TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Transportudvalget 2023-24
TRU Alm.del Bilag 52
Offentligt

Welkom

Redesigning the

railways of the

Netherlands

Wout Knijnenburg

Klaas Hofstra

Utrecht, October 2023

Eigenaar: Klaas Hofstra

14 april 2016

Status: Definitief

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Train Punctuality Netherlands (measured at 2'59" )

COVID

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

The lesson from Japan

We designed a railway system that is impossible to operate

punctually. And if it goes wrong, we blame our operational staff.

We cannot expect our operational staff

to solve the problems we created.

Instead we should enable them to do their job.

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Japanese Traffic Control

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Dutch Traffic Control

(Amsterdam)

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

First better, then more

We want to run more trains. But that only works with a stable system. So we

took hundreds of steps in improving timetable, infrastructure and

operations. Most small, some big.



Most ideas are no-brainers, but get lost in a debate on how much it will

help exactly. “Cash is King” => If it helps punctuality, implement it ASAP!



Some ideas are good for punctuality, but hurt on other aspects. In these

cases, obviously a good balance needs to be found.



Bear in mind: punctuality measures should always contribute to

increasing capacity and/or lower operational cost.

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Integrated “Taktfahrplan” is

the foundation

• Recognizable and easy to use

for passengers as well as

operational staff

•Optimal deployment of

personnel and rolling stock

• Better optimization possibilities

for infrastructure

• Basis for punctuality

improvements on a busy

network (NL, CH, DE)

Redesigning the Dutch Railways

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Nobody creates delays for fun

Delays occur if the dispersion in the operation is larger than the

buffers in the timetable.

This is why most punctuality improvement plans in European Railway

companies do not work. Intensive research taught us that there are only a

few ways to improve punctuality. They are heavily dependent on each other:

1. Reduce the amount of crossing movements

2. Validate process times used for the timetable design

3. Allow for enough (but not too much) buffer in the timetable

4. If the timetable is good, then reduce dispersion in everyday operation

5. Improve the infrastructure where-ever possible

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

1. Reduce crossing movements

The best headway is no headway

If trains can run parallel instead of crossing each other, that is always the

best choice. Even without changing the timetable, the routing on the nodes

can solve or create a lot of problems.

No timetable change!

Only routing in the node

Common railway practice:

Trains separated in time

More important:

Trains separated in space

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

2. Validate process times

It is not possible to plan enough

buffer if you don’t know the exact

process times

The calculation of running, dwell and

turnaround times contains many errors.

Running time calculations are therefore

constantly validated with each other

(Donna/OT) AND with the realisation of

thousands of trains (Trento).

Differences are analysed and used to

improve the calculation.

Minimum running times are based on the

percentile, minimum dwell times are

based on the median.

Ekz

Bkf

Bkg

Hks

Hnk

Hna

Pmo

Whe

Pmr

Nhk

Pmw

Zdk

Zdb

Donna Trento OT

VIRM VIRM VIRM

3,1

2,2

3,3

5,9

2,9

17,3

1,2

8,8

0,7

0,4

0,5

0,6

4,0

0,9

1,5

18,5

2,9

2,1

3,3

6,0

2,9

17,2

0,0

9,8

0,0

1,1

0,0

1,1

4,4

0,0

2,0

18,4

3,1

2,1

3,5

6,0

3,0

17,7

9,7

1,1

1,2

4,4

1,9

18,3

Kaal

3,0

2,1

3,3

6,0

2,9

17,3

1,2

8,7

1,1

4,4

2,1

18,6

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Examples of mistakes we found in the

“perfect” travel time calculation



Stopping locations are not correct

Starting trains start at the signal instead of at the platform

Signal aspects were calculated incorrectly

Arrival at a red/green signal at a stopping station

Rolling stock characteristics had multiple mistakes

Same characteristics were used for trains with different engines

Catenary-free sections (incl. bridges) were not taken into account

Gradients were not taken into account for passenger trains

Catenary voltage incorrect

Braking in the calculation was not according to actual driver behaviour

Bug in the calculation of freight trains

Etc.

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

3. Enough (but not too much) buffer

We need to know whether we planned enough buffer. At the

second precise. Always. Everywhere.

The only way to do this is to run one nationwide

deterministic

simulation,

with buffers:



We use “93% reduced performance” to simulate running time margin of

7%. So each train always runs 7% slower than technically possible.



We use “additional track occupation of 60s” to simulate headway buffer

Why deterministic simulation is better than conflict detection?



Continuous check for all trains, everywhere



Not only detection of the conflict, but also the resulting delays (if any)



Hindered headways instead of unhindered



Second opinion / double check (addition to the planning system)

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Example of simulation result

If the simulations shows delays (>15s) and/or unplanned

stops, there is not enough buffer in the timetable.

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

4. Operational Quality

If we would like our trains to arrive on time,

it might just help to tell the onboard staff

what we mean by “on time”.

After an open discussion with

staff members, NS decided to give

all their onboard staff extra

training, tooling (watch/tablet)

and an updated performance

measurement.

IMPORTANT:

the success of these

measures depends on the quality

of the timetable.

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

5. Improve the infrastructure

Focus existing and new infrastructure projects on simplification

In all infrastructure projects (newly built or maintenance) we try to

add punctuality/capacity scope:



Improving signal aspects and/or placement



Improve speeds



Reduce crossing movements

If infrastructure capacity starts to become a serious problem, the best

option might be to start a large infrastructure project, in which the

complete layout of a node is optimised (“greenfield”).

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Utrecht Central Station

Several big rail projects integrated into one mega-project:

•

Widening of platforms 1-4

New Public Transport terminal (“OVT”)

Track layout south side (“Vleugel / Randstadspoor”)

Track layout north side (“Doorstroomstation Utrecht”)

+ major urban redevelopment around the station

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Platforms1-4

Platform 5/7

Platforms 8-11

Platform 14/15

Platform 18/19

Platform 20/21

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Design of track layout

Old

•

A lot of simultaneous routes

Flexible platform allocation

Large distance between signals

40 km/h

High flexibility in operation

Many switches & crossings

Unreliable diamond crossings

•

New

Conflict free corridors

Dedicated platform allocation

Short headways

80 km/h

Limited flexibility in operation

Elimination of switches & crossings

Standardised 1:15 switches

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

In service since december 2016

Utrecht C.

2015 2018

Infra failures

10 -85%

Delay mins 20.085 2.964 -85%

Train punctuality Utrecht vs. the rest of the NL (2'59'')

96,0%

94,0%

92,0%

90,0%

88,0%

86,0%

84,0%

82,0%

80,0%

2015 Q1

2015 Q2

2015 Q3

2015 Q4

2016 Q1

2016 Q2

2016 Q3

2016 Q4

2017 Q1

2017 Q2

2017 Q3

2017 Q4

2018 Q1

2018 Q2

2018 Q3

2018 Q4

2019 Q1

2019 Q2

2019 Q3

2019 Q4

Utrecht

NL excl. Utrecht

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

“From Utrecht to the whole country”

Main themes for network optimisation

Operational cost

Reduce the amount

of assets, mainly

switches

Reduce switch-

heating

Phase out non-

standard assets

Reduce slack in the

timetable

Operational

performance

Capacity /

Frequency

Sustainablity Safety

Reduce and

reuse materials

and assets

Reduce switch-

heating

Remove local

speed

restrictions

Timetable for

energy

efficiency

Simplify the

track layout

Remove and

improve level

crossings

Phase out non-

compliant

signalling

Enough margin

in the timetable

Reduce the amount Shorter blocks,

of possible routes

higher speeds

and “derusting”

Reduce disruptions ETCS, but only

and possessions for on decent

maintenance

infrastructure

Longer/wider

Enough margin in

the timetable:

platforms,

always, everywhere 740m

Give train drivers

more precise

information

Match

timetable and

infrastructure

ESSA: Eliminate, Simplify, Standardize and Automate

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Example business case

Maintenance cost

2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 Total 15yrs Total 20yrs

yearly maintenance cost s10k/swit #### #### #### #### #### #### #### #### #### #### #### #### #### #### #### #### -1.500.000 -2.000.000

Renewal of points (every 20k/switch

Renewal costs

Renewal of switch heating

50k/switch

Renewal cost of switch

Total

Not included:

Grinding of switches

Catenary renewal

Renewal of Interlocking

250k/sw ####

####

-200.000

####

-500.000

-2.500.000 -2.500.000

#### #### #### #### #### #### #### #### #### #### #### #### #### #### #### ####

-4.700.000 -5.200.000

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Current status

Number of switches

1-1-2018

New built since 2018

Keep / renew

Potential removal

Removed

In projects

Urgent

Not urgent

Mainline

3898

107

2149

1856

397

911

507

(excl. Rtd Port)

Hoofdspoor

Yards

2339

1381

1047

395

324

124

204

Behouden

Is gesaneerd

Wordt gesaneerd

Urgent

Niet urgent

Zijspoor

This table excludes around 300 new mainline switches

being built in optimised track layouts.

Behouden

Is gesaneerd

Wordt gesaneerd

Urgent

Niet urgent

TRU, Alm.del - 2023-24 - Bilag 52: Diverse præsentationer fra besøg i forbindelse med udvalgets studietur til Holland og Tyskland d. 30/10-2/11-23

Questions?