Rail Net Denmark. BN2-1-1: Struktur, udseende og udvikling af banenormer. Technical report, Rail Net Denmak, Copenhagen, URL filarkiv/5032/bn2-1-1.pdf. Rail Net Denmark. BN : Sporafstand og frispormærker. Technical report, Rail Net Denmark, Copenhagen, Rail Net Denmark. Kapacitetsanalyse for overbelastet infrastruktur: Årsager og foranstaltninger. Technical report, Rail Net Denmark: Drift, Copenhagen, URL /index.htm. Rail Net Denmark. BN1-9-2: Sikkerheds- og opholdszoner på perroner. Technical report, Rail Net Denmak, Copenhagen, 2012a. Rail Net Denmark. Mere gods påbanen -via markedet mod de politiske mål. ft.dk, (30): 1 54, 2012b. URL / /index.htm. Rail Net Denmark. Ekspropriationer: Lejre-Vipperød, 2012c. URL visartikel.asp?artikelid= Rail Net Denmark. Nu fortsætter sporfornyelsen på Lille Syd, July 2013a. URL http: // Rail Net Denmark. Ringsted-Femern Banen Jernbanen, der binder Europa sammen. page 13, 2013b. URL Rail Net Denmark. Sporregler Technical report, Rail Net Denmak, Copenhagen, 2013c. URL

188 172 BIBLIOGRAPHY Rail Net Denmark. Beslutningsgrundlag: Hastighedsopgradering Køge Nord - Næstved, 2013d. URL pdf. Rail Net Denmark. BN1-6-5: Tværprofiler for ballasteret spor. Technical report, Rail Net Denmark, Copenhagen, 2014a. URL BN1-6-5.pdf. Rail Net Denmark. Nye TEN-T retningslinjer og Connecting Europe Facility (CEF), June 2014b. URL &soegeord=1316/2013. Rail Net Denmark. DSB infrastruktur: Fritrumsprofiler. Technical Report , Rail Net Denmak: Teknisk Drift, Copenhagen, 2014c. URL filarkiv/5051/fritrumsprofiler pdf. Rail Net Denmark. SODB Sikringsanlæggene og deres betjening: Anlægsbestemmelser for Automatisk sikrede overkørsler. Technical report, Rail Net Denmak, 2014d. Rail Net Denmark. Fornyelsesplan e. URL filarkiv/16061/fornyelsesplan pdf. Rail Net Denmark. Screening: Hastighedsopgradering Køge Nord - Næstved, 2014f. Rail Net Denmark. Elektrificering Køge Nord - Næstved, 2015a. URL dk/visbanearbejde.asp?artikelid= Rail Net Denmark. TIB-S: Strækningsoversigter. 2015b. URL filarkiv/18418/tibs_ pdf. C. Rutter. Hvorfor er højdeforskellen mellem tog og perron så stor?, URL hvorfor-er-hojdeforskellen-mellem-tog-og-perron-sa-stor B. Schittenhelm. Planning With Time Supplements in Railway Timetables. Technical report, DTU Transport - Department for transport and Rail Net Denmark, Lyngby, L. Transport Minister: Barfoed. Forslag til Lov om anlæg af en udbygning af Nordvestbanen mellem Lejre og Vipperød, URL R0710.aspx?id= H. D. Transport Minister: Kristensen. Transportudvalget L 156 endeligt svar på spørgsmå l 9 Offentligt. Technical report, Ministry of Transport, Copenhagen, UIC - International Union of Railways. UIC leaflets - Rail Freight Portal, URL

189

190 Technical University of Denmark Anker Engelunds Vej Kgs. Lyngby