TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Transport-, Bygnings- og Boligudvalget 2016-17
TRU Alm.del Bilag 365
Offentligt

Revision

Made by

Checked by

Approved by

Description

Rambøll/Parsons

JACP

Main report

Ref

Document ID

1100021338

147981-8

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

CONTENTS

1.1

3.1

3.2

3.3

3.4

5.1

5.2

5.3

5.4

5.5

6.1

6.2

6.3

6.4

6.5

6.6

6.7

6.8

6.9

7.1

7.2

7.3

7.4

8.1

8.2

9.1

9.2

9.3

9.4

9.5

9.6

10.

10.1

10.2

10.3

11.

EXECUTIVE SUMMARY

Results

READING GUIDE AND REPORT STRUCTURE

INTRODUCTION

Objective for the study

Driverless potential

Background

Preconditions for the investigation

APPROACH AND METHODOLOGY

OPERATIONAL SCENARIOS

Introduction

Conclusion

Analysis

Description of scenarios

Future development

–

passengers and other projects

INFRASTRUCTURE AND ROLLING STOCK

Platform Safety

Safety between stations

Stepless Boarding

Flexibility and Capacity Enhancing Infrastructure

Power Supply

Driverless Rolling Stock

Combining and Splitting of Trainsets

Depot and Workshop

Rollout Plan

ORGANISATION

Introduction

Conclusion

Analysis

Natural staff reduction

RISKS AND COST ESTIMATE

Overall Assessment

Event based Risk overview

FINANCIAL ANALYSIS

Summary

Introduction

Project timeline and discount rate

Detailed results

Break Even Analysis and Payback periods

Sensitivity analysis

SOCIO ECONOMIC ANALYSIS

Introduction

Conclusion and results

Sensitivity

CONCLUSION

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

EXECUTIVE SUMMARY

In accordance with the agreement on

metro, light rail, suburban rail lines and bicycles

of 12

June 2014, Ramboll Denmark, as main advisor together with associated international sub-

consultants, conducted an investigation on the reorganisation of the S-bane to metro operation.

The purpose of the investigation has been to explore the possibilities and perspectives of

reorganising the S-bane to automatic and driverless operation (metro operation) in connection

with the purchase of the next generation of S-trains. The current S-trains lifespan expires in the

years 2026-2036.

Automatic and driverless train operation can form the basis for a vision of better train service for

less money. Automation creates the possibility to achieve higher frequency, increased punctuality,

fewer cancelled trains and lower costs. A reorganisation of the S-bane to metro operation is a self-

financing project, where the necessary investments associated with the reorganisation can be paid

with gains in the operational costs. The project will also allow for improvement of the level of

service on the S-bane, combined with a reduction of government costs for the S-bane.

reorganisation

of the S-bane to metro operation is technically feasible and can be carried out

without significant risks, as automatic operation is part of the basic functionality of the Signalling

Programme which is rolled out on the S-bane in these years. The

reorganisation

to driverless

operation will only require minor adjustments to the Signalling Programme.

Forecasts for the development of the Capital Region in the next decades show that there will be

significant population growth and thus also a significant increase in traffic. Traffic patterns will also

change and further after the commissioning of the Cityringen and the light rail on Ring 3, which

both runs across the S-bane’s

fingers. To reduce congestion and support green

transport, the

public transport system must be able to be expanded smoothly. Therefore there is a need for the

S-bane to be operated in a way that will enable this development to be adapted continuously.

reorganisation

of the S-bane for automatic and driverless operation will make future S-bane

investments significantly more profitable so that the expected increases in demand can be met

without necessarily having to increase the state's subsidy for rail traffic. In addition, the service

level may increase during the evenings, weekends and night hours, where there are no material

and capacity restrictions as in daytime. Thus, the S-bane can offer round the clock service, as we

know it from the metro.

In this investigation, the possibilities and perspectives of driverless operation of the S-trains has

been investigated. The investigation has included operational, capacity related, technical, financial

and personnel matters. The possibilities are compared with the current operation of the S-bane,

where there is a driver in all trains.

1.1

Results

The investigation shows that the overall operational cost savings associated with the transition to

driverless operation are expected to exceed the cost of the necessary investments associated with

the project. Furthermore, it will be possible to extend the scope of operations on the S-bane,

combined with a reduction of the cost of S-train traffic as compared to today.

The investigation has identified two overall scenarios for a

reorganisation

of the S-bane to metro

operation:



Scenario 1:

The S-bane is automated without changing the scope of operation on the S-

bane compared with today. Scenario 1 has a net revenue of DKK 1,311 million and is

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

economically viable based upon an

internal interest rate of return of 6 percent.



Scenario 2:

The S-bane is automated and the scope of operation on the S- bane is

increased by 24 percent compared to today. In scenario 2, the annual number of travellers

will increase by 11 percent. Scenario 2 has a net revenue of DKK 309 million and is

economically viable based upon an

internal interest rate of return of 9 percent.

The savings for each of the scenarios are based on an average life of 30 years for the investments

and are calculated by discounting all investments, costs and gains to 2016 with a

discount rate of

4%.

reorganisation

of the S-bane to metro operation is thus a self-financing project, which in

addition could also improve the S-bane service level to the benefit of passengers.

1.1.1 Investment needs

In addition to the minor adjustments of the Signalling Programme, the reorganisation of the S-

bane to metro operation will require investments in infrastructure and new driverless trains. In

addition, there will be additional costs for the operation and maintenance of a driverless S-bane.

The driver currently handles the arrival and departure of the S-train and ensures that the S-train

can drive arrive at the station and open and close doors at arrival and departure. In case of

driverless operation, a monitoring system is established that detects contaminants on the tracks

and, if necessary, causes an emergency stop of a train. Experience from other railways with

driverless operation shows that by establishing such a system the safety level at the platforms can

be at least at the same level as for the current operations. Along the tracks between the stations

fences shall be established - also on bridges.

At all stations a possibility for level free access (stepless entry) to the train will be established and

the train will be equipped with a gap-filler. In this way, disabled people will be self-reliant on

boarding and leaving the train, where it today requires intervention by the driver with increased

dwell time as a consequence.

The analysis has shown that a combination of 3-car train sets that can be combined to up to three

train sets with a maximum train length of 165 m will be most appropriate to accommodate the

passenger needs. There will be no need for additional workshops or depots for any of the two

scenarios.

Driverless trains are more expensive to acquire than traditionally manned trains. However, this is

offset by the fact that the turnaround time at end stations for driverless S-trains is shorter than for

manned trains. Thus a higher utilization rate can be achieved and the total number of trains can

be reduced when the system is rendered driverless.

1.1.2 Roll-out plan

The investigations have shown that a decision to introduce driverless operation on the S-bane

must be taken before the purchase of new S-trains commences to ensure that the new S-trains

can be used for driverless operation without having to be changed after delivery. As it is a

comprehensive project that requires changes in infrastructure, purchase of new driverless trains

and the completion of a series of tests, it a decision on the project should be made no later than in

2019.

The roll-out of driverless operation on the S-bane must be carried out in a manner that affects the

current operation as little as possible. To ensure this, it is proposed to complete the rollout in

phases where the first phase is a pilot phase, which comprises the Ringbane (Hellerup-Ny

Ellebjerg). The pilot phase shall be used to test the solution and collect experience without

affecting the other lines on the S-bane. Subsequent roll-out of the project on the other lines of the

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

S-bane will then take place in phases where experience from the Ringbane is utilised.

If a decision is made to carry out the project has been taken not later than in 2019, the pilot

phase on the Ringbane can take place in the years 2022-2026, while the roll-out of the other

sections can take place in the years 2025-2030. This is a conceptual schedule, as the rollout of the

project can be postponed by lifetime extension of existing S-trains, although this can be costly and

impact the operating stability of the S-bane.

The technology-based technological solutions on which the project is based are all in operation on

one or more existing driverless railways in urban areas. The envisaged rollout plan, in which the

project will roll out over the next decade, will probably imply that new technologies in the

meantime will emerge and can possibly be used for the final rollout of the project. These yet

unknown technologies may potentially contribute to further improvement of the safety level and

thus provide even more attractive solutions than those described - also in regards to investments

and operational purposes.

1.1.3 Human resources

Remuneration of personnel directly involved in the operation of the S-bane is today a significant

cost driver. With the reorganisation of the S-bane to driverless operation, the will be no need for

drivers and other operational personnel, while a small number of train stewards will be required to

carry out service tasks currently carried out by drivers.

It is considered possible to carry out the project without having to dismiss staff, when taking into

account retirements and potential transfer of staff to the long distance and regional lines. Specific

models for dealing with excess staff should be investigated in a possible later phase. The

immediate human resource issue is primarily a question of maintaining the required number of

employees at the S-bane during the transition period, as some employees might be expected to

consider changing jobs as a result of a political decision to make the S-bane driverless. As part of

the project it must be ensured that a sufficient number of drivers are available during the

transition phase.

As a result of a reduction of the need for operational-oriented personnel by 90%, it will be possible

for each of the proposed scenarios to increase the capacity in periods when it is not fully utilized

without significant changes in operating costs. This implies that the business model for operating

the S-bane with driverless operation will be significantly different from the business model for the

current mode of operation, where the operating costs are highly dependent on the amount of

trains operated. Thus, the driverless operation will contribute to a significant expansion of the

capacity and new more flexible operational opportunities.

1.1.4 Metro-like operation and classical operation

As part of the investigation, a comparison was made of the advantages and disadvantages of

classical and metro-like operations on the S-bane in connection with a conversion to driverless

operation. Classical operation involves driving a mix of fast trains that stop at selected stations

and trains that stop at all stations. Metro-like operation implies that all trains stop at all stations.

The studies show that the choice between the two types of operation does not significantly affect

the outcome of the project. Thus, a reorganisation of the S-bane to automatic and driverless

operation can be carried out independently of the choice between the two operating models. The

choice between metro-like and classic operation - or possibly a combination of both - can be taken

in conjunction with the timetable planning in the wake of an S-bane reorganisation to automatic

and driverless operation. In today's situation, DSB switches between the two operating models

during the operating day.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

1.1.5

Summary - key figures

The passenger related and financial key figures for the project are shown in the tables below.

The effects of the project are calculated as additional costs to a base scenario where the S-trains

are replaced with new manned S-trains, and the level of service of the S-bane remains unchanged

as compared with today.

Economic effects in relation to basis

Present values*, DKK Mio. 2016 prices

Costs

Infrastructure

- CBTC upgrade

- Stepless entry

- Platform-edge safety

- Fencing along rails

- Upgrade of traction power

Trains

- Train needs

- Driverless technology

Operations and maintenance

- Train maintenance and power supply etc.

- Maintenance of infrastructure

Contractor administration cost

Total costs

Revenue

Ticket sales

Reduction of staff

Total revenue

Net revenue

Payback time (Break-even)

Note: "Summary error" is due to rounding up.

* This is the present value related to the financial cash flow, which should not be mixed up with the socio-

economic present value from the socio-economic cost-benefit analysis.

** S-train demand is reduced in scenario 1, although the production volumes remain unchanged as compared

with today, as driverless S-trains have shorter turnaround times and thus have a higher utilization rate.

*** Ticket sales are reduced in scenario 1, due to the marginally extended dwell times at stations (and thus

travel times) as a result of on-board gap fillers (stepless boarding). This marginal change will affect the traffic

model calculations, but is unlikely to make any significance in practice.

Scenario 1

Scenario 2

1.086

671

329

-16

-201**

184

810

138

672

134

2.014

-39***

3.363

3.324

1.141

671

329

756

521

235

1.989

1.328

662

262

4.149

1.095

3.363

4.458

1.311

18 years

309

19,5 years

Passager-based effects in relations to

basis (DKK)

Scope of production

Growth as compared with basis

Number of annual travellers on the S-bane

Growth as compared with basis

Number of annual collective travellers in

the capital area

Growth as compared with basis

Economy (Internal

rate of return)

Scenario 1

15,9 mio trainkm

111,9 mio.

373,8 mio.

6,1 %

Scenario 2

19,7 mio. trainkm

24 %

124,7 mio.

11 %

378,1 mio.

9,6 %

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

READING GUIDE AND REPORT STRUCTURE

The study of reorganisation of the S-bane for driverless operation is documented in a main report

(this paper) and a number of sub-reports.

The main report provides an overview of the study highlighting and describing conclusions and key

findings within the different focus areas of the study. Each of focus areas has an attached sub-

report, where a more detailed analysis and documentation of the conclusions and findings can be

found. The main report is divided into 9 main sections:

1. Executive summary

This section summarises the objective and conclusion of the study.

2. Reading guide and report structure

This section explains the structure of the main report and links to sub-reports where a

more comprehensive analysis can be found.

3. Introduction

The introduction outlines the objective and background of the study and highlights its key

preconditions.

4. Methodology

The methodology section describes the overall approach and method applied in the study.

5. Operational scenarios

This section describes the analysis, assessment and conclusions made in relation to the

operational concepts and scenarios. The detailed analysis can be found in Appendix 1:

Operational Plans.

6. Infrastructure and rolling stock

This section concerns the different infrastructure and rolling stock focus areas of the study.

Each item has its own sub-report:

Appendix 2: Safety (Platforms, etc.)

Appendix 3: Stepless Boarding

Appendix 4: Track Capacity in the Central Section

Appendix 5: Flexibility and Capacity Enhancing Infrastructure

Appendix 6: Driverless Rolling Stock

Appendix 7: Combining and Splitting or fixed Trainsets

Appendix 8: Depot and Workshop Facilities

Appendix 9: Power

Appendix 10: Rollout Plan

7. Organisation

This section describes the organisational analysis and assessment. The detailed

investigation can be found in Appendix 12: Organisation.

8. Risk and cost estimate

This section describes the risk analysis. The detailed analysis can be found in appendix 11:

Risk and cost estimate.

9. Financial analysis

This section describes the financial analysis. The detailed investigation can be found in

Appendix 13: Financial and socio economic assessment.

10. Socio economic assessment

This section describes the socio economic analysis. The detailed investigation can be found

in Appendix 13: Financial and socio economic assessment.

11. Conclusion

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

3.1

INTRODUCTION

Objective for the study

According to plan, the next generation of Rolling Stock on the S-bane in Copenhagen shall be

procured within the next decade. As part of the long term preparation for that the general

requirements, feasibilities and context related factors must be investigated and assessed.

Technological innovation must e.g. expect to lead to potentials quite different from the mid-

nineties where the existing generation of Rolling Stock were implemented. Equally, Greater

Copenhagen has changed substantially during the last decade, driven by growth and prosperity,

which has influenced the general traffic patterns.

In this overall context, the objective of the study is to deliver fact based pointers and feasibility

assessments that can initiate the next step in a decision making process that will eventually bring

the next generation of the S-bane successfully into operation.

A particular focus in the study is related to unattended train operation, i.e. mode without any

driver. An increasing number of urban traffic systems worldwide have already transferred, or will

in the upcoming years introduce UTO. The key questions related to that are to which extent will

such solutions could fit to the Copenhagen S-bane from a technical, a financial and a service

provision perspective.

Accordingly, the methodologic frame for assessment of the overall objective is to:



Investigate a financial potential associated with UTO, documented in a formalised Business

Case



Clarify the span of feasible technical solutions that can support the business case



Ensure that certain service requirements, defined as more or less formal precondition, can

be met

3.2

Driverless potential

Driverless systems have several superior characteristics compared to a regular attended system.

It’s more flexible

in situations where it is necessary to recover from irregular operation, as the

operation is independent of the drivers and their rosters. Also, extra trains are easier to put into

operation, e.g. if additional capacity is needed due to special events. Furthermore, the absence of

drivers ensures that dwells are kept within the defined dwell time. Situations with a shortage of

drivers (e.g. due to illness) are also avoided in a driverless system.

However, from an optimization perspective of the overall operation, the single most powerful

potential of the UTO mode emerged from the fact, that costs related to operation of one long train

equals two trains of half the size and with double the frequency. Accordingly,

it’s possible to satisfy

a given capacity demand by smaller units operated with higher frequency for the benefit of

customers. Often, the higher frequency makes official time plans redundant due to the very high

availability. Equally, transfers from one line to another, or between different transportation

systems are perceived as very smooth and easy. These factors therefore attract more customers

and make the public transportion more competitive.

This optimized operation is usually enabled by simple network layouts, where train units and

frequency can be combined very flexibly. As the S-bane merges out of several distinct lines into

one central section, makes it a little more challenging. Also, the layout makes the impact of

disruptions or significant delays more complex. The driverless potential needs therefore to be

derived without creating bottlenecks in the inner section, or a too low frequency provision on the

outer lines.

That’s the fundamental

objective in context.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Remain current safety level

A major challenge for converting a system like the S-bane into a driverless

system is the safety requirements and approval process. The investigation

is therefore based on an approach where the safety level needs to be at

least the same level going from STO to UTO operation.

cal

Same service provision for special segments

The service provision is likewise assumed to be at least the same across

needs of different user groups; in particular access for disabled people

needs to be at least the same.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

APPROACH AND METHODOLOGY

An assessment of the possibilities for introduction of UTO on the S-bane requires investigations

into a wide range of operational, technical and organisational focus areas. The areas are to a wide

extent mutually interdependent and the result from one focus area serve, might have an impact in

the investigations in other areas. For that reason, the study has been divided into distinct focus

areas making cross analysis more transparent. The areas are:

Best performing operational scenarios

Feasible and high potential technical solutions related to infrastructure and rolling stock

Analysis of associated organisational impact

Total financial and social economic assessment

The investigation of operational scenarios has been designed in order to develop a number of

potential alternatives making determination of the best performing as clear and significant as

possible. Through an iterative process, preferred scenarios have been defined and selected for

further investigations based upon relevant key parameters covering operational and financial

indicators. The basic idea has been to explore how different potential

baseline cases

defined from

the particular design characteristics of the total railway network

–

bottlenecks in the central

section, distant outer lines, existence of a ring line, etc.

–

can be used to optimize performance.

Subsequently, derived from that approach a considerable number of different operational

scenarios, equally distributed between Classic and Metro, have been assessed. Each one of them

has been investigated in order to find promising patterns, potentially ready for further in-depth

financial analyses. Variations associated with any non performing baseline cases at the other hand,

has been excluded. The outcome of these analyses has led to adjustment and a narrowing of

competing scenarios.

For the above mentioned processes PRIME

has been used for technical and operational analysis,

and the OTM

has been used for calculation of the passenger impact

The technical investigations

–

covering safety, infrastructure and rolling stock

–

have included

identification of alternative solutions and assessment of their suitability for the S-Bane with a clear

separation between

need to have

and

nice to have

investments. For each area of investment

budgets and operational budgets have been established to serve as input for the financial analysis.

Some elements of the technical investigation and investigation of operational scenarios are directly

linked to driverless operation (e.g. platform safety), while others are more indirectly linked (e.g.

coupling of train units). The study will highlight the distinction between elements and solutions

that are directly and indirectly linked to driverless operation.

Both the investigation of operational scenarios and technical solutions end up with a number of

recommendations. It is however important to emphasize that the recommendations shall be

perceived as

conceptual design choices,

aiming at proof of concept for driverless operation. This

implies that the recommended solutions

–

and their attributes and characteristics

–

will change

along with the specification of the project and the technical development within the industry.

PRIME is a simulation model developed and provided by Parsons enabling a structured investigation of travel time, dwell time, required

The OTM model is based on data (matrixes of passenger patterns) from which impact on different operational models can be estimated.

number of rolling stock, etc. related to a given operational pattern

Existing version of OTM are based on the latest calibration from 2009. Accordingly, trends observed during the last 7 to 8 years are not

explicitly captured or explained in OTM

Results from PRIME and OTM have been quality assured separately and validated cross functionally in order to ensure quality assurance

of the outcome

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

An example could be the recommended safety solution at the platform

–

the recommended

solution is relatively specific in order to define its characteristics (cost, safely level, dwell time,

etc.). It is however relatively certain that the final design of this solution will differ from the

described solution, due to the technical development, etc.

When it comes to the organisation analysis, it has been based upon the current situation which

serves as base for assessment and for the changes to be imposed. Based upon the operational and

technical assessments, the required changes can be identified. In particular, the organisational

analysis derived from the suggested roll out scenarios in setting the pace of transformation of

staff, equally figures and assumptions about current relevant staff are included. The financial

impact of these changes serves equally as input for the financial analysis.

The financial and economical assessment is based on the results from the three other

investigations. The business case is made from a delta perspective from STO to UTO mode. The

starting point for the investments in the business case is a fully implemented signalling

programme and procurement of new rolling stock in any case. That has the consequence that

comprehended financial impacts are isolated to any income, capex or opex that differs from STO

compared to UTO. The business case is furthermore prepared in line with

Ny Anlægsbudgettering.

Finally, approach and methodology have been customized and impacted according to international

operational real life studies. Visits to Nuremberg, Munich, the Copenhagen Metro as well as visit

from Stockholm, have brought in valuable consistency checks and inspiration to the definition of

solutions: Recommended platform safety solution, operational handling of false alarms, coupling

feasibility and operational scenarios, are examples of subjects that have been heavily influenced

by these real life studies. In general, none of the suggested solutions are truly green fields. Any

mentioned technology has proven its feasibility in commercial operation, only the application in the

specific context, environment and layout of the S-bane is unique.

Throughout the process, results and findings have been discussed and assessed with the Ministry

of Transport, Building and Housing, Banedanmark, DSB, Copenhagen Metro, Banedanmark (Rail

Net Denmark) and the Danish Transport and Construction Agency, who jointly have in-depth

knowledge and experience in operation of metropolitan railways.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

5.1

OPERATIONAL SCENARIOS

Introduction

The current operation of the S-Bane is based upon trains attended by a driver. By introduction of

driverless trains it will be possible to change the operation and thus improve the utilisation of the

S-Bane network. Previous investigations

have shown that a better utilisation of the S-Bane would

lead to an increased number of passengers.

In order to identify the operation solution that can provide the best possible business case, a range

of scenarios has been analysed. The overall approach has been to investigate the impact of

different benchmark scenarios, each with the objective to derive the maximum potential from a

driverless operation in a rail network with a structure that does not have the typical driverless

simplicity. Goals of the analysis have thus been to reduce the pressure on the central section, to

increase frequency at some of the outermost sections, or to reduce cost by removal of lines with

the fewest number of passengers. The benchmark scenarios have included:



Use of

Ringbanen

instead of the central section by one of the northern lines

Use of turn tracks enabling a train going back to an outer line before reaching the central

section

Alternative uses of capacity derived from use of Ringbanen or use of turn tracks

Each of the benchmark scenarios has then been combined with different operational styles, i.e.

Classic and Metro, and has been assessed with alternative ranges of frequency from 30 over 36 to

39 hourly departures in the central section.

16 scenarios in total have been derived from a combination of these parameters. The tools OTM

and PRIME have then been used for assessments in several steps, and increasingly more detailed,

to identify the best performing scenario.

The whole analysis regarding operational scenarios can be found in appendix 1.

5.2

Conclusion

Based on the analysis it is found that the single best performing scenario consists of a flexible

combination of Classic- and Metro style with 36 hourly departures in the central section, named

Flex 36. This conclusion is based on the entire assessment comprehending both the financial and

the socio economic analysis, while a partial assessment relying only on the financial business case

makes the UTO30 as the best performing. For further details about these analyses, please refer to

chapter 9 and chapter 10.

Classic style is used in peak periods, while Metro style is used outside peak periods. This is a

principle that DSB, to a lesser extent, will already initiates by the end of January 2017 with metro

style operation in the evenings and classic the remaining part of the day. However, in the Flex 36

the extent to which the styles are mixed is higher. The following conclusions can be derived from

this, as well as from the total analysis:

1. Combined benefits from a reduction in the total required number of rolling stock units, as

fewer trains are needed in a Classic style during peak hours and - at the same time - the

advantages of Metro that can be exploited the remaining part of time

2. The outer lines do not have enough passengers to let a Metro style serve to the very end.

Instead, it is beneficial to let half of the

train’s

turnaround between the beginning and the

See e.g. Parsons 2010 or Screening from 2013

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

end of each

finger,

e.g. Solrød, Ballerup, Buddinge and Holte. Only at Klampenborg and

Høje Taastrup every train goes to the end

3. A frequency of 39 hourly departures in the central section does not yield significantly more

passengers and the marginal extra revenue does not match the cost of additional train

units. At the same time 39 trains per hour is a heavy utilisation of the central section that

will cause a significant degradation of the service quality. A maximum of 36 trains per hour

is therefore also recommended in the central section

4. Offloading of the central section by introducing different utilisations using the Ring line, or

establishing turnarounds in order to increase train capacity does not lead to better

performing scenarios

5. In general the distinction between Metro and Classic shall rather be considered as a

continuous space, each concept with multiple variants, enabling the two styles to become

almost identical. E.g. a train only skipping a few stations on the outer part of the network

is per

definition considered as ‘Classic’ even though it

operates according to the metro

definition on the remaining part of the network. This acknowledgement lead to interesting

results: An optimized traffic pattern is much more about identifying and combining the

potentials across the two concepts, rather than taking a binary choice. Even more

interesting: That allows for a potential decision for UTO operation without requirement of a

detailed defined operational pattern in parallel. The latter can easily be a part of the initial

analysis after a formalized project has been initiated

The key operational characteristics for the UTO Flex 36 are shown in the table below.

Flexible scenario -

36 trains per hour

(Best Performing Scenario)

63.249

175

399.714

6,3

Indicator

Base

STO

50.918

161

359.111

7,1

Train km

Train sets

Passenger trips

Trips per train km

Table 1: Key operational data for base scenario and flexible, best performing scenario

5.3

Analysis

The analysis aiming at identifying the most suitable operational scenario has included a definition

of number of scenarios covering metro style and classic operation. Through that process the

scenarios have been compared and the most favourable have been brought forward for further

analysis.

It has been a main objective for the analysis to keep a comparison between metro- and classic

style. The initial pool of scenarios is therefore characterised by scenarios in pairs, i.e. metro and

classic scenarios based on the same overall concept. The following concepts have been analysed in

different combinations:



36 to 39 trains per hour in the central section

All trains running to the terminal stations (end of the fingers) or only some of the trains

Alternative utilisation of the S-Bane network aiming at offloading the central section incl.:



Additional trains from Hillerød via the Ring line to Ny Ellebjerg



Establishing turnaround at Enghave for some trains from Ballerup/Frederikssund and

thus allowing for extra trains to/from the other fingers through the central section

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

5.3.1

Assumptions and preconditions for the analysis

Utilisation in the central section

The number of trains that can be operated in the central section (Dybbølsbro-Svanemøllen) has

been analysed using the PRIME model. Based on recommendations by the International Union of

Railways (UIC), the maximum utilisation has been derived as 40 trains in the peak period and 33

trains in the off-peak period. Consequently, operational scenarios with a high capacity utilisation of

39 trains per hour and medium capacity utilisation scenarios with 36 trains per hour in the central

section have been composed. Simulations of delay propagation has subsequently led to the

conclusion that 36 trains per hour is the maximum obtainable capacity, as 39 trains per hour will

lead to a significant decrease in punctuality compared to the base STO scenario.

Public timetable

In the Copenhagen Metro timetables are not published. This provides additional flexibility as the

train services can be easily adjusted. In turn, this added flexibility can be an incentive to reduce

the running time supplements and thus provide faster travel times. Furthermore, the arrival

pattern of passengers is affected by the timetable being published or not. An assessment has

therefore been carried out in order to investigate whether the same flexibility can be obtained on

the S-Bane if the timetable is not published.

The assessment shows that a published timetable is the preferable solution for the scenarios as

the flexibility obtained by not publishing a timetable cannot offset the increased waiting time. A

main factor in this is that the Copenhagen S-bane network consists of six branches of varying

length which all pass the heavily utilised central section. A timetable is needed to be able to

effectively utilise the central section capacity and at the same time serve different destinations.

This “timetable” does not necessarily have to be published, but

emphasizes the fact that the

structure of the system reduces the dispatching flexibility that could otherwise be obtained (e.g. if

the capacity in central section was significantly higher).

Because of low frequency at the outer stations, especially at mornings and evenings, and also

because of the importance of connections to low frequency bus-services, a published timetable is

clearly recommended.

Train Frequencies during the Day

To obtain a suitable balance between production and demand, i.e. the service provided and the

number of passengers, the frequency is adjusted outside the peak periods. Table 2 shows the

frequencies during the day based on the frequency in the central section in the morning peak. It

has been the objective to achieve a good service in the off-peak periods, but also to cut cost to

optimise the business case.

Time period

Trains in central

section

30 (base)

05-06

06-09

09-15

15-19

19-01

Table 2: Frequencies during the day

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

5.3.2

Simulations

The scenario analyses have comprised technical analyses and passenger impact analysis.

For the technical analyses Parson’s PRIME tool has been used to calculate minimum running time,

assess travel times, crowding and delays taking into consideration the constraints defined, gaps

and necessary running time supplement, required in order to have a stable and regular operation.

Furthermore, the PRIME analyses show the required number of train sets in order to perform the

operation according to the timetable.

The demand in the form of the number of passengers has been calculated using the OTM. The OTM

is used for traffic model calculations in the greater Copenhagen area. 2030 is used as the

calculation year. All decided road- and rail projects are included in the model. The OTM

calculations are based upon the results of the PRIME calculations and simulations.

5.4

Description of scenarios

For the analysis each assessed scenario has been described in details, including its characteristics,

preconditions and constraints. A

complete description is included in the Appendix “Operational

Scenarios”.

In the following the base scenario, base scenario with UTO operation (UTO 30) and the preferred

classic (C.A.36) and metro (M.A.36) scenarios for a future UTO operated S-Bane are described.

Subsequently, the flexible scenario based upon partly metro style operation and partly classic

operation is described.

5.4.1

Base STO and UTO scenario

The base STO scenario and base UTO are based on the 2014 operational pattern with 30 trains per

hour through the central section. 12 trains per hour are operated on the Ring line. On the fingers

the following frequencies are used during the peak periods: 6 trains per hour for: A (inner

section), B, C (inner section), and E as well as 3 trains per hour for lines A (outer section), Bx

(morning only), H, and C (outer section). Running times, dwell times, and the rolling stock is

adjusted to 2030 situation. As depicted in Figure 1, the base timetable is a classic, heterogeneous

timetable with a mix trains stopping at all stations combined, with fast trains that skip some

stations. Compared to the 2014 situation, the signalling system is updated.

The base STO scenario and the base UTO follow the same operational concept. The difference

between the two is that the base UTO scenario is an unattended (driverless) version of the base

scenario (where a driver is needed). The shift to UTO has a positive effect on circulation times of

trains resulting in a more effective operation than without a driver. Thus, the base UTO requires

fewer train sets than the base STO scenario.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Figure 1: Stop pattern for the base STO scenario and the base UTO scenario

5.4.2

Alternative scenarios

In relation to the business case, it was found optimal to operate the S-Bane with half the

frequency on the outer sections of the fingers, as the number of passengers is substantially lower

on these parts of the network than on the inner sections. This is similar to

today’s operation. Of

the analysed scenarios, the short classic and short metro scenarios (C.A.36 and M.A.36) have

been selected as the best. These two scenarios performed best based on a cross sectional analysis,

i.e. operating economy, robustness against delays and passenger pr. train km.

Short Metro scenario M.A.36 and short Classic scenario C.A.36

The short metro and classis scenarios feature a combination of short and long lines. The short lines

serve the inner parts of the network with the highest demand. Thus, the frequency on the inner

and outer sections of the network is adapted to match the number of passengers. Both scenarios

operate with 36 trains per hour in the central section and 30 trains per hour on the Ring line. The

frequency on the outer sections is 6 trains per hour and 12 trains per hour on the the inner

sections (12 trains per hour to Høje Taastrup and 6 trains per hour to Klampenborg). The main

difference between the two scenarios is the origin and destination of lines and the operational

style, i.e. whether the long lines skip stations on the inner part of the network, or stops at all

stations. The two scenarios are depicted in Figure 2.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Figure 2: Stop patterns for the Short Metro scenario, M.A.36 (left), and the Short Classic scenario, C.A.36

(right)

5.4.2.1

Flexible scenario, Flex 36

Following the detailed analysis of the short classic and metro scenarios, C.A.36 and M.A.36, the

following is concluded:



The short metro scenario (M.A.36) generates the most passengers of the two, resulting in

higher passenger revenue



The short classic scenario (C.A.36) has a reduced need of rolling stock compared to the

metro scenario (due to a higher average speed) and thus lower investments costs in rolling

stock

The lower investment costs for the classic scenario compared with the higher passenger revenues

in the metro scenario results in an almost equal operating economy for the two scenarios. To

exploit the different strengths of the two scenarios, they have been combined to a single optimised

scenario named the flexible scenario.

The flexible scenario is based on the classic scenario, C.A.36 (depicted in Figure 2 (right)) with the

following modifications:



Line Bx is run in metro style, i.e. Bx trains stops at all stations

The fast train service, line E and H, stops at all stations outside the peak periods as shown

in Figure 2

The operation will be Classic style during peak periods, i.e. from 5

–

9 and 15

–

19, and Metro

style outside these periods. The additional 6 hourly trains compared to the base scenarios are

added as an extra B line between Høje Taastrup and Østerport. The Flex 36 scenario is depicted in

Figure 3.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Figure 3: Flex 36 scenario. Stop pattern in peak periods (left). Stop pattern in non-peak periods

(right)

As the need for rolling stock depends on the peak period requirements, the need for rolling stock is

kept low by operating a classic operation in the peak periods. Metro style operation is used outside

the peak periods to attract more passengers. Studies have shown that most commuters plan their

arrival at stations when the frequency is 10 minutes. Thus, commuters may benefit from a fast

train services. Outside the peak periods there are fewer commuters, and the passengers arrive

more randomly. As the frequency is evenly distributed with a metro style operation, it is more

suitable for passengers arriving randomly and thus attractive in the off peak periods.

5.5

Future development

–

passengers and other projects

The exact future passenger development is surrounded with some degree of uncertainty. It is

however very likely that an increase in passengers will lead to a demand of more capacity and

thereby an increased number of rolling stock. The increased capacity demand can at first be

handled by increasing the frequency from 36 trains pr. hour to 39 trains pr. hour and by coupling

of train units (section 6.7), but at some point it will be necessary to purchase extra train units to

maintain the same service level.

This issue (even therefore it is positive) is not linked in particular to the implementation of UTO on

the S-bane, and a similar issue will occur in a STO scenario. A potential risk in this scenario is

overpricing of the extra train units, caused by the fact that rolling stock supplier of the main

delivery will have a major advantage. A way to mitigate this scenario is include a later smaller

delivery in the contract with the rolling stock supplier.

A sensitivity analysis regarding the passenger numbers is included in the financial analysis (section

9.5).

Another key factor that will affect the final UTO operational concept is the development of the

infrastructure

–

e.g. implementation of a turn track at Enghave or similar. These improvements of

the infrastructure will be beneficial for the implementation of UTO and in turn it will improve the

flexibility of the S-bane.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

6.1

6.1.1

INFRASTRUCTURE AND ROLLING STOCK

Platform Safety

Introduction

Safety and punctuality are key issues for a successful operation of a railway. By a reorganisation of

the S-Bane to driverless operation, it must be ensured that the current level of safety is at least

maintained and the punctuality/reliability is not reduced. A cornerstone of a driverless system is

thereby the technical solution that can provide the required passenger safety without decreasing

the punctuality.

The investigation contains analysis of different technical safety systems in order to determine the

most preferable solution for the S-bane. Platform safety can in general either be handled by an

intrusion prevention system, or an intrusion detection system. Whereas an intrusion prevention

system (e.g. platform screen doors) segregates the track and platform areas, an intrusion

detection system detects objects in the track. The objects may be persons falling onto the tracks

or other large objects (items and animals) which prevent the train operation.

The analyses leading to the proposed solution is based upon an assessment of different solutions

and their safety level, reliability, maturity and cost.

The whole analysis regarding platform safety can be found in appendix 2.

6.1.2

Conclusion

For the S-Bane stations it will be sufficient

–

from a safety perspective

–

to install an intrusion

detection system in the form of an object detection system (ODS). The introduction of ODS will

actually improve the safety at the stations as the system will immediately detect and stop the train

operation if a person falls onto the track. The ODS has furthermore a significantly lower cost than

intrusion prevention systems and at the same provides the required reliability.

An ODS consists of electronic barriers and a camera system as shown in Figure 4 below. The

recommended solution for the S-bane

–

as a design choice

–

contains two infrared barriers, one

under the platform high above ground and one at a just above track level. In case one of the

barriers identifies an object, a message is sent to the control centre for further action. In order to

minimise traffic interruptions due to false alarms, cameras covering the platform train interface

with video analytics shall be positioned above the platforms with a view along the tracks.

Figure 4: Obstacle Detection System (ODS)

As per Figure 4 an ODS has been tested at the Stockholm metro where two station have been

equipped with different solutions for a period. This trial has demonstrated that the ODS can sense

and identify events of persons and items on the track correctly and at a level required for the S-

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Bane. By means of cameras with video analytics, the assessment can be improved significantly.

Furthermore, tests have indicated that video analytics can detect persons being trapped in train

doors.

The field trial also showed that the ODS shown an acceptable performance for outdoor stations in

different weather conditions. The assessment of the Stockholm metro field trial results are

furthermore supported from commercial operation in Nurnberg (where electronic barriers similar to

those installed in Stockholm) confirms the system robustness.

The experiences from the Copenhagen Metro

–

which converted from an ODS system to automatic

platform gates on the outdoor stations due to performance/reliability issues

–

have been part of

the assessment. The technology development within sensors combined with camera technology

makes it difficult to compare the two generations of ODS systems directly

. Which the field trail

from Stockholm also supports.

The recommended ODS is based upon standard commercially available components which

supports that the technology is mature. For final design of the ODS on the S-Bane, there will be a

need for further studies and adaptations in order to cover curved platforms, multi-track stations

and stations with platforms on the outer tracksides. These conditions are however not seen as

potential showstoppers for the ODS, but known and manageable technical issues.

The technical development within the field of ODS and similar safety solutions (e.g. advanced

driver assistance systems (ADAS) known from the car industry) is rapid changing technologies

–

improving system performance and reducing cost. The overall technical development is pointed

towards higher reliability of the systems due to more accurate measurements/higher component

sensitivity. The cost of the future ODS systems is of course dependent on the technical

development

–

e.g. the development pace of video analytics combined with behavioural algorithms

which can reduce the number of false alarms and improve the reaction time significantly. This

development could influence the technologies on which the ODS will be based and thereby reduce

its cost. It is therefore also important to let the market development be a key driver for the

decision of the final ODS solution.

The introduction of the ODS system will imply a new staff organisation in order to manage

disruptions. An event requiring staff intervention will cause delays and thus decrease the

punctuality. The staff shall be organised in a way which allows presence at any location of the

network within 10 minutes. This will limit the duration of a disruption and the socio economic costs

caused by increased waiting time for the passengers.

Even though an ODS based solution can provide the needed level of safety at the lowest cost, it

might be recommendable to install an intrusion prevention system at a few stations at the central

section in order to achieve a higher reliability of the S-bane. It is however recommended that the

ODS is first installed at all stations, and based on the in-service performance of the ODS, the need

for an intrusion prevention system at specific stations can be reviewed.

The assessment of different intrusion prevention system (e.g. platform screen doors and platform

gates) shows that the most promising solution for the S-bane is a Rope Vertical Platform Screen

Doors (VPSD) solution.

The ODS at the Copenhagen Metro was implemented in 2002-2003 and 2007 and developed around 1990 and 2000. The first roll-out

at the S-bane will take place in 2022.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Figure 5: Rope Vertical PSD at a platform

Rope VPSD is a relatively new technology which has been put into operation in Korea and Japan

and been tested at the Stockholm metro (outside environment). A Rope VPSD, see Figure 5, is a

wall made of rope (metal wires) with distance of app. 50 mm. The space between the ropes/wires

is closed with a flexible plastic material creating a closed surface, and thereby preventing people

from getting into contact with the train. The VPSD is established along the whole platform, fixed at

post every 8-12 meters and with at height of app. 1,75 meter. Once a train has stopped at the

platform the rope VPSD is rolled up along the train, similar to a blind in a window, thereby allowing

passengers to leave and enter the train. The rope VPSD is lowered again before train departure.

The rope VPSD is not yet a fully mature technology. It is a promising technology, which has

significantly lower costs than PSDs as known from the Copenhagen Metro. Furthermore, as the

technology is simpler than PSD it must be envisaged that maintenance costs will be lower.

The rope VPSD and associated costs are not part of the business case due to the fact that it is a

potential add-on if the ODS system performance needs to be improved at a few stations.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

The protection of the tracks can be established before the driverless operation is introduced, as it

does not rely on the driverless functionality. The field trial should be carried out timely to allow for

this.

6.3

6.3.1

Stepless Boarding

Introduction

Following the transition to driverless operation, it will not be possible for a driver to assist disabled

people (e.g. people in wheelchairs, etc.) entering or leaving the train as is the situation today. In

order to comply with existing legislation

, disabled persons shall be able to enter and leave the

driverless trains, accordingly a driverless/staff-free solution will be required.

The key issue at the S-bane is the gaps (vertical and horizontal) between the train and the

platform edge. An investigation has been carried out on an analysis of the magnitude of the issue

(the size of the gaps at all stations) and potential solutions.

The solution recommended is based on the requirements in the field (including BOStrab regulation

which is used for the Copenhagen Metro and in the EU regulation for person with reduces mobility,

TSI PRM) and that the service level requirement for disabled people should be at least the same as

today.

The whole analysis regarding stepless boarding can be found in appendix 3.

6.3.2

Conclusion

The analysis shows that the gaps are manageable with known solutions. The recommended

solution is:



Horizontal gaps are managed by an onboard gap filler

Vertical gaps are managed by construction of ramps on the platforms

The horizontal gap can be bridged by a gap filler on the train which can cover the maximum gap

between a train and a platform. Gap fillers are part of standard UTO trains and can bridge the gaps

on the S-bane which are between 88mm and 403mm without additional costs.

The vertical gap can be bridged through construction of two fixed ramps in each train direction on

the platforms. Ramps will be required on up to 85 of the 194 platforms where the gap is more than

50mm. A fixed ramp localised at one of the door positions will allow access to a train. It will not

require maintenance and not cause any increase in dwell time. Establishment of ramps on parts of

the platforms complies with TSI PRM.

For a few platforms the vertical gap is negative. According to TSI PRM this is allowed, but not

according to BOStrab. A solution to this issue shall be found in a dialogue with the National Safety

Authority.

The analysis of stepless boarding also reveals that the current manual solution is an outdated

solution seen from both a service and operational perspective. In fact, it could easily be in

contradiction with the relevant UN convention not to provide an automated solution once new

trains will be purchased. An improvement of the current situation

–

e.g. removing the vertical gab

by infrastructural adjustments as mentioned above

–

could then be required in any case and then

not a particular UTO related issue. Stepless boarding is however a

need-to-have

solution for a

driverless system and it is therefore included as part of the UTO project.

UN Convention on the Rights of Persons with Disabilities

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

6.4

6.4.1

Flexibility and Capacity Enhancing Infrastructure

Introduction

This section describes the infrastructure investigation. The focus for the infrastructure investigation

has been the infrastructure readiness going from STO to UTO, including an increase of operation

from 30 to 36 trains pr. hour. This focus investigation is made from a

need-to-have

perspective.

Beside the

need-to-have

investigations,

nice-to-have

investigations have also been made in this

part of the report. The operational concept for UTO will change the timetable from 30 trains/hour

to 36 trains/hour in the central section in the peak hours compared to today’s timetable.

Normally

when a train frequency is increased, the risk of failure and operation errors will also increase. The

investigation analyses is if these failures can be minimised by implementing flexibility-increasing

initiatives in the S-bane infrastructure. These investigations are from a

nice-to-have

perspective.

The whole analysis regarding Flexibility and Capacity Enhancing Infrastructure can be found in

appendix 5.

6.4.2

Conclusion

The most important conclusion is that the current S-bane infrastructure can handle the

implementation of UTO without need for any changes. The 36 trains/hour are possible through the

central section and the different fingers can also handle their frequency, inclusive turnaround

movements at the end stations.

The extra frequencies will create extra passengers, but the platforms will not need to be expanded

in order to handle the extra passengers. This is due to the extra frequencies will distribute the

passengers more evenly on the platforms over time. The only places on the S-train network were

the trains will have more passengers boarding and un-boarding pr. train are at some of the

stations on the fingers. These platforms have not met their capacity limit, whereby the platforms

can handle the extra passengers without a problem.

In UTO the Ring Line is scheduled to 30 trains/hour in each direction in the peak hours. With UTO

the trains will have a

1 minute scheduled turnaround time compared to 6 minutes for today’s

timetable. In theory, 30 trains/hour on the Ring Line should therefore be possible (giving 2

minutes turnaround time), and with the new Signalling Programme implemented, the signalling

system will also be able to handle 30 trains/hour. Ny Ellebjerg station will though be the restriction

for the Ring Line’s timetable

due to the placement of the crossovers at the station. To improve the

robustness at Ny Ellebjerg station for the Ring Line, the crossovers can be optimised by changing

them into a diamond crossing. This is a robustness improvement, but not necessary in order to

obtain 30 trains/hour for the Ring Line.

A reversing track at Carlsberg, Copenhagen Central or Flintholm station has also been investigated.

A scenario with one of these turnaround possibilities were deselected in the earlier phases of the

project due to the financial calculations.

With UTO implemented for the S-bane network, the timetable will be more robust due to the extra

flexibility of turning the trains around faster than the STO timetable. This can also be used in case

of a fall-back situation, where the trains can run easier with one track operation and thereby have

a positive effect during disruptions. The UTO timetable therefore has not a need for extra

crossovers. The recommendations are still to implement new crossovers at Hvidovre and Avedøre

stations, as this will make the timetable even more robust since it will result in the possibility of

crossing over to the opposite track on the S-bane network every 10 minute (20 minute pr.

direction).

These crossovers will benefit both the UTO operation and today’s operation.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

6.5

6.5.1

Power Supply

Introduction

This section describes the investigation related to power supply focusing on the additional need for

power supply in the different scenarios, and the development of power efficiency of future trains.

The whole analysis regarding Power supply can be found in appendix 9.

6.5.2

Conclusion

Banedanmark has earlier made calculations of operational scenarios comparable to the Classic and

the Metro UTO solutions resulting in the report “Effektanalyse af S-banen” (Analysis of power on

the S-bane). The following costs for upgrading the traction power system to the UTO project are

based on this data foundation.

Scenario

STO

–

Baseline

UTO

–

Baseline

UTO

–

Classic Style

UTO

–

Metro Style

UTO

–

Flexible

Number of trains

Cost

11.000.000 €

17.000.000 €

19.400.000 €

17.000.000€ – 19.400.000€

Table 3

–

Cost for implementing UTO

The costs for already planned/conducted upgrades have been taken out of the cost estimate.

There are significant costs for upgrading the traction power system when implementing UTO, both

for the Classic Style and for the Metro Style. Approximately 2/3 of the costs relates to upgrades

that should already have been done regardless of the UTO project. These upgrades are however

not planned/approved and are therefore included in the cost estimate due to the preconditions for

the investigation (only approved projects shall be taken into account).

The future power consumption of the rolling stock is difficult to predict as there are two opposite

trends in the market, pulling consumptions in opposite directions. The first trend is the

development for rolling stock to be more efficient and with lower weight, thereby causing lower

consumption for the propulsion system. The second trend is that there are more and heavier

auxiliary systems, such as air-condition and passenger services in the rolling stock, thereby

causing higher consumption for the auxiliary systems.

The trends will cause the peak consumptions to be reduced but the minimum consumption to be

increased. This will have a positive effect on the traction power system, as the peak consumptions

normally are the ones that define the capacity of the system.

Implementing new rolling stock to the system does not cause additional costs due to power

consumption. Even though the new rolling stock is assessed to be an advantage to the system, the

positive effect on the cost is assessed to be negligible.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

6.6

6.6.1

Driverless Rolling Stock

Introduction

Introducing driverless operation on the S-bane will require the purchase of new rolling stock, or

retrofitting of existing rolling stock. This study is based on the assumption that the S-trains will

need to be replaced as they reach the end of their lifetime (2026-2036)

This section summarizes the investigation of driverless rolling stock

–

the key elements of the

investigation are:



A proposal concerning the future train unit layout, including relevant design parameters

Equipment for driverless rolling stock and cost estimate

The whole analysis regarding driverless rolling stock can be found in appendix 6.

6.6.2

Conclusion

Train design

Both future STO and UTO rolling stock should be based on a standard vehicle platform. The

marked survey has confirmed that the future rolling stock for the S-bane

can be based on “metro

rolling stock”.

Based on passenger data from

OTM, it’s concluded that

3-car trains (55m long) which can be

coupled up to three times (max. train length: 165m) are the most viable solution for both UTO and

STO rolling stock.

Figure 7: Proposal for interior layout of 3-sectional train set

A three car unit of 55m length provides for optimal use of the available infrastructure as three

coupled units (9 cars) will utilise the full available platform length. Three carriage train sets are

within the scope of the main manufacturer’s standard products. Reference

cases are Metro Line 1

Panama (Alstom), Downtown Line Singapore (Bombardier), Metro Oslo (Siemens) and Metro

Madrid (AnsaldoBreda/ Hitachi Rail, CAF).

The concept provides:



Adequately high seat numbers for passengers travelling longer distances on the S-Bane

A slightly increased proportion of door width per train set, leading to an improved

boarding/alighting performance

A similar amount of multi-purpose space for mobility impaired, baby strollers, bikes, etc.

An equal spacing of doors along the train even when operating as multiple units

The above concept recommendations complies with the train sets which the rolling stock suppliers

can deliver. Further refinements will be required, but will be part of the rolling stock procurement

process.

According to the assessment in section 5.1.3.2 of appendix 3 taking into account the existing

infrastructure, an entrance height of 950 mm is recommended.

The lifetime is estimated to 30 years. The current 4

generation s-trains were delivered from 1996 to 2006

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Rolling stock with a maximum speed of 120 km/h in UTO mode can be offered by all major

manufacturers. Though 120 km/h is not common for metro platforms, major rolling stock

manufacturers have indicated that upgrading of the body structure and bogies will not lead to

major cost increases. For the structural design of a train, the classification of the operational

environment is decisive. For the S-bane system a categorisation as C-II (rail vehicles exclusively

designed for closed networks without interfaces to other traffic) according to EN 15227 is assumed

(and should also be strived for in any discussions with the approval authorities).

Equipment for driverless rolling stock and cost estimate

The study includes a market survey in order to determine any additional on-board equipment

requirements and cost of driverless rolling stock. The baseline for the market survey is in

compliance with current legal framework for driverless operation (BOStrab

and FoF

The market survey shows that the following additional equipment will be needed for a UTO train:



For the vehicle body: collision detection, derailment detection and gap fillers/detection

For the doors: “remote control” (disabling doors) from operations control centre (OCC),

“emergency release” and door control for automatic door control before

departure

For the infrastructure: Uninterrupted/continuous communication between OCC and trains

at all times, including voice, video and (optional) the ability to check that the infrastructure

is obstacle free. Alternatively, this can be achieved by operating the first train with a

sweeper train (a train in automatic mode, but with a steward in front with the emergency

control panel open and ready to push the emergency stop bottom) after a stand-still

period, e.g. during the night or due to a possession

Investment estimates for the required fleet is based on recent procurements for similar rolling

stock. Investment for UTO equipment is added. The investment cost is show in the following table.

Item

Cost for one 55 m STO train set

Based upon procurement of new train sets for the Berlin S-Bahn which have a

similar operating environment as the Copenhagen S-Bane

Additional cost for inclusion of the UTO functionality

Based upon data from the Nuremberg Metro and assessments due to increased train

length and higher requirements in relation to reliability and worse accessibility for

intervention

Total costs per UTO trainset

Table 4: investment estimates

Cost in DKK

DKK 44 mio

DKK 1,5 mio

DKK 45,5 mio

Fleet size

The following table shows the fleet sizes for the various scenarios. Size of reserve fleet,

maintenance and operational spare units, is equal to a proportion of 10% of the operationally

required fleet size.

Operational

required units

146 units

141 units

159 units

173 units

184 units

185 units

Maintenance

spare

13 units

12 units

14 units

15 units

16 units

Operational

spare

2 units

3 units

Total fleet

size

161 units

155 units

175 units

191 units

203 units

204 units

Base STO

Base UTO

Flexible +

10%

15%

Table 5: Fleet size

BOStrab

–

German regulations for the construction and operation of light railways (also applying to metros)

Fahren ohne Fahrzeugführer

–

operation without train drivers

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

The remaining 20% of service will require trains with a larger capacity due to 5% of the section

demand. If a concept without coupling is chosen, this will result in a scenario with a huge

overproduction of capacity, because all trains would be dimensioned to handle the maximum

demand.

As shown in the figure, a train set of 2 coupled units is sufficient for a majority exceeding 99% of

services to cover the demand in the flexible scenario +0%. In case of increasing passenger

numbers, it is possible at any time to operate with a full train set length of 3 units.

The cost of train sets without coupling facility in regular operations is assumed to be 5% lower

than for train sets with coupling facility. Higher efforts are required for design of platform safety

system if a system relying on intrusion detection is utilised.

The signalling system should be capable of coupling, as coupling takes place already today.

Automated coupling can be executed at all stations and in stabling areas.

The annual operational costs for train sets with coupling facility is assessed to be at least 30%

lower than for train sets without coupling facility on the S-Bane, as demand fluctuations between

lines and throughout the day can be better taken into account.

A 3-sectional train design will lead to a yearly saving

on 31% or 62 mio. € compared to a concept

without coupling for the flexible scenario (without any increase of the OTM-figures). When allowing

free bikes and subsequently increased train capacity for peak hour traffic, there will be a further

increase in savings due to coupling.

If coupling should prove to be especially challenging in terms of operational reliability in bad

weather periods (heavy winter conditions), it could be chosen not to uncouple the train sets (i.e.

run with long trains) causing a higher operating cost in minor periods. This approach is already

used by DSB today.

All manufacturers have indicated that the unattended coupling is something they either have

already delivered, or will deliver in the future. Siemens has delivered unattended coupling as a

feature for the Nuremberg unattended metro system.

The table below summarises pro for the two concepts (operation with or without coupling) on key

parameters.

Arguments pro short units with capability

to couple

Vehicle

procurement

Arguments pro longer units

without capability to couple



Enables procurement of only one type of train

unit for the whole system, also reducing the

need of different maintenance spares (both

parts and rolling stock) compared to the

procurement of different vehicle types



As demand fluctuates not only during the day

but also throughout the network, the

purchase of long unit trains would require

invest without the corresponding need



Optimization of train length in

regard to existing

infrastructure also for

Ringbane possible



Lower invest per capacity for

vehicle due to longer units and

no need of coupling equipment



Time for de-/coupling

extending circulation reduces

turnaround buffer which could

be used for delay

compensation

Service and

operation



Significant reduction in operating cost



Increasing service quality by running short

units several times instead of one long unit

running once - at comparable marginal costs



Increase of operation flexibility with only one

type of train unit



Increased capacity according to temporary

events causing higher service demand

–

e.g.



Combining and splitting adds

uncertainties to the system,

which have to be compensated

by reliable technical solutions

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

concerts, sports events, etc.



Rescue of broken/degraded trains

–

coupling

makes recue of degraded trains in the

network much easier thus it will be possible

to transport a degraded train to the workshop

by a functioning train



Higher ability to meet future demand

–

e.g. if

the rush hour expands (which is a general

tendency in metropolitan areas) it will be

possible to meet this demand by increasing

capacity enabled by coupling

Maintenance



A higher share of maintenance work, which

can be done during low peak periods at

daytime



Lower running performance per individual

unit extending fleet’s life span



Vehicle length will fit into the current

workshop capacity

Stabling

and operational workarounds



Current stabling capacity sufficient for new

fleet of coupleable units but not for a fleet

without coupling feature, additional invest

would be required

Table 6: Comparison of concepts with or without coupling

Taking the above into consideration the most advantageous solution for the S-Bane will be

coupling of the train sets.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

6.8

6.8.1

Depot and Workshop

Introduction

A procurement of new rolling stock for driverless operation on the S-bane could require an

adaptation to maintenance facilities and stabling areas. The current facilities are designed and

optimized for the layout of the current SA and SE units.

This section summarizes the investigations of identified adaptations.

The key elements of the investigation are:



A proposal concerning the stabling concept and assessment of required empty running

mileage

Assessment of current maintenance facilities

Estimation of future needs and development of a refurbishment concept for the workshops

The whole analysis regarding depots and workshops can be found in appendix 8.

6.8.2

Conclusion

Today the stabling facilities already have sufficient capacity for the future fleet size. However,

some dead running will be required. To reduce this, it may be prudent to construct some additional

stabling facilities. The economic benefits of doing so should be evaluated once the future operating

concept has been chosen.

While some adaptations to the workshops are required, these can be accommodated largely within

the footprints of the existing workshops.

Stabling areas

Referring to Appendix 6, the train unit length is assumed to be 56 m. In particular, smaller

stabling areas at the terminal stations seem to be designed for a train set consisting of two SA-

units. As the unit length for the proposed concept is one third this length, existing capacity can be

used very efficiently. Comparatively, low capacities can be found north of the central section, in

particular along the Farum- and Hillerød-fingers. At the terminal stations Holte, Buddinge,

Østerport and Solrød Strand, no stabling areas are available.

The total stabling capacity on the S-bane network would be sufficient for 240 train units of 56m

length. The working assumption for the elaboration of the stabling concept is that all stabling will

be equipped for UTO. Exceptions are the stabling tracks and workshop in Høje Taastrup and the

workshop in Hundige, where manual shunting would be required.

As the maximum fleet size including spares is 204 units in Scenario Flex 36 +15%, the need for

additional stabling capacity can be avoided if some deadheading and/or capacity enhancements at

the other end of the line are accepted.

Already in the Base STO scenario a significant amount of empty running is required. UTO-cases,

Base UTO and Flex 36 require fewer services without corresponding demand.

The table give an overview of extra running performance:

Scenario

Base STO

Base UTO

Flex 36

Flex 36 +5%

Operati

onal

required

train

units

Add. daily

running

performance

in train unit-km

Additional

train-km

per day

Annual operating

costs due to lack

of stabling

capacities

in mio. DKK

Difference in

annual costs to

Base STO-case

in mio. DKK

146

141

159

173

1.785

1.793

1.621

1.323

1.192

1.390

14,2

14,1

13,2

18,9

- 0,1

- 1,0

+ 4,7

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Flex 36 +10%

Flex 36 +15%

184

185

3.007

2.070

23,3

23,7

+ 9,1

+ 9,5

A reduction of the mentioned costs can be achieved when measures to extend the current stabling

capacity and the construction of new stabling areas are realized. Once the future operating

concept has been chosen, a detailed analysis should assess the most economically effective

actions.

Workshops

Two workshop sites near Høje Taastrup and Hundige stations provide maintenances services to the

S-tog fleet. Whereas Høje provides the majority of services including heavy maintenance, Hundige

is only equipped with facilities for light maintenance and exterior cleaning. As the new proposed

train layout differs in various points from the existing S-tog units (two axles per bogie instead of

one, two bogies per body section, increased section length, diverging train unit length), significant

changes to the configuration of work stands and equipment are expected.

Depending on the scenario, fleet sizes increases by 6 to 56 train units. Maximum fleet size of 191

units represents an expansion by 40 % referring to the number of train units.

It is assumed that today three work stands are kept for exterior cleaning, 9 for light maintenance

and 7 for heavy maintenance.

The new concept requires that on workshop tracks for heavy maintenance in Høje Taastrup and

tracks in Hundige, where today one unit is maintained, in future two units in-line will be treated.

For light maintenance facilities in Høje Taastrup, where today two units per track are processed,

the elaborated concept proposes to maintain three units at once.

UTO related equipment does not require additional track capacity.

The increased number of work stands without a direct access will mean that the average

occupancy per work stand could decrease. As far as possible, service and maintenance plans

should therefore be adapted to counteract bottlenecks.

After refurbishment, 13 working stands would be available for heavy maintenance. Two lots in

Hundige and 12 in Høje Taastrup are reserved for light maintenance. Graffiti removal and exterior

washing can be done for 5 units at once. In total, 27 units could be handled for maintenance

activities at once.

In total, investments of 17,3 mio. € resp.

128,7 mio. DKK are required for fitting of existing

structures to the new fleet. Extensive extensions of yard and buildings will not be required.

More detailed design will have to be carried out in future project phases to confirm, or adjust any

assumptions made for this assessment.

It is proposed to apply all modifications already for the Base STO-case, to exploit the potential of

existing structures.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

The overall roll-out plan shall be considered as a conceptual, but a robust plan. However, external

projects and political processes may affect the plan

–

e.g. postpone the starting point of the

project.

Overall approach

The proposed strategy includes the implementation of a pilot scheme using the Ring line first to

acquire experience on newly installed UTO systems and to minimize risks before rolling out the

new UTO systems on the rest of the S-bane. In order to use the Ring line as a valuable pilot

project

–

and a key risk mitigation

–

it has to timely placed so knowledge can be transferred to the

main project. It is therefore recommended, to place the pilot at the Ring line so knowledge from

testing and commissioning can be taken into account in the design phase of the main project.

Furthermore, the advantage to use the Ring line is that it is not interfering with the central section

and that any new timetable can be easily implemented. The roll-out on the Ring line will take place

from 2022-2026.

Once the Ring line has been converted to UTO, the following step-wise rollout is recommended:

1. Migrate Høje Taastrup

–

Klampenborg (2025-2028)

2. Repeat the migration steps on Farum

–

Frederiksund (2026-2030)

3. Repeat the migration steps on Hillerød

–

Køge (2026-2030)

The decision on which line to be migrated first should is based on the complexity of the line (start

with the most simple) and the consideration for permanent depot and workshop accessibility.

Roll-out strategy

The specific roll-out strategy for each line will take place in two main steps:

1. Preparing and testing of UTO infrastructure

This step includes upgrading of the signalling system from STO to UTO, installation of ODS

including new control-center, and fencing of the tracks. This step also includes testing of the

systems and their interfaces

2. Rolling out new UTO trains including testing

This step includes the roll-out of UTO trains on a given line. The first sub-step will include

testing of train-infrastructure interface during night/non-commercial operation. Hereafter,

when the test is completed, the new trains will be put into commercial operation as they are

delivered.

This implementation approach implies that the new UTO trains (operated in UTO) and existing S-

trains will be operated in mixed operation during the roll-out. This reduces the need for drivers,

allowing smooth driver attrition and will reduce cost for Rolling stock (no temporary driver cab).

This type of migration

–

mixed operation

–

has been implemented on a range of projects e.g. the

RATP Paris Line 1 and Nuremberg.

The roll-out period will furthermore be used to gather data from the ODS system in order to

determine if Rope VPSD are needed at some stations in order to obtain the required service level.

Procurement strategies

–

Time Plan

There are three types of procurements to be planned:

1. Contract change to S-Bane Siemens CBTC contract (with Banedanmark) for the upgrade of

their CBTC system to UTO. This contract change can be prepared in advance and negotiated

well ahead of the 2022 deadline for starting the roll-out of UTO on the F-line. We estimate that

this contract change could be negotiated over a period of 6 to 12 month maximum

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

2. Procurements of all supporting infrastructure UTO systems: ODS, fencing, power, adjustments

of platforms (stepless boarding), etc. This procurement can also start well ahead of the 2022

deadline for starting the roll-out UTO on the F-line. It is estimate that this procurement would

last 1 year

3. Procurement of new UTO rolling stock. The tender phase for procuring new trains typically

requires 2 years from the start of writing the specifications and contract award. The

procurement should include a base order for buying new trains for the F-line with an option to

be called for buying the rest of the rolling stock to be rolled out on the entire system. Equal

prices for new Onboard CBTC system will be ensured by Banedanmark who will execute a

contract change to the S-bane CBTC contract with Siemens

Safety approval

One of the benefits of implementing the new UTO system on the F-line is to try out the safety

approval process for:



Upgrading the CBTC system from STO to UTO



Introducing new rolling stock in UTO mixed with STO trains



Introducing ODS and ADAS system

The main risk that is foreseen is to obtain an approval for the operating rules for mixed traffic

(STO/UTO).

Approval of UTO rolling stock is not particularly risky as it has been done before for the Metro.

Impact on the business case

The proposed rollout strategy offers several economic advantages for the business case:



Direct introduction of the new train in UTO. This allows to avoid having a temporary cab for

drivers (additional asset cost) and to start early realizing the savings by removing the

drivers from operation



By rolling out and proving the new UTO system on the F-line, this will de-risk the whole

project and reduce the risk of additional development, delayed safety approvals, etc.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

exception being when trains are moved to maintenance areas, where the movement of trains will

be manual. Thus, the current 63 FTE’s are reduced to 16.

Smaller changes will happen to the remaining part of the observed population.

7.3.2 Transition period

The roll-out phase is set to be 2024-2030 as defined in chapter [6.9], and with the driverless

operations being fully functional from 2030. Thus, staff reductions are assumed to take place

gradually in the years 2024-29, following the pace for introduction of new material as shown in the

table below.

2024

2025

2026

15%

2027

15%

2028

30%

2029

30%

2030

Table 8 - Expected roll-out pattern from 2026-2030

Distributed into actual numbers and staff categories, the adjustments are expected to follow the

pattern illustrated in the table below.

Year

% roll-out

Shunting personnel

Supervisors - train drivers

Supervisors - train stewards

Coordinators

Maintenance personnel

Train drivers

Train stewards

Total personnel change due to UTO

2024

-2

-1

-26

2025

-2

-1

-26

2026

15%

-7

-2

-1

-79

2027

15%

-7

-2

-1

-79

2028

30%

-14

-3

-2

-158

-157

2029

30%

-14

-3

-2

-158

-157

2030

Total

-47

-11

-5

-526

-524

Table 9 - expected personnel change due to UTO, 2024-2030

Based on the employee adjustments described above, all savings are expected to be achieved in

2030. The expected realization of savings for each year is shown in

Table 10

below.

Year

% roll-out

Savings from staff reductions

Indirect savings

Total savings

2024

2025

2026

2027

2028

2029

15%

30%

13.480.407 13.480.407 40.441.220 40.441.220 80.882.439 80.882.439

2.874.401 2.874.401 8.623.204 8.623.204 17.246.407 17.246.407

16.354.808 16.354.808 49.064.423 49.064.423 98.128.846 98.128.846

Table 10 - Expected realization of savings

2030

Total

269.608.130

57.488.025

327.096.155

7.4

Natural staff reduction

A key question of interest for the business case will be estimates around the transformational staff

profile from current to future state. For that reason, the natural staff reduction defined as the total

sum of retirement and resignation (e.g. new job) has been estimated. Equally, as the cost and

conditions for employees differs according to different employment terms the figures are divided

into regular employees and public servants as illustrated in the table below.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

From 2016-2024, it is estimated that 51 civil servants will retire and be replaced with train drivers

on regular employment conditions. From 2024-2029 another 42 will retire, but these will not be

replaced. This leaves a remainder of 57 out of 150 civil servants.

7.4.2 Handling of excess employees

Although the substantial natural staff reduction, it will not be sufficient to absorb the entire work

force. As shown in Table 12, a total of 339 train drivers will be in excess and will have to be

terminated or employed elsewhere in the organization. Out of these, 57 are estimated to be civil

servants, of which many are expected to retire in the following years. This number may in reality

be lower, assuming that some train drivers may retire before the age of 65.

Needed employee reduction

Natural attrition

Excess employees

2026

-36

2026

-35

2026

-35

2027

-34

2028

158

-33

125

2029

158

-33

125

Total

526

-206

339

Note: in 2024 and 2025, the natural attrition is

higher than the needed employee reduction. Since this “deficit” of employees

cannot be transferred to the following years, the number of excess employees is set to be 0 rather than -10 and -9

respectively

Table 12 - Excess train drivers, 2024-2029

An opportunity to reduce the number of excess train drivers would be a transfer to regional or long

distance trains (fjernbanen), although this would require an additional 6.5 month training period

It’s estimated that

around 1000 train drivers are employed within this area at DSB. Assuming

similar characteristics with the S-trains, the annual attrition is expected to be around 78 train

drivers, equalling a total number of 310 for the entire period, since no train drivers are transferred

in 2024 and 2025.

Based on the estimated numbers and experience from the signalling program, the most significant

challenge for DSB may then in reality be retention rather than handling of excess employees. With

339 train drivers in excess and an estimated absorption capacity from intercity and regional trains

of 310 as well as other opportunities for reemployment, there may be very few problems with

excess employees. Rather it would be important to ensure that enough train drivers stay on board

to secure operations until a full transition to UTO, since many may be expected to start seeking

opportunities elsewhere as soon as the plans are revealed.

It’s recommended that dedicated strategies and formulated initiatives supporting a smooth

organisational change management process are incorporated in a potential future transformation.

A 14 month training period is required for being a regional train driver. S-train drivers need a 6.5 month training upgrade (including

litra training) when making the transition from S-train drivers to regional train drivers

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

RISKS AND COST ESTIMATE

Any new infrastructure project must be based on a formalised risk assessment covering

identification of potential risks, assessment of their probability and impact which defines the

projet’s risk profile. It is part of the initial decision process and leads to an evaluation of the

project’s reliability. During the later preparation and implementations steps it is an associated

management process defining risk ownership, mitigation and follow-up activities continuously.

From an initial perspective the UTO does have a particular risk profile.

Many elements pull the entire project in a very promising direction: UTO is a well proven

operational mode, infrastructural detecting systems on platforms and in between station have

been successfully in operation for years, and several types of driverless rolling stock has been in

operation worldwide during the last decades. Only to add on top, that technical stability and

refinements are continuously improving each of these elements. Easier, cheaper, more well-

functioning and more reliable solutions have emerged during the last decade and improved

solutions compared to what mentioned in this report will definitely be available when UTO

potentially shall be implemented at the S-bane.

The other way around: UTO with the particular characteristics of the S-bane is not in function

anywhere. The size and complexity of the total S-bane, almost entirely above ground and high

speed compared too many rail metro systems are all well-known elements, however not in

combination. The general risk assessment must be related to this fact. In the following sections a

further analysis is conducted.

The whole analysis regarding risk assessment can be found in appendix 11.

8.1

Overall Assessment

As an introduction to a traditional event based risk assessment it is relevant to reflect on the

overall risk level associated with the project as a whole. Looking into this focusing on the

development and transformation, leaving the future operation excluded, two main sources of risk

shall be mentioned.

The first source is related to the fact that the feasibility of the project relies heavily on external

infrastructural factors outside the UTOs span of control: CBTC signalling technology upgrade

enabled by the Signalling Programme is required and a firm precondition for further development

into fully driverless operation. Equally, the ability not only to operate driverless, but also to

increase the frequency in the central section of the network will rely on the same external factor.

The UTO project relies on a future infrastructural state which neither a reality nor in control of the

UTO project to obtain.

The second main source is related to the

combination

of various well proven technologies and

standardised solutions, that, when brought together in context, might cause a risk per se due to

the increased complexity of the total system. All suggested solutions related to rolling stock,

platform edge safety, coupling, etc. are in commercial operation today. Also, driverless concept is

very well functioning in several above ground cases, required max. speed is standard delivered by

manufactures, etc. However, some of these elements will be combined in new patterns and

concepts that have not been tested elsewhere in a similar configuration. In that aspect driverless

operation at the S-bane does bring a new situation, which as circumstance will introduce a risk.

Based on these two risk sources,

it’s the assessment that the specific UTO project can be

associated to an overall risk level characterized as medium at the current state of preparations.

The likelihood of actual occurrence of events, that might have a substantial impact on ability to

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

deliver as planned shall not be underestimated or neglected. Thus, continued careful risk

assessment shall be carried out and cover all parts of the project throughout the continued project

formulation process. Mitigation of all identified events identified through this process shall be

incorporated in the project. This process shall continue during the subsequent tendering and

implementation phases.

On the basis of the project analysis carried out the risk assessment, it indicates an overall score of

35-50 in a range from zero to 100, thus the overall risk assessment for that reason can be

classified as medium. This assessment shall be seen in the light of the fact that much more

evidence about the mentioned external infrastructural development and its performance, i.e. the

proof of the CBTC concept, will be available before crucial UTO decision is required. This proof and

reliable information about the performance of the CBTC system will be based upon operational

experience and be in place way before decisive choices must be made in a potential UTO project.

8.2

Event based Risk overview

The distinct most substantial risks for implementation of UTO are described in further details

below.

Risk

Organisational resistance

Resistance from train drivers and

Unions becomes too strong and

that results in skilled personnel

may leave DSB due to uncertain

future and make train operation

unstable in transition period and

increased price due to mitigating

actions or unexpected negotiated

incentives

Procurement strategy

A procurement strategy which

does not reflect the project

context and objectives can lead

higher prices than expected

Mitigation



Well-balanced roll-out plan

based upon staff parameters

e.g. age, potential transfers

from S-bane to Fjernbane, etc.

Evaluation

Probability:

35-60%

Impact:

1-5%



The procurement strategy shall

ensure a “correct” division of

deliveries

–

e.g. separation of

the CBTC system (client

delivery) and the rolling stock

(supplier delivery)

Use the Ringbanen as a pilot

project including a lesson-learn

period in order to implement

the key findings for the rest of

the rolling stock procurement

Ensure separation of the two

projects

–

the potential

organisational change is

completed before or after the

technical change

Probability:

1-10%

Impact:

>25%

Procurement of rolling stock

The requirement specification

leads to over- or under-specified

which results in higher prices than

expected or a low quality product

Organisational and technical

changes

Parallel implementation of

organisational and technical

change projects can result in

project chaos

Complexity and interfaces

The combination of the following

three factors makes the S-bane



Probability:

1-10%

Impact:

>25%

Probability:

10-35%

Impact:

15-25%



Use the Ringbanen as a pilot

project including a lesson-learn

period in order to implement

Probability:

35-60%

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

unique:

1. The size of the S-bane

2. It is above the ground

3. The solution is driverless (UTO)

This complexity can cause events

to which there are no prior

experiences about, e.g. too costly

changes to existing infrastructure

“Brown field” project

The S-bane is not designed for a

driverless system

–

changing an

existing railway will cause

unforeseen challenges



the key findings for the rest of

the project

Impact:

15-25%

Involve stakeholders in project

formulation and preparation in

order to ensure compliance

with actual conditions and

minimising disruption of

ongoing operation

Continue collection of

technology experience

worldwide and assess them

Regular update of traffic

forecasts using

Ørestadmodellen

Follow the Signalling

Programme development and

raise concerns towards the

Signalling Programme in case

of timing, quality, performance

or functionality changes for the

CBTC system in order to ensure

that UTO project preconditions

will be fulfilled with respect to

the CBTC system

Probability:

35-60%

Impact:

5-15%

Future transportation needs

Uncertainty about future

technology and transportation

patterns make decided concepts

outdated



Probability:

1-10%

Impact:

1-5%

Delay or quality degradations

of the Signalling Programme

The UTO project relies on the

signalling programme. Additional

delays or quality degradation of

the signalling programme will

have directly impact on scope of

the UTO project

–

e.g. delays of

the signalling programme will in

case of further delays also create

delays of the UTO project.

Potential quality reductions of the

final STO signalling system will

cause a need for extension of the

scope of the UTO project in order

to obtain the necessary

functionalities

Automatically coupling

The automatically coupling does

not work as assumed



Probability:

10-35%

Impact:

5-15%



Use the Ringbanen as a pilot

project including a lesson-learn

period in order to implement

the key findings for the rest of

the project

If the difficulties are limited to

periods with bad weather

conditions it can be managed

with running full configuration

(coupled units)

Use the Ringbanen as a pilot

project including a lesson-learn

period in order to implement

Probability:

10-35%

Impact:

5-15%

Platform safety

The chosen platform safety

solution

–

ODS

–

does not work as



Probability:

1-10%

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

assumed

–

e.g. detects too many

false events (birds, papers, etc.)

which results in reduced reliability

or does not work in the outdoor

conditions



the key findings for the rest of

the project

Base the chosen solution on

proven concepts/international

experience from

implementation in comparable

environment

Use the Ringbanen as a pilot

project including a lesson-learn

period in order to implement

the key findings for the rest of

the project

Early involvement of the

approval authority throughout

the project

Use the Ringbanen as a pilot

project including a lesson-learn

period in order to implement

the key findings for the rest of

the project

Impact:

15-25%

Safety approval (CSM)

The safety approval process is

underestimated which results in

delays and cost overruns



Probability:

10-35%

Impact:

15-25%



Infrastructure works

The infrastructure works need to

be executed at night or during line

closures resulting in delays or cost

overruns

Power supply

Power supply upgrades are

postponed after political approval

of the UTO project which leads to

additional costs

Poor data basis

Poor infrastructure data leads to

underestimated sub-projects (e.g.

stepless boarding, fencing need,

etc.)

Stepless boarding

–

onboard

gap-fillers

The onboard gap-fillers cannot

stand the weather conditions and

daily use. This has been a risk in

other projects

Project implementation

The complexity of the project with

more parties involved, many

stakeholders and project

interfaces may cause unexpected

amendments or changes to the

project and call for more

resources on the Client’s

side.

This will endanger the project

costs and the time schedule



Probability:

35-60%

Impact:

1-5%

Probability:

10-35%

Impact:

1-5%



Sample check of available data

in order to assess quality of

data during project formulation

and preparation before project

approval

Clear specifications related to

the gap-fillers in the rolling

stock tender material

Testing in the right weather

conditions

–

also by supplier

Probability:

10-35%

Impact:

1-5%

Probability:

10-35%

Impact:

5-15%



Careful project formulation and

preparation involving all key

project parties and

stakeholders

Continued involvement of all

project parties and key

stakeholders during project

implementation after

contracting

Probability:

10-35%

Impact:

5-10%

The total risk add-on is calculated to approximately 780 million DKK which constitutes about 39%

of the budget for Flex36 (2 billion DKK).

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

The overall risk evaluation is shown on the figure below.

Figure 13: Risk evaluation

The most severely (red) risks are:



Complexity and interfaces (contributing with a risk add-on cost of 190 mio. DKK)

“Brown field” project (contributing with a risk add-on

cost of 95 mio. DKK)

Organisational and technical changes (contributing with a risk add-on cost of 90 mio. DKK)

Safety approval (contributing with a risk add-on cost of 90 mio. DKK)

Procurement strategy (contributing with a risk add-on cost of 82 mio. DKK)

These risks and the evaluations reflect that the project is first of its kind in Denmark (first mover

project). The Metro in Copenhagen was designed as a driverless system from the beginning.

Especially,

the “Complexity and interfaces” and “Brown field project” risks describe and capture the

challenge of transforming an existing railway system into a driverless system.

The most important mitigation

–

across a broad number of risks

–

is the use of Ringbanen as pilot

project. The main purpose of the pilot project is first of all to test the technical solution and make

sure it is feasible (adjust the scope

–

e.g. technical modifications) before it is implemented at the

rest of the S-bane.

The different scenarios have been taken into consideration by the risk evaluation. There are

however not identified major differences between the scenarios from a risk perspective:



The difference between “Classic style” and “Metro style” is basically

a time table question

and it does not affect the investments in infrastructure or rolling stock differently. Thus, it

does not influence the above risk assessment

The difference between “36” and “39” is likewise mostly a time table question –

however it

is highlighted that the 39 scenarios are close to technical limitations of the signalling

system and thus an operational risk related to the 39 scenarios. This operational risk has

been taken into consideration in the scenario selection and was one of the reasons to

deselect the 39 scenarios

The difference between UTO30 and Flex36 is primary the required number of rolling stock

units and time table. The risk assessment will however contain the same overall risks and

evaluations



TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

9.1

FINANCIAL ANALYSIS

Summary

The purpose of the financial analysis is to provide a comparison of the feasibility of each

alternative based on a consolidated examination of all costs and revenues related to the

investments in and operations of the new driverless trains. The analysis includes two different

alternatives to the baseline scenario, namely the Base UTO 30 and Flexible UTO 36.

The Base UTO 30 has a 2016 NPV of 1.311 mill. DKK, whilst the Flexible UTO 36 scenario results in

an NPV of 309 mill. DKK. Based on the NPV calculations, therefore, both scenarios are financially

beneficial compared to the baseline, although the former proves significantly stronger. This is

further backed up by the breakeven analysis, which shows that the Base UTO breaks even primo

2038, meaning that the project pays back within 18 years after the first project costs are held. The

Flex 36 scenario breaks even 18 months later in mid-2039, slightly later than the Base STO, which

breaks even ultimo 2038.

The main driver of the business case in favour of UTO is the savings of staff operating costs. The

transition to driverless operation saves 3.363 mill. DKK in NPV, equivalent to an annual saving of

327 mill. DKK after 2030. The difference in NPVs between the two UTO is roughly 1.000 mill. DKK,

and it is primarily driven by higher rolling stock investments, as well as higher train set operating

costs, the latter of which mainly covers train maintenance and energy consumption. Thus, the

larger passenger base and resulting 1.095 mill. DKK larger revenue base in the Flex 36 is

outweighed by the higher costs base.

The sensitivity analysis assesses the sensitivity of the UTO business case to changes in key

parameters; namely, passenger revenues, rolling stock investments, infrastructure investments

and savings on staff operating costs. The business case for the UTO scenario proved highly robust

to changes in the key parameters, whilst the Flex 36 show less leeway in terms of changes in the

size of key parameters. This scenario is especially sensitive to changes in the savings on staff

operating costs, as well as the price of new rolling stock.

Conclusively, there is a strong business case for introducing driverless operation in the S-train

network. However, a simultaneous increase in train frequency to 36 trains does not improve on

the business case, and the current frequency of 30 trains is therefore suggested as the best option

from a purely financial perspective. Confer Appendix 13 for more detailed information about the

calculations behind the financial analysis.

9.2

Introduction

The overall objective of this project is to investigate the financial costs and benefits of

implementing unattended train operation (UTO) on the S-bane compared to the current semi

unattended train operation (STO). The STO will be fully implemented after the implementation of

the signalling programme.

The financial analysis measures all costs and benefits in financial terms by addressing whether or

not UTO provides increased revenue (more passengers), reduced costs (CAPEX or OPEX

), or any

combination of these that brings a better financial result compared to the existing operational

setup. Specifically, the performance of the UTO alternatives compared to the baseline is

constrained based on preconditions that should satisfy the following: 1) At least same overall

safety level should be kept, 2) Equally special segments of passengers should be provided with

same service level, e.g. disabled people shall have at least the same level of access, despite the

lack of driver support. 3) Other service intensions, on the other hand, are less strict: No explicit

CAPEX = Capital expenses, OPEX = Operating expenses.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

requirements are associated to for example availability, reliability or comfort. Such parameters are

assumed to fulfil a sufficient level, as long as it does not damage the financial result.

In practice, the financial analysis provides a financial examination of the costs and revenues

related to the project of transforming the S-train network to unattended train operation. The

analysis is conducted in accordance with the principles in the Ministry of Transport, Building and

Housing’s

“Manual for samfundsøkonomisk analyse på transportområdet”

The results of the

financial analysis are based on figures and values provided by other working groups, as well as

underlying assumptions related hereto. The project calculation period spans from 2020, where the

first project costs are held, to 2053, where the lifetime of all capital investments end. The

following section provides a short discussion on the timeline of the project and the calculation and

timing of costs and benefits. For a more detailed description of the assumptions and methods

underlying the analysis, please refer to Appendix 13.

9.3

Project timeline and discount rate

According to TRM, the project timeline shall follow the lifetime of the acquired rolling stock

. This

is based on the fact that it does not seem feasible to extend the timeline beyond the lifetime of the

rolling stock, as the main interest lies with the analysis of the operations of the new rolling stock.

In the screening report from 2013, the lifetime of the new driverless rolling stock was estimated to

25-30 years

. This is consistent with the lifetime of the current rolling stock, which is 30 years (cf.

section 6.6 on rolling stock). As a result, a lifetime of

maximum

of 30 years is applied in the

financial analysis.

In practice, the project timeline consists of two phases; 1) a construction phase and 2) an

operational phase. The construction phase spans from 2020 to 2029, both years included.

Effectively, this phase begins when the first project costs are held in 2020 (cf. section 9.4.2 on

details about the project costs), and ends when the last UTO trains have been rolled out on the

lines Farum

–

Frederikssund and Hillerød

–

Køge. For details on the rollout of the new trains, see

section 6.9.

The operational phase spans from 2030 to 2053. Technically, however, the operational phase

overlaps the construction phase, as the first trains are rolled out on the Ring line already from

2024-2026, after which the operations hereof starts in 2027. The operations of the new UTO trains

on the rest of the lines begin in 2030, as the operations of hereof are rolled out later than on the

Ring line.

In accordance with the rollout plan in section 6.9, the first trains to operate on the Ring line are

acquired in 2024, and it is the lifetime of these trains, which determines the length of the project

timeline. As this is assumed to be maximum 30 years, the timeline of the project ends in 2053. In

practice, however, this means that trains acquired at a later date are yet to exhaust their lifetime

at the end of the calculation period. However, no terminal value is included at the end of the

period due to the fact that the average lifetime of new driverless rolling stock is defined to be

between 25 and 30 years. As a result, some trains may in fact have shorter effective lifetimes than

the 30 years assumed. This applies both to the project alternatives and the baseline scenario,

however. Thus, as the business case calculations are based on the difference (delta) between the

alternatives and the base scenario, on average there is no effect of this assumption on the overall

results of the analysis.

Transportministeriet (2015). Manual for samfundsøkonomisk metode på transportområdet.

Transportministeriet (2015). Manual for samfundsøkonomisk metode på transportområdet, p. 17.

Transportministeriet: Rapport om mulighederne for automatisk S-bane drift (2013).

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

The Ministry of Finance and the Ministry of Transport, Building and Housing sets the official

discount rate (the rate at which all future costs and benefits are discounted back to 2016 present

values (PV)) for projects not exceeding 35 years at

4 percent.

Hence a discount rate of 4

percent is applied in both the financial and socio-economic analyses.

9.4

Detailed results

Table 13 below displays the results in delta figures (Alternative

–

Base), i.e. as the differences

between Base UTO/Flex 36 and Base STO, respectively. The following paragraphs provide short

descriptions of each of the accounts displayed below, along with explanations as to why they differ

between the two alternatives.

NPV mill. DKK 2016

Base UTO

201

-140

-44

-671

-329

-39

-47

-1.086

-1.070

-134

3.363

-139

-137

-172

-501

-672

-810

-39

Flex 36

-521

-158

-77

-756

-671

-329

-55

-39

-47

-1.141

-1.898

-262

3.363

-198

-97

-1.033

-1.328

-172

-501

-662

-1.989

1.095

ALL ACCOUNTS

–

DELTA FIGURES

Procurement of new rolling stock

Additional costs for UTO functionality

Upgrade of the on-board unit (CBTC)

TOTAL OPERATIONAL RELATED CAPITAL

EXPENDITURE

Platform Edge Safety

Safety between stations

Traction Power

Track

Stepless boarding

Upgrade of maintenance facilities (workshops)

Upgrade of the signalling system (CBTC)

TOTAL INFRASTRUCTURE COSTS

TOTAL INVESTMENT CAPITAL EXPENDITURE (CAPEX)

PROJECT COSTS

SAVINGS ON STAFF OPERATING COSTS

Rail access fees

Insurance costs

Train set operating costs (maintenance and energy)

TOTAL OPERATIONAL COSTS

Safety between Stations

Platform Safety

Costs for stabling areas

TOTAL INFRASTRUCTURE MAINTENANCE COSTS

TOTAL OPERATION AND MAINTENANCE COSTS (OPEX)

TOTAL PASSENGER REVENUE

TOTAL OPERATING SURPLUS

TOTAL COSTS

TOTAL SAVINGS ON STAFF OPERATING COSTS

TOTAL REVENUE

NET RESULT

-2.014

3.363

-39

1.311

Table 13: Results of the financial analysis, 2016 NPV mill. DKK

-4.149

3.363

1.095

309

Transportministeriet (2015). Manual for samfundsøkonomisk metode på transportområdet

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

30 percent is there for used in a balance between practice and expert assessment. The early phase

of assessment and few historical very challenging projects argues for a high factor. At the other

hand, Rolling Stock does typically implies less risk imposing external stakeholder engagement

(noise, resistance, complains, legislations etc.) than huge construction- or installation projects.

Equally the rolling stock will based on standardized platforms wherever possible. From these

weighted considerations a 30 percent risk factor seems reasonable for Rolling Stock purchase

9.4.1.2

Operational related capital expenditures (rolling stock)

The operational related capital expenditures include all investments related to the acquisition of

new rolling stock, and they are held in accordance with the rollout period. They are divided in 1)

procurement of new rolling stock, 2) additional costs for UTO functionality and 3) upgrades of the

on-board signalling systems (CTBC).

The main cost driver of all three sub-accounts above is the number of new trains purchased; cf.

chapter 6.6 and appendix 6 for more information on rolling stock. More trains are purchased in the

flexible scenario, as the train frequency is increased to operations with 36 trains, which results in

higher costs. Furthermore, the additional costs for UTO functionality are driven by the price

difference between STO and UTO trains, as there are extra costs related to the UTO specific

functionality of new S-trains. The upgrades to the signalling systems (CTBC) cost 2 million DKK pr.

train in the UTO scenarios, and 1.6 million DKK pr. train in the STO scenario.

9.4.1.3

Infrastructure capital expenditures

The infrastructure capital expenditures include the following two groups of investments (notice

that the figures provided are excl. NAB

1) UTO specific investments:



Platform Edge Safety covers costs for safety systems such as IR-barriers at the stations. Total

investment is 652 mill. DKK throughout the rollout period in both UTO scenarios.



Safety between Stations includes a total investment of 320 mill. DKK in both UTO scenarios.



Stepless boarding accounts for investments of 38.25 mill. DKK for technology improvements to

enable stepless boarding solutions.



Upgrades of the signalling systems (CBTC) covers investments of 45.5 mill. DKK across the

rollout period to adapt the current signalling systems to UTO trains.

2) Necessary investments for both STO and UTO:



Traction Power upgrades includes investments to increase the capacity of the power supply in

the S-train network, as a result of the increase power demand from a higher number of

operating trains. Investments are identical in the Base STO and Base UTO scenarios at 82 mill.

DKK, but 65 percent higher in the Flex 36 scenario at 136 mill. DKK, as the power system

needs to generate power for a larger number of operating trains.



Track investments are identical across all scenarios and include total investments of 22 mill.

DKK.



Upgrades of maintenance facilities (workshops) are identical across scenarios and include total

investments of 129 mill. DKK across the rollout period. This account covers the costs for

upgrading the current workshops to be adaptable to servicing the new rolling stock.

9.4.2

Project costs

Project costs cover expenses for project management and administration, safety approvals, testing

and commissioning, mobilization prior to trial operation, as well as revenue operation. They are

calculated as 18 percent of the total CAPEX of rolling stock and infrastructure excl. of NAB. Figure

The net present values provided earlier in table 14 are inclusive of the NAB uncertainty correction.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

-800

-700

-600

Mio. DKK

-500

-400

-300

-200

-724

-614

-545

-100

Base STO

Base UTO

Flex 36

Figure 16: Operational and maintenance costs, mill. DKK

As Figure 16 above shows, the operating and maintenance costs are significantly higher in the Flex

36 scenario at an annual cost of 724 mill. DKK from 2030 to 2053.

9.4.3.1

Operational costs

The operational costs are divided into 1) rail access fees, 2) insurance costs, and 3) train set

operating costs.

1) Rail access fees account for the cost, which the train operator pays to Bane Danmark for

access to the infrastructure and rails. Based on “Netredegørelsen 2018”, the current access fee

(infrastructure charge) is calculated to 4.8 DKK pr. train km in 2016 prices.

Thus, the higher

the number of train kms driven, the higher will be the costs for access fees (cf. the larger costs

in the Flex 36 scenario of 198 mill. DKK).

2) Insurance costs (ansvar, kasko, forsikringsdækning af passagerer) are calculated based on

unit costs provided by Niras in

“Enhedsomkostninger ved persontogsdrift”.

Here, the unit cost

of insurance is estimated at 0.38 mill. DKK pr. S-train unit

. The unit costs in 2016 prices were

obtained by extrapolation using the Consumer Price Index (Forbrugerprisindekset). This

resulted in unit costs of 0.44 mill. DKK and 0.46 mill. DKK pr. STO and UTO train unit,

respectively. The difference in unit prices is based on the fact that the price of a new UTO train

is 3.4 percent higher than a new STO train (cf. appendix 13 for more details).

3) Train set operating costs account for the largest part of the operational costs, and they cover

costs for rolling stock maintenance and energy. They are calculated as 21.1 DKK pr. STO train

unit km, and 21.8 DKK pr. UTO train unit km. Operating costs are thus higher in the Flex 36

scenario, as they are driven by a larger number of train unit kms.

BaneDanmark “Netredegørelse 2018” can be found at

http://www.bane.dk/db/filarkiv/21502/Netredeg%F8relse%202018%20endelig%20version%20v2.pdf

and current infrastructure charges can be

found at

https://www.retsinformation.dk/forms/R0710.aspx?id=175591#id42ce8ed4-7bfd-4cf5-b0d3-ffe853f183b3

http://www.radikale.net/Files/379/niras-enhedsomkostninger-ved-togdrift-samlet-afrapportering-071025-endelig-1.pdf

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

9.4.3.2

Infrastructure maintenance costs

Maintenance costs for infrastructure are held in accordance with the rollout plan for the related

infrastructure investments, and they cover maintenance costs for platform edge safety, which

amounts to 1.047 mill. DKK and 2) safety between stations, which totals 360 mill. DKK. These are

UTO specific maintenance costs (as they relate to UTO specific investments), and they are identical

for both UTO scenarios. Finally, infrastructure maintenance costs include annual costs for stabling

areas, which amount to 14.2 mill. DKK for Base STO, 14.1 mill. DKK for Base UTO and 13.2 mill.

DKK for the Flex 36 scenario. See chapter 6 for more details.

9.4.4

Savings on staff operating costs

The savings on staff operating costs from introduced driverless operation accounts for by far the

largest benefit in the business case at an NPV of 3.363 mill. DKK in both UTO scenarios. Confer

chapter 7 “Organisation”

for a more detailed description of the consequences of UTO operation on

personnel.

Figure 17 below illustrates the annual savings on staff operating costs. Note that the annual

savings on staff costs is lower in 2030, as a so-called

“stay bonus” of

27 mill. DKK is included as a

lump sum in both UTO scenarios. The bonus covers the extra costs estimated for retaining relevant

train drivers until the end of the transition phase, as there is a risk that some personnel is not

willing to stay until the end of the rollout period, knowing that they will either need to find another

job or face reallocation.

350

300

250

Mio. DKK

200

150

100

UTO Staff operating savings

Figure 17: Savings on staff operating costs, mill. DKK

9.4.5

Passenger revenue

Passenger revenue is calculated by multiplying the number of passenger trips, according to OTM

data, by the estimated average tariff per trip. In the financial analysis, the average trip tariff is

calculated based on the average number of zone crossings pr. trip, in order to obtain the most

reliable estimate possible. The average number of zone crossings differs between the different

scenarios (Base UTO and Flex 36), as do therefore the calculated average tariffs. Cf. appendix 13

for detailed descriptions on the calculation methods.

The average tariffs pr. passenger trip are hence calculated as:

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report



Base STO (current situation): 12.88 DKK

Base UTO : 12.87 DKK

Flex 36: 12.52 DKK

The number of zone crossings is shorter in the flexible scenario, as the metro operation results in

shorter average trips, which lead to a lower average tariff.

The number of passenger trips is assumed to grow 0.5 percent annually from 2030-2053. This is

based on the forecasted population growth for the period 2030-2040 in the Greater Copenhagen

municipalities, in which S-trains currently operate. This growth rate follows the recommendations

provided by key stakeholders in the project, and is therefore deemed a reasonable estimate of

future passenger growth.

It was considered to apply a higher growth rate in the project, based on the average passenger

growth rate the last 10 years of 2.4 percent

. However, this past growth arguably mainly stems

from the decision to allow free bikes on S-trains in 2010

. Evidently, in the 10-year period prior to

the decision of free bikes allowance, average annual growth rate was 0.1 percent. This suggests

that applying the average growth since 2005 is overly optimistic, thus validating the assumption of

a 0.5 percent passenger growth rate from 2030 and onwards

As evident from figure 15 below, annual passenger revenues are roughly at the same level across

all three scenarios until 2030, after which the larger passenger base in the Flex 36 scenario leads

to higher passenger revenues compared to the Base STO and Base UTO scenarios.

2.500

2.000

Mio. DKK

1.500

1.000

500

Revenue - Base STO

Revenue - Base UTO

Revenue - Flex 36

Figure 18: Annual passenger revenues, mill. DKK

For a detailed description of the underlying calculations, please see appendix 13.

The 2.4 percent average growth is calculated as the Compound Annual Growth Rate (CAGR) for the number of S-train passengers in

https://www.dsb.dk/Om-DSB/Presse/Nyheder/Tag-cyklen-gratis-med-i-S-toget/

Additionally, one may consider the fact that increased passenger demand in the S-train network (from the 0.5 percent annual

the period 2005-2015. Source: Statistics Denmark

increase) may at some point in the future result in overcrowding and thus may create the need for additional rolling stock acquisitions.

However, as this would be true for the Base STO scenario as well, the delta effect of this consideration is deemed insignificant and

therefore excluded from the analysis. Additionally, the amount of overcrowding greatly depends on the distribution of these passenger

increases across the day; an increase in non-peak hours is thus less likely to result in overcrowding, compared to an increase in peak-

hours. Based on these considerations, therefore, the 0.5 percent estimate involves some underlying uncertainty. This is acknowledged

and addressed in the sensitivity analysis in section 9.6.1.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

As a result of the combination of classic operation during peak hours and metro operation during

non-peak hours, the flexible scenario offers more passenger trips compared to both the Base STO

and Base UTO. This translates into higher passenger revenues even though the average tariff pr.

trip is smaller in Flex 36.

9.5

Break Even Analysis and Payback periods

Figure 19 below illustrates the accumulated net cash flows, or the liquidity development, across

scenarios throughout the analysis period. Base UTO breaks even in primo 2038 and hence has the

shortest payback time at 18 years after the first project costs is held in 2020. The flexible scenario

pays back 18 months later in mid-2039, whilst Base STO breaks even in ultimo-2038. The

estimated payback periods are consistent with the screening report from 2013, in which the

payback period was estimated at 19 years.

25000

Accumulated net result (cash

flows), mio. DKK

20000

15000

10000

5000

-5000

-10000

-15000

Base STO

Base UTO

Flex 36

Figure 19: Accumulated net cash flows (net result), mill. DKK

The break-even times of the three scenarios thus fall within a short interval. Additionally, it is

interesting to notice that the flexible scenario only outperforms the base STO after 2042.

Conversely, the base UTO already outperforms the baseline after 2036, based on accumulated

cash flows. This is directly related to the large extra investments made at the beginning of the

period in the flexible scenario. As is evident from the fact that the Base UTO has the highest NPV

(cf. 7) and the shortest payback period, the flexible scenario never makes up for the extra

investments related to increasing the frequency to 36 trains.

9.6

Sensitivity analysis

The purpose of the sensitivity analysis is to examine how the results are affected when key

parameters are changed. In the financial analysis, the following four parameters are key

determinants for the overall results, and they will be the subject of the following sensitivities.

Passenger revenue

Operational related CAPEX (rolling stock investments)

Infrastructure CAPEX

Savings on staff operations

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

9.6.1

Passenger revenue

The calculation of passenger revenue is based on traffic simulations on the number of passengers

from the OTM-model, and they are as such not the subject in a financial sensitivity analysis.

Instead, the assumed passenger growth rate is the subject of interest.

It is assumed that the passenger base will increase by 0.5 percent per year from 2030-2053. The

growth rate is an important parameter in the financial analysis, and it may be argued that letting

the number of passengers increase linearly over the period and simultaneously keeping the

number of operating trains constant is in contrast with reality. Hence, it may be contested that the

capacity of the trains acquired during the rollout period is sufficient in relation to the assumed

future passenger growth. Testing this assumption is therefore of high importance. The table below

shows the how the result of the business case changes when the passenger growth rate is altered.

Delta NPVs, million DKK 2016 prices

Base UTO

1.312

1.311

1.309

1.277

Flex 36

257

309

365

1.277

Sensitivity scenarios,

annual growth rates

Worst case: 0 percent

Current case: 0.5 percent

Best case: 1 percent

Converging rate: 6.3 percent

Table 14: Passenger growth rate sensitivities

The analysis suggests that the results of the business case are robust to changes in the passenger

growth rate. Both UTO scenarios are financially viable compared to the base case at zero growth in

passengers. Furthermore, the analysis illustrates that the two scenarios converge when the growth

in passengers is very high at 6.3 percent, which in reality seems quite unrealistic. This supports

the conclusion that the Base UTO scenario is the strongest, and equally importantly that the

transition to driverless operation in the S-train network is in fact financially viable.

9.6.2

Operational related capital expenditure (rolling stock investments)

The size of the rolling stock investments is a key factor in the business case, especially evident

from the large cost in the Flex 36 alternative, where a higher number of trains are purchased.

Although the uncertainty related to the operational related capital expenditures is already partially

accounted for in the analysis through the 30 percent risk add-on, a sensitivity analysis is still

deemed highly informative.

The table below shows the sensitivity of the business case to changes in the operational related

capital expenditure in both the baseline and the project alternatives. It distinguishes between

changes to the estimated unit price of a new rolling stock (cf. section 6.9), as well as changes to

the quantity of new trains purchased.

Sensitivity scenarios,

rolling stock capital

expenditure

Current price and quantity

- Price sensitivity:

- 10 percent higher price

Delta NPVs, million DKK 2016 prices

Base UTO

1.311

Flex 36

309

1.317

231

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

- 20 percent higher price

Quantity sensitivity:

- 10 percent more trains

- 20 percent more trains

1.324

152

1.313

1.315

Table 15: Operational related CAPEX sensitivities

212

105

The table shows that both alternatives still have a positive NPV when the price and number of

rolling stock is changed. Furthermore the table shows that the Base UTO has a higher NPV than

that Flex 36 in all cases.

9.6.3

Infrastructure capital expenditures

Changes to investments that are identical across all three scenarios are not relevant to test, as

they will affect the three scenarios in the same way and hence be irrelevant.

Therefore, in this sensitivity analysis, only infrastructure investments that differ in size between

the Base STO and the UTO scenarios have been altered. The table below shows the results of the

sensitivity analysis.

Delta NPVs, million DKK 2016 prices

Base UTO

1.311

1.188

1.066

Table 16: Infrastructure CAPEX sensitivities

Sensitivity scenarios,

infrastructure CAPEX

Current level

- 10 percent increase

- 20 percent increase

Flex 36

309

181

As the table show, changing the level of infrastructure investments does not alter the overall

conclusion; there is a positive business case for the introduction of UTO in the S-train network,

and the Base UTO still outperforms the Flex 36.

Looking at the specific numbers, a 10 percent increase in investment costs results in a decrease in

NPV of 123 mill. DKK for Base UTO, whilst the NPV of the Flex 36 decreases by 128 mill. DKK. The

small difference stems from the fact that the only investment costs, which differ between the two

UTO scenarios are the costs for upgrading the Traction Power System. This cost is initially 82 mill.

DKK and 136 mill. DKK in the Base UTO and Flex 36, respectively. All other infrastructure

investments are the same between the two project alternatives, cf. section 9.4.1.3.

At a 20 percent increase, the Flex 36 alternative results in an NPV of 52 mill. DKK, while the Base

UTO still significantly outperforms the Base STO at an NPV of 1.066 mill. DKK.

9.6.4

Savings on staff operating costs

The savings on staff operating costs is the main driver of the positive business case for driverless

operation. For this reason, this probably prevails as the most significant parameter to test in a

sensitivity analysis. The table below shows the effect of altering the annual savings on staff

operating costs, which was initially estimated at 327 mill. DKK, contributing with an NPV of 3.3 bill.

DKK in both UTO scenarios.

Sensitivity scenarios,

Delta NPVs, million DKK 2016 prices

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

savings on staff costs

Current level

- 7.05 percent decrease

- 10 percent decrease

- 20 percent decrease

- 30 percent decrease

- 38.78 percent decrease

Base UTO

1.311

1.001

973

635

297

Flex 36

309

-29

-367

-704

-1.001

Table 17: Savings on operating staff costs sensitivities

Each 10 percent decrease in the annual savings on staff operating costs amount to a decrease in

the NPV of both scenarios of 338 mill. DKK. This effectively means that the Flex 36 already

becomes unprofitable compared to the Base STO at a decrease in annual staff savings of about 7

percent, whilst it takes a significant decrease of 38.78 percent before the Base UTO converges to

the level of the Base STO result.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

10. SOCIO ECONOMIC ANALYSIS

10.1 Introduction

The socio-economic analysis describes the costs and benefits for the Danish society when replacing

the existing S-trains with driverless trains. The analysis is conducted using the principles from the

manual for socio-economic analysis within transportation

published by the Danish Ministry of

Transport, Building and Housing. The calculations are performed using the TERESA model

administered by same authority.

The socio-economic analysis is based on the assumptions provided by the other working groups as

well as the financial analysis conducted by working group 3. This includes fundamental parameters

such as project investments, operational and maintenance costs. User benefits including time and

distance savings are based on the results from the OTM- and PRIME-models. Furthermore, the

derived effects (for example externalities and tax distortions) as consequences of the changed

consumption patterns are included in the analysis as well.

10.2 Conclusion and results

Table 18 shows the socio-economic results divided in categories project investments, operational

and maintenance costs, user effects, external effects and other effects.

Entry

Project investments

Operational and maintenance costs

User benefits

Externalities

Other effects

Total net present value

Internal rate of return

Benefit-cost ratio

NPV per DKK project investment

Alternative 1

–

Base UTO

-1.595

2.235

-295

-8

257

594

6,1%

1,22

0,37

Alternative 2

–

Flex36

-2.861

1.595

4.746

-180

-185

3.114

9,6%

1,67

1,09

Table 18: Net present value, 2016 price level in market prices (mDKK)

Both alternatives have a positive net present value indicating that both alternatives are profitable

in a socio-economic perspective, cf. Table 18. However, alternative Flex36 has the most favorable

socio-economic result when comparing the two alternatives. The difference between the two

alternatives is primarily due to the large user benefits generated by the Flex36 alternative, even

though both operational costs and investments are significantly higher.

The internal rate of return and the Cost-benefit ratio resolve in the same conclusion for both

alternatives. Hence for both alternatives, the internal rate of return is higher than 4 pct. which is

expected for this type of project. Furthermore, for both alternatives the benefit-cost ratio is higher

than

1. The ‘NPV per DKK project investment’ is calculated as ‘total net present value’ divided by

the absolute value of ‘project investments’ and indicates that

for each DKK spend on project

investment the state can expect a positive net present return in either alternative.

In the following sections the different cost categories are described in more detail. For a

comprehensive explanation of each cost category the reader is referred to the appendix 13

describing the socio-economic analysis in more detail.

”Manual

for samfundsøkonomisk analyse på transportområdet”

fra Transportministeriet

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

10.2.1

Project investments (CAPEX)

The investment costs included in the socio-economic analysis is based on the investments

described in the financial analysis. Since the investments are established in factor prices a tax

factor

(henceforth NAF) is added to the price to convert the prices into market prices. The NAF

factor is assumed to be 1,325 which correspond to the assumptions in the manual for socio-

economic analysis within transportation. NAF can be seen as an estimate of the average tax

burden for private expenditures and is used to make private, public and citizen expenditures

comparable.

As the lifespan of the conducted investments, including rolling stock, are equal to the length of the

analysis no terminal values are included in the socio-economic analysis. This corresponds to the

assumptions in the financial analysis.

Table 19 gives an overview of the included investments in capital separated into subgroups. The

results are after including the NAF tax factor.

Entry

Operational related investments (rolling stock)

Investment in infrastructure costs

Project administration costs

Total net project investments

Alternative 1

–

Base UTO

-1.439

-178

-1.595

Alternative 2

–

Flex36

-1.002

-1.512

-347

-2.861

Table 19: Project investments (NPV in mDKK)

During the constructing period certain inconveniences due to noise, dust, emissions and other

disturbances are expected to occur. These inconveniences can be expected to affect individuals in

close proximity to the construction area. None of these inconveniences caused in the construction

period are included in the analysis as these are hard to quantify and are considered to have minor

impact. This results in a very minor overestimation of the benefits of the project.

10.2.2 Operational and maintenance costs (OPEX)

The operational and maintenance costs of the project are based on the financial analysis. Some of

the operational costs results from acquisition of product and services which are estimated in factor

prices. These costs are converted into market prices by the use of the NAF tax factor as specified

by the manual for socio-economic analysis within transportation.

Passenger revenue has been calculated using the OTM-model based on the train timetable and

expected number of passengers within the TERESA-model to take into consideration the loss in

revenues from other public transportation modes. Some train passengers are expected to change

from bus to train and from a socio-economic view this is simply a reallocation between public

transportation modes. This is different from the financial analysis since the socio-economic

analysis takes the society as a whole into consideration. The financial analysis directs its attention

to the business case of the train operation whereas the socio-economic analysis looks at a broader

picture.

Table 20 shows the included operational and maintenance costs after adding the NAF tax where

applicable.

Nettoafgiftsfaktor (NAF) jf. ”Manual

for samfundsøkonomisk analyse på transportområdet”

fra Transportministeriet

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Entry

Operational costs

Infrastructure maintenance costs

Passenger revenue

Staff operating costs

Total net operational and maintenance costs

Alternative 1

–

Base UTO

-182

-890

-40

3.348

2.235

Alternative 2

–

Flex36

-1.759

-877

882

3.348

1.595

Table 20: Operational and maintenance costs (NPV in mDKK)

10.2.3 User benefits

When implementing either of the two alternatives the transportation consumption pattern changes

which result in either a benefit or a cost for the individual consumer. The user benefits or costs are

based on the time and distance change determined by the OTM- and PRIME-model results

. The

OTM and PRIME results describe how much more/less time passengers spend in either public or

private traffic and how many more/less kilometers they drive in private transportation. These

changes are valued through the use of transportation unit prices as defined by the Ministry of

Transport, Building and Housing and their unit price catalog

. A unit price thereby indicates a

value for society in relation to individuals saving or spending more time in traffic. E.g. when an

individual saves one leisure hour in public transportation this generates a value of approximately

85 DKK. Had this hour been saved in a working relation the monetary value would have been

approximately 406 DKK.

The analysis includes changes in how individuals use public transportation and road transportation

and estimate the monetary value incurred compared to the base scenario. If individuals spend less

time in traffic or drive fewer kilometers when an alternative is implemented then this count as a

positive value from a socio-economic perspective. The opposite apply when more kilometers are

driven and more time is spent in traffic.

Public transportation includes bus and train transportation. The value of time savings are

separated into four categories for public transportation: Traveling time saved, waiting time saved,

time saved traveling to and from mode of transportation and time saved changing mode of

transportation. Table 21

indicates that most of the ‘benefits’ occur from time savings in public

transportation. The TERESA-model does not differ between modes of transportation within public

transportation. This means that the result is only highlighted as one source of benefit and not

separated into bus and train transportation

. From Table 21 it can be seen that the Base UTO

alternative result in more time spend in public transportation which result in a negative socio-

economic value. On the other hand, Flex36 creates large time savings within public transportation

which results in a large positive value for society as a whole.

Time and distance changes for road transportation are separated into three types: Cars,

commercial vehicles and trucks. From Table 21 it can be seen that Base UTO result in more time

being spend in road transportation and more kilometres are driven resulting in a negative socio-

economic value. The opposite apply for the Flex36 alternative where hours are spend in road

traffic and less kilometres are driven.

Table 21 shows that the individual passenger saves money in ticket expenditures when comparing

the base case with the two project alternatives. This is due to the fact that each individual

generally travels a shorter distance and thereby travels through fewer zones on an average. As

See the appendix section 11 for a review of the OTM- and PRIME-model results

Transportministeriets nøgletalskatalog

From Appendix 1 section 6.1 we know that Flex 36 provides 11% more passenger trips per weekday compared to the base scenarios.

Approximate 10.000 of these trips originate from trips that were previously done by car, foot or bike. 3 500 trips are new and the

remaining 26 000 are trips that were previously taken by bus, metro or other public transport modes.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

ticket prices are determined through the number of zones the passenger travels through this

means that the lower distance results in lower ticket expenditures in general. This means that

even though ticket revenues are rising in business case as more individuals are travelling by public

transportation each individual are saving money on public transportation.

As the project is implemented in steps the user benefits are as well. In the phase-in period only

part of the benefit effect is initiated. A phase-in period is assumed for each line opening. The first

line

–

the Ring Line

–

initiates at 2026 with 40 pct. of the potential passenger benefit for this line

and is increased each year in the sequence 70, 90 and 100 pct. in 2027, 2028 and 2029

respectively. The initiation

of the “fingers” uses the same percentages with start in 2030. Hence,

the phase-in period resembles the assumed phase-in period for the passengers in the financial

analysis.

The user benefits from time and distance changes for each category can be seen from Table 21.

Entry

Time savings, road

Time savings, public transportation

Time savings, freight transportation

Distance savings, road

Ticket expenditures, public transportation

Total net present value (NPV)

Alternative 1

–

Base UTO

-61

-205

-34

-295

Alternative 2

–

Flex36

552

4.165

-12

4.746

Table 21: User benefits (NPV in mDKK)

The negative result of Alternative 1 indicates that consumers of transportation spend more time in

traffic and drives a longer distance compared to the base scenario. For Alternative 2 the

consumers experience a time saving but at the same time experience a small increase in the

distance travelled. In total, alternative 2 implies large positive benefits for the consumers.

10.2.4 External effects

Externalities include the effects to the society that the individual traveller does not take into

consideration when choosing which mode of transportation to use. Individuals are not expected to

consider the noise or air pollution generated when considering their favourite travel plans. The

external effects can either appear as benefits or costs depending on how the consumption pattern

changes. All externalities are determined by the amount of kilometres driven in each alternative.

This means that if more kilometres are driven more externalities occur. When more kilometres are

travelled externalities are occurring causing negative influence to society and when less is

travelled externalities influence society positively.

The socio-economic analysis includes changes in externalities such as climate, air pollution, noise

and accidents that arise due to the changes in travel consumption patterns. When more kilometers

are driven more accidents, pollution and noise is expected to occur.

‘Air pollution’ includes

emission of NOx, HC, SO

, CO and PM2.5 whereas ‘climate’ only includes emissions of CO

. The

inconvenience due to noise is also estimated using unit costs. Lastly, the number of accidents

depends upon the kilometres driven and the costs to society increase as number of accidents

increase. An extended description of externalities and how they are determined can be found in

appendix section 13.

The benefits or costs of externalities are valued using unit prices as a function of train and car

kilometers driven in each alternative compared to the basis scenario. E.g. for each additional

kilometer driven by individuals due to implementation of the project society is expected to lose

value as more accidents, noise and pollution is expected to occur. Table 22 shows the net present

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

value of the externalities in the two alternatives. As both alternatives result in net more kilometers

driven both alternatives have negative externalities.

Entry

Alternative 1

–

Alternative 2

–

Base UTO

Flex36

Accidents

-5

-128

Noise

-1

-13

Air pollution

-1

-48

Climate (CO2)

-1

Total net present value (NPV)

Table 22: Externalities (NPV in mDKK)

-8

-180

10.2.5 Other effects

The category ‘other effects’ include tax consequences for the public, work supply distortion and

work supply gains. Each has been calculated using the TERESA-model and adhere to the

assumptions in the manual for socio-economic analysis within transportation. Table 23 specifies

each of the three categories. An elaboration of each of the separate subcategories can be found in

the appendix section 13 or in the manual for socio-economic analysis with transportation projects.

Entry

Tax consequences

Work supply distortion

Work supply gains

Total net present value (NPV)

Table 23: Externalities (NPV in mDKK)

Alternative 1

–

Base UTO

271

-38

257

Alternative 2

–

HA36

-353

-415

582

-185

10.3 Sensitivity

Socio-economic analysis always involves some uncertainties due to uncertainty surrounding the

used input and assumptions. To determine the robustness of the analysis, these uncertain

parameters have been varied to investigate how important these are to the socio-economic result.

We have identified the following parameters which we consider crucial for the analysis and which

might involve uncertainty.



Investment in capital



Socio-economic interest rate (the rate of return which can be expected from similar

projects)



Value generated through time savings (time saving unit prices)



Expected passenger growth rate

Entry

Main result

Investment in capital: +25 pct.

Investment in capital: -25 pct.

Socio-economic interest rate: 3 pct.

Socio-economic interest rate: 5 pct.

Unit prices for time savings: -25 pct.

Unit prices for time savings: +25 pct.

Passenger growth: 0 pct.

Passenger growth: 2 pct.

Alternative 1

–

Base UTO

594

106

1.081

1.052

261

669

518

572

531

Alternative 2

–

Flex36

3.114

2.551

3.676

4.415

2.148

1.791

4.436

2.393

3.038

Table 24: Sensitivity (NPV in mDKK)

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

Table 24 indicates that the socio-economic results are rather robust as neither of the sensitivity

scenarios resolve in negative net present values. This is the case when changing the amount of

investment in capital, the interest rate, unit price for time savings and passenger growth.

TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report

11. CONCLUSION

The conclusion is derived from the overall analysis where there is focus on the strongest key

findings and can be summarized as follows:



UTO attraction

When purchasing the next generation of rolling stock for the S-Bane in Copenhagen, it is

possible to define UTO scenarios as much more attractive and viable option, than a

comparable STO mode. From a financial perspective the attraction can be clearly

demonstrated in the possibility to provide the same service at reduced cost (frequency of 30

trains per hour in central section with an operational pattern as it is today), or to increase the

service level at the same cost (frequency of 36 trains per hour in central section with a

combination of classic and metro style operation). If also taking the Social Economic analysis

into account, the scenario with best service provision also exhibits the best economic

performance.

Operational flexibility

The UTO attraction and the investigated scenarios should be understood as a confirmed proof-

of-concept,

rather than suggestions for specific future operational plans. For that reason, it’s

not required that a decision about introducing UTO is necessarily followed by decisions about

exact operational plan, or a specific traffic pattern. This could be addressed as a part of the

potential future tender process.

Technical feasibility

From a technical point of view, feasibility can be provided by different solutions that are all

functioning in commercial operation today. The major technical elements comprehend CBTC

based rolling stock with a max. speed of 120 km/h, coupling into units from 1 to 3 sets,

stepless boarding, objective detection systems (ODS) on platforms and fences in between

stations. No of the suggested technical solutions are characterized as green field. However,

due to innovation, even more efficient and robust solutions are expected to be developed and

introduced within the next 10 years.

Rollout based on early deployment and stepwise implementation

The recommended rollout scenario is based on the test and early deployment on Ringbanen,

leaving the vast majority of the network undisturbed until a more robust solution is

implemented and sufficient lessons learned are collected, reviewed and assessed. Once the

Ringbane is converted into a UTO mode, a three step successive rollout plan encompassing

the outer lines, shall be planed taking the complexity of each line into account, including the

accessibility to depots and workshops.

Organisational impact

The most obvious organizational impact will be that +500 of the existing train drivers that will

be made redundant on the S-bane. The existing on-train and platforms service personnel may

probably be enlarged to a team of an estimated 200 service stewards, as known from the

Metro today. The changes give a major positive contribution to the final financial results.

However, a combination of natural reduction in staff (retention, job changes) and potential

transfer of train drivers into regional- and long distance trains (fjernbanen), makes the

redundancy among train drivers and the need for layoffs manageable. It’s recommended

keep a strong focus on the general transition process, which should be supported by change

management strategies and strong leadership.



TRU, Alm.del - 2016-17 - Bilag 365: Offentliggørelse af rapport om førerløse S-tog, fra transport-, bygnings- og boligministeren

Main report



Risk Assessment

The overall risk assessment points in the direction that the most considerable incidents may

well be related to dependencies on other external development projects, which are outside

the control of potential UTO project. And, the combination of newer proven solutions mixed

with an exsisting system, i.e. UTO mode in a large railway network not originally designed for

driverless solution almost entirely above ground, may also be a challenge. Due to these

uncertainties and the early stage in the investigation, a risk premium of 50 pct. is added to all

infrastructural investments, while 30 pct. is added to the rolling stock. In addition, when it

comes to external dependencies, a strong mitigating response will be, that much more

evidence about the actual performance will be available long before the necessary decisions

will need to be made in regards to UTO.

Financial and socioeconomic results

From a financial perspective, the UTO attraction is derived from four main sources: Extra cost

related to new infrastructure, extra cost to UTO compliant Rolling Stock, increased income

due to more passengers, and reduced staff costs due to UTO mode. All this is estimated as a

delta result, compared to a STO mode. From a Socio economic perspective, the major

contributor comes from user benefits, in particular, time savings in public transportation, in

the scenario with 36 hourly trains.



To summarize, the analysis emphasises that UTO is the best option with the best obvious potential

for future benefits. It’s therefore, the recommendation to take the next step in the decision

process in the direction of UTO. And as it stands, further investigation will be required. However,

this can be incorporated in a long term procurement strategy based on a controlled decision gates,

and thereby allowing the process to continue as soon as the required decisions are made.