Intermediate report on the development of

railway safety in the European Union

2013
Intermediate report on the
development of railway safety in the
European Union
2013

European Railway Agency

Safety Unit

Contact: Vojtech EKSLER

Publication date: 15 May 2013
CONTENTS

BACKGROUND .................................................................................................................................................. II

SUMMARY ...................................................................................................................................................... III

SAFETY OVERVIEW ........................................................................................................................................... 1

HISTORICAL DEVELOPMENT OF RAILWAY SAFETY ............................................................................................................. 2

RISK LEVELS ............................................................................................................................................................ 4
Second set of CSTs .......................................................................................................................................... 6
Second assessment of the second set in 2013 ................................................................................................ 8

ACCIDENT OUTCOMES ................................................................................................................................... 10

SIGNIFICANT ACCIDENTS .......................................................................................................................................... 10

DANGEROUS GOODS ACCIDENTS ............................................................................................................................... 12
CASUALTIES FROM SIGNIFICANT ACCIDENTS ................................................................................................................. 13
PRECURSORS TO ACCIDENTS ..................................................................................................................................... 16
LEVEL CROSSING SAFETY .......................................................................................................................................... 17
RAILWAY SUICIDES ................................................................................................................................................. 20
SAFETY AND MARKET OPENING ................................................................................................................................. 22

ACCIDENT COSTS ............................................................................................................................................ 24

SAFETY OF INFRASTRUCTURE ......................................................................................................................... 25

TRAFFIC VOLUMES ......................................................................................................................................... 27

RAILWAY TRANSPORT OPERATION EFFICIENCY .............................................................................................................. 29

Passenger trains ........................................................................................................................................... 29
Freight trains ................................................................................................................................................ 29

INDEPENDENT ACCIDENT INVESTIGATION ..................................................................................................... 31

BACKGROUND INFORMATION ....................................................................................................................... 34

ANNEXES ........................................................................................................................................................ 36

ANNEX 1 — COMMON SAFETY INDICATORS................................................................................................... 37

LIST OF CSI DATA TABLES......................................................................................................................................... 37

LEGEND ............................................................................................................................................................... 37

ANNEX 2 — LIST OF SERIOUS ACCIDENTS – COLLISIONS AND DERAILMENTS - OCCURRING IN 2012 AND
NOTIFIED TO ERA ........................................................................................................................................... 43

ANNEX 3 — LIST OF NATIONAL SAFETY AUTHORITIES AND NATIONAL INVESTIGATION BODIES .................... 44
List of abbreviations

ATP automatic train protection

CSI common safety indicator
CSTs common safety targets
CT Channel Tunnel
ERA European Railway Agency
ERADIS European Railway Agency Database of Interoperability and Safety
ERAIL European railway accident information links
EU European Union
FWSI fatalities and weighted serious injuries
IM infrastructure manager
KPI key performance indicator
LC Level crossing
NIB national investigation body
NRV national reference value
NSA national safety authority
OSP observed safety performance
p.a. Per annum (per year)
RSD railway safety directive
RU railway undertaking
SMS safety management system

I
BACKGROUND
Safety of European railways is relatively high, being one of the safest modes of transport. Even so, it is essential
to maintain and improve the current level of safety for the benefit of European citizens. A safe railway is more
efficient and also a more attractive transport choice, enabling society to address the environmental and
st
economic challenges of the 21 century.

The European Railway Agency (ERA) is a cornerstone of the EU strategy for railway safety. It supports national
safety authorities (NSAs) and national investigation bodies (NIBs) in their tasks and provides evidence for policy
actions at EU level. It develops and promotes the common safety framework as a means for achieving an open
railway market in the EU. The Agency also coordinates activities such as monitoring and provides support for
the further development of EU legislation.

Monitoring safety performance is one of the key tasks of the ERA. The ERA collects, processes and analyses
different sets of data, in order to support recommendations on actions to be taken. In this way, the Agency
facilitates evidence-based policy- making at the EU level. By continuously monitoring and analysing safety
performance, the Agency provides the assurance that the objective of maintaining and improving safety where
reasonably practicable can be achieved.
1
The Agency is requested by EU legislation ( ) to produce a report on safety performance on biannual basis. As a
biannual report was published in 2012, there is no requirement to produce such a report in 2013; therefore no
regular “Railway Safety in the EU” report is published by the Agency. Nevertheless, the availability of annual
evaluation of railway safety performance is of vital importance for the Agency itself, EU and national policy
makers, public agencies and experts from consultancies and academia.

This intermediate report is intended to provide an updated overview of railway safety performance across the
EU and present results of various analyses of that performance. It also includes the results of various
benchmarking exercises that have not been previously published by the Agency. They are all intended to share
knowledge available to the Agency with all interested parties and enable them to make the railway system
safer, more efficient and more competitive.

This report is unique in its nature - it is a once-only publication prepared exclusively in electronic format and its
content has been customised to meet the present needs of the Agency and of its stakeholders, including the
European Commission. The methodology applied throughout this report stems from the presumption that one
2
cannot improve what one cannot measure ( ).

Although this report relies heavily on quantitative data, the ERA is starting to use qualitative information for
the evaluation of railway safety in the EU and its Member States in the coming years. The assessment of
processes carried by National Safety Authorities and National Investigation Bodies extends the possibility for
safety monitoring beyond safety outcomes and changes its nature from reactive to proactive.

1
( ) Article 9(2) of the Agency Regulation (881/2004/EC)
2
( ) In the words of Lord Kelvin: “When you can measure what you are speaking about, and express it in numbers, you know
something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a
meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced
to the state of Science, whatever the matter may be.”

II
SUMMARY
This report provides an annual overview of railway safety performance of the European Union. Following the
statutory biennial report published by the European Railway Agency in 2012 in accordance with the
requirements of the Railway Safety Directive, this is intended as intermediate report.

Most recently available figures confirm that railways remain one of the safest modes of transport in the
European Union and worldwide. However, it has become difficult to sustain the trend in reducing casualties on
railways. In particular, it appears that the sector continues to struggle to reduce the number of third-party
victims, which represents 90 % of all casualties on railways, excluding suicides.

Safety performance of EU Member States varies considerably, with a more than ten-fold difference in risk for
all categories of railway users. These differences have not reduced over the past few years and represent a
major challenge for EU policy makers.

The continuous opening of railway market does not appear to be a threat to safety: the countries with
advanced market liberalization have a better safety performance than other countries and they do not appear
to be losing their position with time.

Around 2 400 significant accidents occur each year on the railways of the EU Member States. Accidents to
persons caused by rolling stock in motion and level-crossing accidents constitute more than three quarters of
railway accidents, excluding suicides. In these accidents, around 1 200 persons are killed and a similar number
of persons are seriously injured each year.

In 2011, railway safety continued to improve across the EU, with 2 342 significant accidents resulting in 1 183
fatalities and 1 032 seriously injured. Accident figures have been decreasing considerably over the past five
years; the casualty numbers have seen slight, close to uniform reductions over the same period.

Level crossing users are the only category of third-party/external victims for which the number of causalities
has seen a reduction over the past five years; yet this reduction was less significant than the reduction in road
casualties over the same period.

The number of suicide and trespasser fatalities has not seen any significant reduction over time. In
consequence, while suicide and trespasser fatalities accounted for 84 % of all fatalities in 2007; their share has
increased to 90 % in 2011. External victims, i.e. suicides, trespassers and level crossing users made up 98 % of
railway fatalities in 2011.

Among 2 342 significant accidents that occurred in 2011, 44 were classified as serious accidents by National
Investigation Bodies (NIBs) and as such investigated independently. In total, the NIBs opened an investigation
into 249 accidents and incidents that occurred in 2011.

III
SAFETY OVERVIEW
Around 2 400 significant accidents occur each year on the railways of the EU Member States. Accidents to
persons caused by rolling stock in motion and level-crossing accidents constitute more than three quarters of
railway accidents, excluding suicides. In these accidents, around 1 200 persons are killed and a similar number
of persons seriously injured each year.

In 2011, railway safety continued to improve across Europe, with 2 342 significant accidents resulting in 1 183
fatalities and 1 032 seriously injured. Accident figures have been decreasing considerably over the past five
years; the casualties’ numbers have seen slight, close to uniform reductions over the same period (Figure 1).

4000
3730
3819
3500
3027
3000

2401
2500
2342
1370 1382
2000 1110 1236
1032
1500

1000
1517 1479 1385 1256 1183
500

0
2007 2008 2009 2010 2011

Fatalities Serious injuries Significant accidents

Figure 1: Significant accidents and resulting casualties for the EU-27 (2007–2011)

Comparisons of fatality risks for travelling passengers (occupants) reveals that train is one of the safest mode
of transport. The fatality risk for an average passenger is about 0.15 fatalities per billion kilometres,
comparable with the risk of commercial flight passengers of 0.1 fatalities per billion passenger kilometres.
The fatality risk for a train passenger is three times lower than the risk for a bus/coach passenger (Table 1).

Fatality risk (2008-2010)

Transport mode used by user
Fatalities per billion passenger kilometers
Airline passenger 0.101
Railway passenger 0.156
Car occupant 4.450
Bus/Coach occupant 0.433
Powered two-wheelers 52.593
Vessels passenger N/A
3
Table 1: Fatality risk of passenger using different mode of transport (EU-27 in 2008-2010) ( )

3
( ) Source of data: EU transport in figures (Statistical Pocketbook 2012), DG MOVE 2012, European Commission

Page | 1
Historical development of railway safety
The overall level of railway safety in Europe, as measured by fatal train collisions and derailments per billion
train-kilometres, has gradually improved since 1990, although there is considerable scatter from year to year.
4
The estimated overall trend is a reduction in the accident rate of 6 % per year ( ). This gives a fall of 70 % from
1990 to 2012 (Figure 2). The estimated underlying average number of fatal train collisions and derailments
per billion train-killometers was about 4.4 in 1990 and 1.3 in 2011. Despite a positive long-term trend in the
risk of fatal train collisions and derailments over the past two decades, the data in Figure 2 suggests that the
progress has been slowing down, in particular since 2004.

Figure 2: Fatal train collisions and derailments per billion train-kilometres in 1990–2012 for the EU-27,
5
Switzerland and Norway ( )

The number of fatalities in all railway accidents has seen a distinct, downward trend for all categories of
accidents, except level-crossing accidents. This can be partly explained by the continuous increase in road
traffic across Europe, as contributing to the likelihood of a level-crossing collision. The currently run
programmes to remove or upgrade level-crossings might not be extensive enough to compensate for the
increased risk of a level crossing collision.

Accidents with multiple fatalities rarely escape the attention of the media and the public, so data on these
may be more complete. Figure 3 is based on data from the historical archive of railway accidents maintained
by the Agency; it shows the number of major accidents and resulting fatalities for the 33 years 1980–2012. It
includes not only the train collisions and derailments with 5 or more fatalities, but also the major level-
crossing accidents, train fires, and accidents involving groups of persons struck by rolling stock in motion.

The trend in the accident rate per billion train-kilometres for accidents resulting in five or more fatalities is
strongly downward over the period 1990–2012, but somewhat less steep if taken back to 1980–2012. Figure
3 shows that there were on average eight major railway accidents each year during the 1990s, this figure has

4
( ) A. W. Evans (2011), ‘Fatal train accidents on Europe’s railways: 1980–2009’, Accident Analysis and Prevention 43(1),
391–401.
5
( ) Figure courtesy of Andrew W. Evans (Imperial College and University College London), based on own database of fatal
train accidents and collisions and on the train-km data from the UIC, Eurostat and the ERA.

Page | 2
now come down to five accidents per year in the 2000s. There were four accidents with five or more fatalities
in Europe in 2012, three of which were level crossing accidents.

6
Figure 3: Railway accidents with five or more fatalities (1980–2012) ( )

In conclusion, available historical data on fatal railway accidents shows a gradual improvement in safety over
the past three decades, however restricting the analysis to the past eight years creates uncertainty about the
trend in railway safety in Europe in recent years. This stems from the low number of fatal accidents and from
their random nature.

6
( ) All EU countries, Norway and Switzerland, excluding Romania for the period 1980–1989. Accidents on railway
mainlines not covered by the RSD are also included.

Page | 3
Risk levels
Accident risk expressed in the number of outcomes per exposure is probably the best measure of the safety
level. The framework for the evaluation of CSTs/NRVs also uses it as a basis for the assessment of safety
levels at the level of Member States and the Union. Considering all railway fatalities (excluding suicides), the
fatality risk per million train-km in the period 2009-2011 was 0.31 killed per million train km in the EU level.
th
Yet the values of risk vary greatly between countries: The risk of countries in the lowest and highest 10
percentile differs by a factor of 15. (i.e. the fatality risk for Estonia (1.36) is 15 times higher than the risk for
Ireland (0.09)). Interestingly, the countries with a fatality risk higher than average show much greater
variations in risk than those outperforming the EU average. (The distribution of risk among countries is clearly
asymmetrical, with the median value of 0.31, as compared to mean value of 0.60.) Actually, one third of
countries seriously undermine the low level of risk at EU level. Six EU countries have a level of risk that is at
least four times higher than the EU average as shown in Figure 4.

Railway fatalities per million train-km (2009-2011)

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.4 EU: 0.31

0.2

0.0

Figure 4: Fatality risk on EU railways: Fatalities per million train-km (2009-2011)

Another way to benchmark the level of risk of national railway systems is to look at the fatality risk of
passengers expressed in terms of passenger fatalities per passenger kilometres. Six countries and the Channel
Tunnel have recorded no passenger fatality in the period 2006-2011, thus their passenger fatality risk is zero.
The countries with a level of risk higher than the average are typically those with a high risk for all persons on
railways (8 countries have fatality risks for passenger and all users higher than EU average). This fact,
together with the similarities in the distribution shapes, discards the common belief that the two measures of
risk are not comparable and that the safety of passengers is not correlated to safety of other users.

There are certain limitations in the two benchmarking indicators: They rely on the numbers of fatalities only,
since serious injury data are not believed to be fully comparable between countries and the period
considered is not of the same length, because of limited compatibility of certain data before 2009. However,
these limitations do not seriously undermine the conclusions that can be drawn from Figure 4 and Figure 5.

Page | 4
 Passenger fatalities per billion passenger-km (2006-2011)
1.4

1.2

0.8

0.6

0.4

0.2 EU: 0.16

Figure 5: Passenger fatality risk: Passenger fatalities per billion passenger kilometres (2006-2011)

Similarly, one can show the levels of risk for different types of persons and different exposure. This is done in
the framework of common safety targets (CSTs) and national reference values (NRVs), where the NRVs exist
for 6 categories of persons and sometimes for two types of exposure. The NRVs are valid measures of risk
that can be used for benchmarking similar to the one shown in Figure 5, but their value is limited due to the
fact that they are not updated in regular enough intervals and often rely on relatively old data.

Selected descriptive statistics can be used to assess the development of differences in risk levels between
countries over time. Fatality risk per million train kilometres for three groups of persons is considered:
Passengers and employees, all railway victims except third parties (suicide, trespassers and LC victims) and all
victims except suicides. The fatality risk is estimated for two periods: 2006-2008 and 2009-2011. For all three
user categories under consideration, the fatality risk decreased considerably between the two periods
(decrease in mean and median values). Similarly, the standard deviation characterizing how widely values are
dispersed from the average value decreased over time, yet the relative decrease was more important for
passengers and employees (and other) victims. When also considering level crossing users and trespassers, the
relative reduction is only minor. The distribution of risk values for Member States is relatively peaked
regardless the category of users considered seeing from positive kurtosis values. A relative increase in kurtosis
in time suggests that the distribution has become less flat. With regard to passengers and employees, the
decreases in risk variation have been driven by reductions in risk in countries which had relatively high risk
values in the first period, given the increase in skewness. For other categories, the decrease in the variation of
risk is a result of an overall reduction recorded for all countries.

Fatality risk per million train-km Period Mean Median SD Kurtosis Skewness
2006-2008 0.049 0.039 0.043 0.021 0.882
Passengers and employees
2009-2011 0.030 0.022 0.030 6.280 2.194
2006-2008 0.081 0.056 0.106 15.139 3.568
All except third parties
2009-2011 0.048 0.031 0.054 9.678 2.871
2006-2008 0.724 0.379 0.665 0.075 0.973
All fatalities except suicide
2009-2011 0.596 0.306 0.562 0.163 1.093
Table 2: Descriptive statistics for fatality risk in periods 2006-2008 and 2009-2011

Page | 5
Common Safety Targets
Common safety targets (CSTs) are quantitative tools intended to monitor whether the current safety levels of
the railways in the Member States are at least maintained. In the long term, they could also help to drive
efforts to reduce the current differences in railway safety performance. Railway transport is the only mode of
transport for which the targets have been prescribed by European legislation. The CSTs are EU-wide
maximum risk values, the national reference values (NRVs) are the maximum risk levels set for individual
7
Member States. The risks are measured by the number of weighted fatalities (FWSI ( )) per train-kilometre.
There are risk categories for passengers, employees, level-crossing users, unauthorised persons on railway
premises, ‘others’ and as applied to society as a whole.

Second set of CSTs

As required by the RSD, the Agency proposed and the EC adopted the second set of CSTs and NRVs in 2012.
The CSTs and NRVs of the second set are based on a six-year time series of data, from 2004 to 2009, that
were delivered to Eurostat by Member States. The second set of CSTs and NRVs was calculated with the same
method, defined in the CSM on the assessment of the achievement of CSTs, as the first set. The revisions
made by the countries to Eurostat data were taken into account. Compared with the first set of CSTs and
NRVs, the only difference is the extension of the period for which data was used for calculations. The
calculation method, the data source and the risk categories are the same as in the first set. The values for the
second set of CSTs are shown together with the values of the first set in Table 3.

Risk category / Risk in terms of FWSI per CST 1.set (× E-06) CST 2. set (× E-06)
exposure (2004–2007) (2004–2009)
Risk to passengers - per train-km CST 1.1 0.25 0.17
- per passenger-km CST 1.2 0.00201 0.00165
Risk to employees CST 2 0.0779 0.0779
CST 3.1 0.743 0.710
Risk to level-crossing users
CST 3.2 n.a. n.a.
Risk to ‘others’ CST 4 0.0185 0.0145
Risk to unauthorised persons on railway premises CST 5 2.03 2.05
Risk to the whole society CST 6 2.51 2.59
Table 3: Values of the second set of CSTs for different risk categories

Figure 6 shows the values of the second set of NRVs for train passengers. The two indicators are showed at
the same figure, the FWSI per passenger train-km and FWSI per 100 passenger-km. Not surprisingly, the two
indicators are relatively well correlated, despite the fact that the hypothetical average train occupancy may
vary considerably between MS. There are big discrepancies in terms of risk values among countries. The risk
levels of Member States (estimated as NRV) differ by up to sixty times.

A detailed look at the descriptive statistics on NRV values (2004-2009) unveils huge variations in NRV values
as expressed by the ratio between largest and smallest NRV value. The variation in NRV values for different
categories of railway users is expressed through the coefficient of variation; the variation is most significant
for the categories of passengers and for unauthorized persons.

While the extreme variation recorded for unauthorized persons is partly the result of poor statistics (suicide
fatalities are often confused with unauthorized persons fatalities), the prevailing variation in risk for all
categories of railway users under consideration is enormous and may be reflecting the different levels of
infrastructure safety in Member States and the differences in the level of implementation of the common
safety framework.

7
( ) Weighted fatalities and serious injuries are the normalised measure of railway safety outcome. One seriously injured
person is considered as 0.1 fatalities and added to the number of fatalities in the given year.

Page | 6
 NRV 1.1&1.2: Passenger risk (2004-2009)
1.8E-07

1.6E-07
CST [NRV(EU)] for 1.1: 1.70 E-07
1.4E-07 CST [NRV(EU)] for 1.2: 1.65 E-09
1.2E-07

1.0E-07

8.0E-08

6.0E-08

4.0E-08

2.0E-08

0.0E+00

NRV 1.2: FWSI passengers per 100 pax-km NRV 1.1: FWSI passengers per train-km

Figure 6: CST and NRVs for the passenger category (second set based on 2004–2009 data)

Two additional descriptive statistics are showed in Table 4: kurtosis and skewness. The low values of kurtosis
8
indicate that the single NRV values are widely spread around the mean ( ). The positive values for skewness
(right skewed distribution) indicate that most values are concentrated to the left of the mean, with extreme
values to the right. This means that a small number of countries have significantly higher NRVs compared with
the average.

Unauthorized Whole
Passenger Employees LC users Others
NRV (FWSI) persons society

1.1 1.2 2 3.1 4 5 6

Mean 4.70E-08 4.89E-10 2.21E-08 2.18E-07 8.01E-09 4.89E-07 7.60E-07

Median 3.25E-08 3.00E-10 1.45E-08 1.62E-07 7.00E-09 2.02E-07 3.73E-07
Largest 1.70E-07 1.65E-09 7.79E-08 7.10E-07 1.45E-08 2.05E-06 2.59E-06

Smallest 2.73E-09 2.76E-11 1.36E-09 2.16E-08 2.41E-09 1.59E-08 5.09E-08

Largest/Smallest 62 60 57 33 6 128 51
Coefficient of
1.02 1.04 0.98 0.81 0.53 1.12 0.94
variation
Kurtosis 1.61 1.02 0.93 0.87 -1.46 1.34 0.13
Skewness 1.45 1.36 1.39 1.04 0.32 1.39 1.01
nd
Table 4: Descriptive statistics for NRVs (2 set based on Eurostat 2004-2009 data)

8
( ) Leptokurtic distribution, sharper than a normal distribution, with values concentrated around the mean.

Page | 7
Second assessment of the second set in 2013
The second assessment of the second set was carried out by the Agency in early 2013 and delivered to the
Commission at the end of March. The assessment was based on a five-year time series (2007–2011) of data
on railway casualties that were delivered to Eurostat by Member States according to Annex H to Regulation
(EC) No 2003/91 on rail transport statistics. The assessment was made for six risk categories of CSTs and NRVs
9
using the method set by Commission Decision 2009/460/EC ( ).

In general, the results of the annual assessment of achievements of CSTs/NRVs indicated that railway safety
performance remains acceptable at the EU level for all categories of railway users under consideration. The
results further showed other than acceptable safety performance in four Member States, usually for one
category of railway users (Table 5). Only in one case the result of the assessment was “probable deterioration
of safety performance”. In some cases, the negative result of the assessment was due to poor quality data in
years before 2007, used to set up the second set of CSTs. Following the consolidation of data carried out by
NSAs at national level, the Agency recommended to the EC to revise certain values of NRVs for Slovakia,
Bulgaria and Romania.

Unauthorised Whole
Passengers Employees LC users Others
Risk category persons society
10 11
1.1 ( ) 1.2 ( ) 2 3.1 4 5 6
Romania
Possible Romania
Slovakia Slovakia None Romania Slovakia Romania
deterioration Slovakia
Sweden
Probable
none none Bulgaria None None None None
deterioration
Table 5: Results of the assessment of achievements of the second set of CSTs/NRVs - other than acceptable

Figure 7 shows the intermediate results of the second assessment of the second set of CSTs/NRVs for the
category of whole society. The values of NRVs (second set) are plotted together with the values of OSP
(Observed Safety Performance) in 2011 and MWA (Moving Weighted Average Value) for years 2007-2011.

To achieve an acceptable safety performance after two steps of the method, the OSP, or MWA have to be
lower than the NRV stepped up by 20 %. The OSP value was however higher than NRV in nine countries (NO,
SE, FR, IT, BE, SK, BG, RO, PL) and the similar observation was made for MWA value that was higher than NRV
in nine countries (NO, IE, SE, IT, BE, SI, SK, RO, PL). It is largely thanks to the application of the 20 % margin
that all countries but four showed acceptable safety performance in 2011.

9
( ) Commission Decision 2009/460/EC on a common safety method for assessment of achievement of safety targets.
10
( ) Scaling base: passenger train-km per year
11
( ) Scaling base: passenger-km per year

Page | 8
 NRV & MWA 6: Societal risk (2004-2009)
2.5E-06

2.0E-06

1.5E-06

1.0E-06

5.0E-07

0.0E+00

OSP 6: FWSI per train-km MWA 6: FWSI per train-km NRV 6: FWSI per train-km

Figure 7: NRVs, OSP and MWA risk levels for the whole society (second set based on 2004–2009 data)

In the fourth assessment step, a check is made whether the number of significant accidents per train-km,
with respect to the previous years, remained stable (or decreased). The criteria for this appraisal are whether
there has been a statistically significant increase in the number of relevant significant accidents per train-km.
This is evaluated by using an upper Poisson tolerance bound which will determine the acceptable variability
based on the number of accidents that occurred in the different Member States.

In the 2011 assessment, all countries but one (Bulgaria) passed this test for all types of significant accidents
under consideration. In case of Bulgaria, the number of all significant accidents and the number of accidents
caused by rolling stock in motion in 2011 was higher than the number of accidents in previous years.

Page | 9
Accident outcomes

Significant accidents
Around 2 400 significant accidents occur each year on the railways of the EU Member States. Accidents to
persons caused by rolling stock in motion and level-crossing accidents constitute more than three quarters of
the total number of accidents, excluding suicides. The number of significant accidents per accident type in the
period 2010–2011 is shown in Figure 8.

For collisions of trains, level crossing accidents and other accidents, the reported number of accidents in 2011
was lower than in the previous year. There was an increase in the number of accidents between 2010 and
2011 for derailments, fires and accidents to persons caused by rolling stock in motion.

On average a derailment or a collision is reported at least every second day in the EU, causing significant
disruptions to railway operations.

The Member States reported 1 480 accidents to persons caused by rolling stock in motion in 2011. The risk of
this type of accident is relatively high in the three Baltic countries and in some Central and East European
countries (Czech Republic, Slovakia, Poland, Romania).

The number of fires in rolling stock reported for 2011 (25) is similar to the number of fires reported in
2010 (23). At least one such fire in rolling was recorded in 11 EU countries in 2011 (compared to 8 in 2010).

A wide range of accidents, not included within the specific types of accidents, are included in the category of
other accidents. The 129 cases reported in 2011 include collisions and derailments of shunting rolling
stock/maintenance machines, dangerous goods released during transport, objects projected by the running
train, and electrocution in connection with the rolling stock in motion; the category other accidents is the
third largest group of accidents.

Significant railway accidents (EU-27)

2 500 2401
2342

2 000

1480
1 500 1420

2010

1 000 2011

619
528
500

145 129
99 83 95 97
23 25
0
Collisions of trains Derailments of trains Level-crossing Accidents to persons Fires in rolling stock Other accidents Total nr accidents
accidents caused by rolling
stock in motion

Figure 8: Reported number of significant accidents per accident category (2010–2011)

Page | 10
 Collision types and their outcomes Total EU Share
Trains collisions 11 13 %
Collisions with obstacles 72 87 %
Persons killed and seriously injured in trains collisions 7 12 %
Persons killed and seriously injured in collisions with obstacles 50 88 %
Table 6: Collisions of trains by type and their outcomes (EU-27 in 2011)

In 2012, the number of collisions and relevant outcomes was reported by type for the first time, on a
voluntary basis. Among 83 collisions reported in 2011, only 11 were trains collisions, the remaining 72
collisions involved a train hitting an obstacle within the clearance gauge. The number of casualties per
collision type follows a similar pattern, with on average 1 person killed and 6 seriously injured in trains
collisions and 50 people killed and seriously injured in collisions with an obstacle (Table 6).

Over the past five years, the number of significant accidents has seen a reduction of 38 % (9.3 % p.a. on
average). This is almost twice as much as the reduction achieved for fatalities (22 %) and serious injuries
(25 %). The number of significant accidents has reduced in all categories of accidents, with most important
reductions in the category of collisions, derailments and fires. The lowest reduction was achieved in the
category of accident to persons by rolling stock in motion.

Given the heterogeneous development in the number of significant accidents across the different categories
of significant accidents, the relative share of accident types has changed dramatically over the past years. This
is shown in Figure 9. The relative share of collisions and derailment dropped from 16 % in 2007 to 8 % in
2011, while the number of accidents involving third parties (LC accidents and accidents to persons) increased
from 75 % to 86 %.

2007 2011

3% 1%
Collisions of trains
7% 7% 6% 4% 4%
9%
Derailments of trains
23%
LC accidents

Accidents to persons
44% 31%
63% Fires in rolling stock

Other accidents

Figure 9: Reported number of significant accidents per accident category (2007 and 2011)

Page | 11
Dangerous goods accidents
When a railway accident involves dangerous goods, whether they are being transported or not, it must be
reported under a separate category of accidents: accidents involving dangerous goods. Depending on the
type and consequences, an accident involving dangerous goods may also be reported in duplicate as a
significant accident. In 2011, Member States reported a total of 28 accidents involving dangerous goods; in
nine of these, the transported dangerous goods were released during the accident. The 28 accidents involving
dangerous goods occurred in 11 EU countries.

Dangerous goods accidents 2010 2011

Number of accidents involving at least one railway vehicle transporting dangerous

17 19
goods in which dangerous goods are NOT released

Number of accidents involving at least one railway vehicle transporting dangerous

37 9
goods in which dangerous goods ARE released
Total number of accidents involving at least one railway vehicle transporting
54 28
dangerous goods
Table 7: Railway accidents involving dangerous goods (EU-27)

Page | 12
Casualties from significant accidents
In parallel with the decrease in railway accidents, the total number of casualties, excluding suicides, has fallen
steadily in recent years. There were 1 183 fatalities reported for the year 2011, a six per cent decrease from
the previous year (1 256 fatalities recorded in 2010). The number of passenger casualties (fatalities and
serious injuries) fell down to the pre-2010 level, with 38 passenger fatalities in 2011. The unusually high
number of fatalities among train passengers in 2010 was largely driven by the outcome of one single
occurrence, the collision of trains in Belgium on 15 February 2010 that alone led to 19 fatalities and 35
serious injuries.

Fatalities in railway accidents (EU-27)

1 400

1 256
1183
1 200

1 000

797
800 750 2010

2011
600

400 359
294

200

62 44 41
38 29 25
0
Passengers Employees Level crossing users Unauthorised persons Other persons Total persons

Figure 10: Number of fatalities per victim category (2010–2011)

Figure 10 shows the number of fatalities in different categories of persons over the period 2010–2011. With
797 fatalities in 2011, unauthorised persons represented 67 % of all persons killed on railway premises. The
number of level-crossing fatalities of 294 in 2010 is by far the lowest ever recorded on EU railways. This figure
represents 25 % of railway fatalities, but only 1.1 % of road-user fatalities. Level crossing safety might
therefore be perceived as a marginal problem by the road sector, while it is a key problem for the railway -
also because of its impact on railway operation.

Railway fatalities per victim category

3% 1% 1%
3% 3% 9%
Passengers
Employees
28% 19%
Level crossing users
Unauthorised persons
Other persons
63% 1%
69% Suicides

Figure 11: Relative share of fatalities per victim category among railway and all fatalities (2009–2011)

Page | 13
Suicides are reported separately from accident fatalities. They represent 69 % of all fatalities on railways and,
together with the unauthorised person fatalities, constitute 88 % of all fatalities occurring within the railway
system (period 2009-2011). In 2011, on average 8 suicides were recorded everyday on the EU railways,
totalling 2 868, a record number since 2006. Several Central and Eastern European countries registered a
significant increase of railway suicide fatalities in 2011; only seven EU countries saw their suicide figures
falling in 2011.

Figure 11 shows that if we exclude suicide fatalities, the majority of fatalities are unauthorised persons. Level-
crossing accidents account for 25 % of fatalities, whereas passenger and employees fatalities make up 6 % of
the total number of deaths on railways. People strictly internal to railway operation (passengers, employees
and other persons) represent only three per cent of persons killed on EU railways.

Fatalities in railway accidents by category of victim (EU-27)

4500

4000

3500

3000
2622 Suicides
2773
2422 2743 2868 Other persons
2500
1874
Unauthorised persons
2000 Level crossing users
Employees
1500
Passengers

1000 855 928

940 852 750
797
500
366 504 380 359
405 294
0
2006 2007 2008 2009 2010 2011

Figure 12: Fatalities on EU railways per year and victim category (2006-2011)

Figure 12 shows that although the total number of fatalities on EU railways has been steady since 2007 (with
around 4 000 fatalities in total), there has been an increase in the number of suicide fatalities. Suicide and
unauthorized user fatalities accounted for 84 % of all fatalities in 2007; their share has increased to 90 % in
2011. Victims not inherent to the railway system (suicides, unauthorized persons and level crossing users)
make up 98 % of railway fatalities.

Over and above the number of fatalities, a large number of persons are seriously injured each year on the
railways. Over the past five years, for each 10 persons killed, Member States reported some nine seriously
injured persons. This ratio, illustrating the seriousness of accidents, has been constant over time, with the
exception of 2009, in which there were only eight seriously injured per 10 persons killed in significant railway
accidents.

In 2011, 1 032 persons were seriously injured, a decrease of 204 over 2010 when 1 236 serious injuries were
reported (Figure 13). The numbers of injured passengers and level crossing users reported for the period
2010-2011 show variations beyond what might be expected from natural fluctuation).

Page | 14
 Serious injuries in railway accidents (EU-27)
1 400

1236
1 200

1032
1 000

800 2010
2011

600

433
409
400 354
327
273
217
200
93 74
53 35
0
Passengers Employees Level crossing users Unauthorised persons Other persons Total persons

Figure 13: Number of serious injuries per victim category (2010–2011)

The decrease in the number of casualties (fatalities and serious injuries) in recent years is promising,
especially in the categories of passengers, employees and other persons, who are all users internal to railway
system. At the same time, the trend in the number of unauthorized person casualties is a cause of concern.

Over the past five years, there were 11 fatalities per 10 seriously injured persons on EU railways. Persons
being hit by a train are the users most likely to die. There are almost two killed trespassers per one seriously
injured trespasser. Among all railway users, passengers are most likely to survive in significant accidents.
Passenger fatalities represent only one fifth of all railway casualties. Analysing the seriousness of injuries over
time does not reveal any significant trends since 2006.

Fatalities per serious injury in significant accidents (EU-27)

2

1.8

1.6

1.4

1.2

0.8

0.6

0.4

0.2

0
Passengers Employees Other persons Level crossing Unauthorised Total persons
users persons

Figure 14: Seriousness of injury in significant accidents: fatalities per seriously injured (2007-2011)

Page | 15
Precursors to accidents
As accidents on railways are rare, the monitoring of less serious events occurring on railways is an essential
tool of a proactive SMS. ‘Precursors to accidents’ are indicators of incidents that under other circumstances
could have led to an accident. The indicators reported to the Agency are: broken rails, track buckles, signals
passed at danger, wrong-side signalling failures, broken wheels and broken axles (Figure 15).

Over the period 2010–11, EU countries reported as many as 20 650 precursors to accidents; this is a ratio of
more than four precursors to one significant accident. However, if we discard accidents to persons caused by
rolling stock in motion, the ratio between the precursors and accidents rises to 11:1. This unveils the great
potential benefit in analysing precursors in the proactive monitoring of railway safety.

Signal passed at danger is the most common type of accident precursors; it is also a precursor for which the
highest absolute reduction has been registered in 2011. The most important reduction has been however
achieved for the category of broken wheels and broken axles (a 40 % year-to-year reduction when taken
together).

Precursors to accidents (EU-27)

6 000
5563

5 000

4486

4 000

3 000 2010
2749 2700
2011
2213

2 000 1776

1 000
525
433

56 33 43 28
0
Signals passed at danger Broken rails Track buckles Wrong-side signalling Broken wheels Broken axles
failure

Figure 15: Reported number of precursors in 2010-2011 (EU-27 countries and Norway)
Signals passed at danger is also the most commonly investigated type of incidents as shown by the overview
of the number of incidents investigations carried out by NIBs (Table 8). In the past three years (2008-2012),
there were on average 13 SPADs investigated by NIBs, compared to 6 other types of incidents for which an
investigation has been carried out by a NIB.

Incident 2006 2007 2008 2009 2010 2011 2012

SPAD 4 3 6 16 13 14 15
Other 2 0 4 7 4 8 9

Table 8: Number of incidents investigated by NIBs (2006-2012)

Page | 16
Level crossing safety
Level crossings constitute a significant safety concern. In recent years, on average, every day, one person has
been killed and close to one seriously injured at level crossings in Europe. This is shown in Figure 16 that
summarizes the development of level crossing accident casualties since 2007.

Casulties in level crossing accidents (EU-27)

600
545 550

500
446
410
389
400 372 360
332
300 304
300

200

100

0
2007 2008 2009 2010 2011

Fatalities Serious Injuries

Figure 16: Development of casualties in level crossing accidents (EU-27)

Level crossing accidents and fatalities represent more than one fourth of all railway accidents on EU railway.
However, level crossing fatalities make up only one per-cent of all road deaths. While level crossing safety has
been traditionally viewed as a road safety problem by railway infrastructure managers, road authorities often
struggle to address the problem in relation to other urgent road safety problems they seek to address. The
concept of shared and delegated responsibility often fails to deliver the targeted results when it comes to
level crossing safety and may need to be revised at both EU and national level.

Railway fatalities (2009-2011) Road fatalities (2009-2011)

Passengers Level
Others Level
3% Employees crossings
3% crossing
3% 1%
users
28%

All roads
Unauthoris except LCs
ed persons 99%
63%

Figure 17: Share of fatalities in level crossing accidents in all other railway and road accidents (2009-2011)

There has been a sound reduction in level crossing accident fatalities in recent years; level crossing user
deaths have been decreasing by 5.7 % per year on average. However, the reductions in the number of deaths
in all other types of railway and road accidents have been even more significant; their numbers have been
decreasing at an average annual rate of 6.3 and 8.0 % respectively.

Page | 17
 Fatalities by type 2006 2007 2008 2009 2010 2011 % change p.a.
LC fatalities 366 504 380 405 359 294 5.7 %
All other railway
1105 1013 1099 980 897 897 6.3 %
fatalities
All other road fatalities 42 700 42 000 38 550 34 400 30 500 30 200 8.0 %

Table 9: Development in level crossing accident fatalities compared to all other types of accidents (EU-27)

The fatality risk at level crossings in EU countries was estimated for the period 2009-2011 and the countries
were ranked in Figure 18. The difference in risk between the countries with the smallest fatality risk at level
crossings (United Kingdom with 19 deaths per billion train km) and the country with the highest level of risk
(Greece with 537 deaths per billion train km) is huge, there is a 28 fold difference in the estimated risk of
fatality at level crossings. The variance in risk remains significant, even if we disregard the 10th percentiles of
countries with highest and lowest risk levels.

Fatality risk at level crossings (fatalities per million train-km) - 2009-2011

0.6

0.5

0.4

0.3

0.2

0.1

0.0

Figure 18: Fatality risk at level crossings: Level crossing fatalities per million train-km (2009-2011)

The estimation of trends in accidents and other outcomes for EU countries reveals that while the number of
significant accidents on level crossings has been decreasing at a steady pace since 2006 (by 15 per cent per
year on average), the number of casualties (fatalities and serious injuries taken together) was almost
constant. At the same time, casualties on railways have been decreasing by about 4 % per year, on average.
In other words, while there has been a marked reduction in level crossing accidents (and casualties) over the
past five years, this reduction disappears when one looks at the development in railway casualties. This
finding is surprising; one would expect a strong correlation between the trend in the number of accidents and
related casualties. Possible causes of this discrepancy include poor reporting practice and a sudden increase
in the seriousness of LC accidents. A longer time series is needed to confirm the estimated trends in order to
draw clearer conclusions.

One can assess the level of risk at LCs alongside progress in its reduction in different EU countries. This
analysis is limited to 21 countries, for which the relevant data (casualties, train-km) were available for the
whole period under consideration (2006-2010). In Figure 19, KSI risk (number of killed and seriously injured
people in LC accidents per million train kilometres) for the past three years (2008-2010) is plotted horizontally
against the estimated average annual percentage change in LC casualties. The EU averages of the two
indicators are used to divide the diagram into four quadrants. Three countries with the highest annual

Page | 18
number of train-km (Germany, United Kingdom and France) appear in the left-hand part of the graph among
countries with lower than average casualty risk at LCs, weighting heavily on the EU average. Only France,
Germany, Denmark and Sweden achieved lower than average KSI risk after higher than average reductions in
KSI risk. The graph also shows that there is no correlation between the level of casualty risk and the trend in
risk.

The estimated values of the annual average percentage change in Figure 19 are sensitive to sudden changes
in the number of casualties, behaving as random variables. A longer time series of data would be needed to
assess the trend with a higher degree of reliability.

20
NO

15
Average annual percentage change in KSI per train-km

10
SE
5
DK
SK
(2006-2011)

0 FR BG
IT HU
CZ
DE EU PL LT
-5
NL FI AT LV
-10 UK EE

IE BE
-15
ES

-20 SI RO

PT
-25
0 0,2 0,4 0,6 0,8 1 1,2
KSI per million train-km (2009-2011)

Figure 19: Casualty risk (killed and seriously injured per million train kilometres) versus average annual
percentage change in casualty risk from 2006 to 2011

With about 50 accident investigations into LC accidents by NIBs per year, not even one tenth of all fatal LC
accidents are subject to independent investigation. In many cases, these investigations do not seek root and
underlying causes, significantly limiting their value. Although these investigations cost money, it is a
worthwhile investment, if we consider the costs to society of these accidents.

Infrastructure managers (IMs) of EU countries regularly issue statements that about 95 % of LC accidents are
caused by LC users, who break (road traffic) rules, either intentionally or unintentionally. The liability is then
often confused with responsibility, being quite a different issue. When an in-depth accident investigation into
a LC accident is carried out, problems are frequently identified with specific safety barriers. Moreover, not all
IMs apply a holistic and analytical approach for LC safety improvements.

Page | 19
Railway suicides
Railway suicides are persons recorded and classified as suicide by competent national authorities. While the
classification approach is mature in most Member States, some incertainty prevails in a few countries, where
some suicide fatalities may be confused with trespasser fatalities.

The number of suicide fatalities continues to rise across the EU. Every year, more than 2 500 suicide fatalities
and additional 800 trespasser fatalities occur on EU railways (Figure 25). While accounting for 8 % of all
suicides, the societal impact of suicides on railways remains considerable. The consequences are not only
trauma for all parties involved, but also significant costs incurred by delays, deployment of rescue services,
loss of productivity or employees involved etc..

On average, railway suicides account for 8 % of all suicides across the EU. Only in the Czech Republic,
Germany, the Netherlands and Slovakia is their share higher than 10 %. In Greece, Ireland, Poland and
Romania, the share of railway suicide as a proportion of all suicides is below 1 %. This may be due to the
relatively small railway network and low density of population.

Suicide and trespasser fatalities (EU-27)

4000

3500
852 797
855 750
3000
928

2500
940
2000

1500 2868
2622 2773 2743
2422
1000 1874

500

0
2006 2007 2008 2009 2010 2011

Nr of suicides Unauthorised persons

Figure 20: Suicide and unauthorized person fatalities (EU 27 in 2006-2011)

Railway suicides, and more generally, all suicides are relatively well correlated with unemployment rates. This
may partly explain an increase in railway suicide fatalities in 2007 and 2009, visible at EU level. This
correlation is however not traceable anymore at country level.

Railway suicides are relatively common on EU railways: there were 700 suicides per billion train-km on
average in the past three years. The railway suicide rate is highest in the Netherlands, followed by Portugal,
Hungary and the Czech Republic. In these four countries, the suicide rate is above one suicide per million
train kilometres.

Page | 20
 Suicide fatalities per million train-km (2009-2011)
1.4

1.2

0.8

0.6

0.4

0.2

Figure 21: Suicide rates: number of suicide fatalities per million train kilometres (EU 27 in 2009-2011)

The costs of delays due to suicides (and trespasser fatalities) represent a significant share of total costs of
delays incurred to railway undertakings. It typically takes up to 2 hours to open a railway line when a person
is struck by a train. This is a significantly longer time compared to delays caused by some technical failures.

Page | 21
Safety and market opening
Setting up a common safety regulatory framework for the EU is a part of the wider EU policy to create more
efficient railway market in Europe. Gradual opening of national railway markets carries inherent safety risks
that are being addressed by the provisions of EU safety legislation. The two figures below seek to shed light
on the impact of market opening on safety at country level.

The measurement of the degrees of market opening is relatively complex and any benchmark proposed
would inevitably have its limitation. However, one measurement index, the rail liberalization index, has
recently gained recognition and acceptance by railway community. The index has been produced by IBM
12
Global Business Services ( ) since 2002. It reflects legal and de facto barriers to market access from the
perspective of an external railway undertaking seeking access. It reflects the market shares of external RUs
active in addition to the incumbent as a practical consequence of existing barriers to open market.
st
The fourth edition of the index published in 2011 benchmark the rail market opening of MSs as of 1 January
2011 and ranks countries from those most advanced to those delayed in terms of rail market opening. Three
categories of countries are considered based on the value of the index: six countries are considered as
delayed, 15 countries on schedule and six countries advanced, in terms of rail market opening.

The IBM rail liberalization index is plotted against the casualty risk for passengers and employees in Figure 22.
While it is impossible to find a correlation between the two variables, it appears that countries from
advanced group have lower casualty risk than the countries in the two other groups.

HU BG

PL

RO

EL
SI SK

LV PT
EE IT
CZ
AT SE
ES IT
DE
FI DK
IE FR NO
LU NL
LT UK

Figure 22: Fatality risk versus rail liberalization index (IBM 2011) for EU countries

12
( ) Rail liberalization index 2011, IBM Germany with Prof. Kirschner, Deutsche Bahn, 2011

Page | 22
A separate look at the development of railway safety in the advanced group of countries as compared to all
other countries confirms that there are no significant differences in trends in casualty risk for passengers and
employees in countries with a higher liberalization index.

Railway casulties vs Rail Liberalization Index (IBM)

1.2
Casualties (fatalities and serious injuries)

0.8

0.6

0.4

0.2

0
2006 2007 2008 2009 2010 2011
Advanced Others

Figure 23: Fatality risk versus rail liberalization index (IBM 2011) for EU countries

Page | 23
Accident costs
The data on the cost of accidents show a wide variation over time and between countries. It is also evident
that Member States continue to have problems in establishing reporting regimes for this set of CSIs.

The economic impact of significant accidents in 2011 is shown in Figure 24. It has five components: Costs of
fatalities, cost of injuries, costs of material damage, costs of damage to the environment and costs of delays.
While the first two components are available for all countries, the number of countries providing information
on the costs of damage to infrastructure, to the environment and delays is limited.

Societal costs of casualties represent the majority of costs of significant accidents. For countries that reported
costs for all five categories of costs is their share 73 % (Austria) and higher.

By adding together the costs of fatalities and of serious injuries, we obtain a value exceeding EUR 2.5 billion,
which gives a broad idea of the overall economic burden of rail casualties in 2011. Other reported costs of
accidents for all EU countries account for little more than EUR 200 million.

Figure 24: Economic impact of significant accidents in 2011 (in million EUR)

The significant accident costs of material damage to rolling stock and infrastructure per train-km are relatively
high in Norway, Austria, the Netherlands and Poland (more than EUR 100 per 1 000 train-km).

Page | 24
Safety of infrastructure
Three CSIs concern railway infrastructure, the first is a measure of the coverage of automatic train protection
(ATP) systems on the lines (Figure 25); the second is the number of level crossings (Figure 26), normalised by
the length of the network expressed in track kilometres; and the third gives information on the type of
protection at level crossings (Figure 27).
13
ATP ( ) systems is widely considered to be the most effective railway safety measure that infrastructure
14
managers can implement to reduce the risk of collisions on mainline railways ( ). A relatively high density of
train protection is typical in countries with high traffic density such as the Netherlands, Italy and Germany.
This can be seen in Figure 25.

The percentage of tracks equipped with an ATP system has seen a marked increase of 2 % from 2010 to
15
2011 ( ). This was largely driven by progress in ATP implementation achieved by France, Latvia, Sweden and
Norway. However, the data seem to be reported in an inconsistent manner across the EU, reducing their
comparability.

% of tracks witih ATP in operation

100

90

80

70

60

50

40

30

20

10

0
AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK
2009 67.9 11.0 11.0 100. 17.0 90.0 22.8 0.0 87.4 83.0 47.2 5.0 92.4 34.9 100. 46.4 99.0 74.0 0.0 51.3 48.3 64.0 67.0 18.5 4.2
2010 72.6 11.0 11.0 100. 0.0 93.2 53.0 23.6 0.0 88.4 82.0 52.3 5.0 100. 37.9 100. 47.7 98.0 70.0 0.0 52.6 97.8 65.5 67.0 21.0 4.2
2011 82.0 11.0 11.0 100. 94.3 53.0 23.7 0.0 86.0 82.0 52.3 54.0 5.0 100. 37.3 100. 66.0 96.0 74.0 0.0 58.6 97.5 81.2 67.0 21.0 4.2

Figure 25: Percentage of tracks equipped with automatic train protection (2009–2011)

There were 114 615 level crossings in the EU countries in 2011. On average, there are five level crossings per
16
10 line-km in the EU; only 24 % of them are active level crossings with user-side protection ( ). Sweden,
Austria, the Czech Republic, the Netherlands and Norway have the highest density of level crossings in terms
of level crossings per line-kilometre. Of these, the Netherlands has the highest ratio of active level crossings
to all level crossings. A low ratio of active level crossings to all level crossings is typical for the less densely
populated countries (Figure 26). Spain has the lowest average number of level crossings per line-kilometre:
there is one level crossing per 5 line-km.

13
( ) Automatic train protection (ATP) means a system that enforces obedience to signals and speed restrictions by speed
supervision, including automatic stop at signals. Systems where track signalling information is substituted and/or
supplemented by cab signalling are included. The part of the definition relating to ‘automatic stop at signals’ is
intended to include also automatic stops at conflict points between clearance gauges.
14
( ) Interfleet (2011). Investigating the links between historic accident rate reduction and the underlying changes, Report
prepared for ERA in 2011. Report can be downloaded from the ERA website.
15
( ) Estimate for EU-27 countries excluding France and Denmark.
16
( ) Protection is typically provided by arm barriers.

Page | 25
 Number of level crossings (LCs) per 100 line kilometers (2011)
120
Average niumber of level crossings (LCs) per 10 line kilometers

100

80

60

40

20

0
AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK
Passive LCs per line-km 55 8 4 0 49 21 18 19 19 11 51 12 36 50 7 9 12 8 21 0 44 23 20 73 45 0 32
Active LCs per line-km 33 46 17 0 40 31 22 18 32 7 14 31 33 13 27 22 39 24 67 88 27 16 9 30 28 0 11

Figure 26: Number of active and passive level crossings per 100 line-km in 2011

Detailed statistics are available on the type of active level crossings at European level. In Figure 27, the data
17
for EU countries ( ) show that level crossings with automatic user-side protection and warning (barriers with
lights) (34 %) are the most common type of active crossings (24 %), followed by the level crossings with user-
side warning (11 %). Passive (unprotected) level crossings represent 53 % of all level crossings in the EU.

Level crossings per type (EU-27 excluding FR and DK in 2011)

with automatic user-side warning

11% 1%

with automatic user-side protection

with automatic user-side protection and

23% warning
53%
with automatic user-side protection and
warning, and rail-side protection

4% with manual user-side warning and/or

protection
8%
Passive level crossings

Figure 27: Breakdown of active level crossings according to the level of protection in 2011 (EU countries)

17
( ) EU-27 countries excluding Denmark and France

Page | 26
Traffic volumes
The number of train-kilometres continued to rise in 2011, with a 2 % annual increase at the EU level. The
number of freight train-km increased by 6 % between 2010 and 2011. At the same time, the number of
passenger-kilometres reported in 2011 is similar to that of 2009 and 2010: slightly less than 400 billion train-
km. The average number of passengers per train was 122 in Europe in 2011; i.e. the ratio of number of
passenger-kilometres to passenger train-kilometres.

Train kilometres (million)

1 200

1 000

800

600

400

200

0
AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK
2009 152.3 91.87 31.49 5.652 163.1 1002. 82.15 6.820 19.61 188.1 50.01 504.0 106.2 18.18 350.5 14.05 8.063 18.72 132.0 43.27 208.6 40.58 88.50 143.0 18.20 44.95 568.5
2010 156.1 98.00 30.63 5.714 160.1 1032. 83.08 8.934 16.96 186.7 51.00 484.7 97.40 17.69 323.9 14.13 8.161 16.62 146.2 46.46 219.0 40.00 93.52 141.3 18.84 47.53 520.0
2011 152.2 100.5 31.24 5.600 160.6 1063. 84.78 7.000 12.53 191.3 51.07 501.5 110.2 18.05 317.3 15.32 8.861 18.47 160.5 45.85 227.3 37.20 104.2 140.3 20.33 45.38 528.4

Figure 28: Number of million train-kilometres (2009–2011)

Germany is the country with the highest number of train-kilometres, accounting alone for one quarter of all
train-kilometres in the EU.

Train-km in 2011 / Train-km in 2007

1.20

1.00

0.80

0.60

0.40

0.20

0.00

Figure 29: Relative change in the number of train-kilometres between 2007 and 2011 (Train-km in 2011 /
Train-km in 2007)

Page | 27
Looking at figures for passenger train-kilometres and freight train kilometres separately allows identification
of countries with important share of freight train traffic (three Baltic countries, Slovenia and Poland). At the
level of Union, passenger traffic represented 80 % of all train-kilometres in 2011. The share of passenger
train-kilometres exceeded 90 % in Ireland, Denmark, Greece, the UK and the Netherlands (Figure 30).

Passenger train-km among all train kilometres (2011)

100%

90%
EU (80 %)
80%

70%

60%

50%

40%

30%

20%

10%

0%

Figure 30: Percentage of passenger train-kilometres among all train-kilometres in 2010 in all countries

Four countries with the highest passenger volumes (Germany, France, Italy and the UK) together account for
two thirds of all passenger-kilometres. In two of them (Germany and the UK), have passenger volumes been
increasing over the past three years, as shown in Figure 31.

Passenger kilometres (million)

85 035

90 000
83 702

83 260
82 750
81 750
81 612

80 000

70 000
56 059
55 831

60 000
52 765
46 426

50 000
43 474

2009
41 326

2010
40 000
2011

30 000
21 729
21 399
20 978

18 577
18 049
17 800
16 892
16 800
16 621

20 000
11 434
11 216
11 036
10 900
10 700
10 500

10 493
10 493
9 494

7 945
7 795
7 666
6 889
6 750

6 587
6 553
6 472

6 389

6 177

10 000
5 500
5 141
4 152
4 143
4 111
3 959
3 882
3 876

3 153
3 036
2 996

2 428
2 291
2 247
2 144
2 100
2 067

1 681
1 678
1 639
1 414
1 144
958

840
773
747
733
506

393

389

813
357

354
333

741
232
497

373

350

0
AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK

Figure 31: Number of million passenger-kilometres (2009–2011)

Page | 28
Railway transport operation efficiency
The operational efficiency of railway transport can be implicitly expressed by a simple ratio of passenger
kilometres per passenger train kilometres and freight tonne kilometres per freight train kilometres.

Passenger trains
The theoretical average passenger train load in the EU-27 was 122 passengers in 2011 (1 % increase
compared to 2010). The average passenger train load appears to be primarily a function of country size;
Romania, Estonia and Latvia are an exception to this rule. Given the limits of international passenger traffic
across Europe, the share of domestic long-distance services is the most important factor here.

Passenger kilometers per passenger train-km (2011)

200

180

160

140

120

100

80

60

40

20

0
FR IT PT ES EE PL SE LV NL UK BE FI AT DE IE HU BG NO DK EL SK LT SI RO CZ LU

Figure 32: Passenger kilometres per passenger train kilometres in 2011

In case of passenger rail traffic, France together with Italy has the highest ratio of theoretical average train
occupancy, with more than 150 passengers travelling on board of an average passenger train. The theoretical
average train occupancy is lowest in Luxembourg, Czech Republic and Romania. The ratio is a function of
prevalent types of services operated in the given country, as well as of its geographical aspects. Yet, different
positions of countries with huge number of similarities (Baltic countries, Nordic countries) in the ranking
presented in Figure 32 is surprising and may be pointing to differences in operational arrangements and
minimum public service strategies.

Freight trains
The average load ratio for freight trains on European railways gives only a very limited idea about the
efficiency of freight train operations, since it is a function of the load mix transported. Raw materials
represent a high proportion of transported goods in Baltic countries and in some Central European countries,
leading to a relatively high theoretical average load ratio (Figure 33). A high degree of interoperability of
railway systems of Baltic countries and the Russian one contributes to their relative outperformance.

Page | 29
 Freight tonne kilometers per freight train-km (2011)
3500

3000

2500

2000

1500

1000

500

0
EE LV LT SK HU CZ RO LU PL UK FI DK CT SE AT NL IT NO FR SI DE EL PT ES IE BG BE

Figure 33: Freight tonne kilometres per freight train kilometres in 2011

Page | 30
Independent accident investigation
Every year, more than 200 of accidents and incidents are investigated by NIBs of MSs. This number has been
slightly increasing in time in recent years, since more and more NIBs decide to open an investigation into
other than serious accidents. The number of serious railway accidents investigated by NIBs has been stable
since 2007 with some 40 serious accidents into which a NIB investigation started (Figure 34).

Serious railway accidents investigated by NIBs with resulting casualties

250

200

150

100

50 38
42 42 42
34 30 36

0
2006 2007 2008 2009 2010 2011 2012
Fatalities Serious Injuries Accidents

Figure 34: Serious railway accidents investigated by NIBs together with the resulting casualties (EU-27)

A detailed look into the type of serious accidents investigated by NIBs shows that level crossing accident is
the most commonly investigated type of serious accidents, followed by train derailment and train collisions
(Figure 35).

Serious accidents investigated by NIBs per type of accident

45

40 Other

35 Fire in RS

30
LC accident
25
Accident to person
20

15 Train derailment

10 Train collision with

obstacle
5
Trains collision
0
2006 2007 2008 2009 2010 2011 2012

Figure 35: Serious accidents investigated by NIBs per type of accident (EU-27 countries in 2006-2012)

Page | 31
Not all significant accidents have to be investigated by NIBs. Serious accidents that must be investigated by
18
NIBs represent a fraction of significant accidents . In addition to serious accidents, the NIBs sometimes
investigate accidents and incidents which under slightly different conditions might have led to serious
accidents. This is however not a common practice in all MSs, in particular when it comes to incidents and
other minor accidents.

Each year, the NIBs notify the Agency that they have opened about 250 investigations into serious accidents
and other accidents and incidents. The NSAs report about 2 500 significant accidents a year (Table 10).
Among those occurrences investigated by the NIB, only about 12 % of investigated occurrences were serious
accidents as referred to in Art. 19(1) of the RSD. As the railway undertakings (RUs) and infrastructure
managers (IMs) should normally investigate all serious accidents as part of their safety management systems
(SMSs), those accidents which have not been investigated by the NIBs, will be investigated by railway
operators.

National safety authorities (NSAs) National investigation bodies (NIBs)

Year of reporting
Significant accidents Precursors Notifications of opened investigations
2009 3 027 9 565 177
2010 2 401 10 712 247
2011 2 342 9 893 368
Table 10: Number of events reported to the ERA in the period 2009–2011 (EU-27)

The share of significant accidents (reported under CSIs) as a proportion of all investigated accidents per type
is showed in Figure 36, which unveils that train derailments are the preferred type of accident into which the
NIBs decide to open an investigation, with 46 % of these accident types being investigated by NIBs. Less than
one third of significant train collisions and fires in rolling stock are investigated by NIBs of Member States.
Only 8 % of level crossing accidents are subject to independent investigation. This may appear surprising
since in these accidents, the IMs responsibility for managing the risks of the infrastructure is likely to be
relevant. Establishing causes of accidents to persons caused by rolling stock in motion is usually
straightforward; the investigation into this type of accident is typically limited and carried out by the
operators in cooperation with judicial authorities.

Share of significant accidents investigated (2009-2011)

5000
1%
4500
4000
3500
3000
2500
8% 4461
2000
1500
1000 1827
29% 46%
27%
500
247 202
85 175 158
0 31 99 53
Fires in rolling stock Collisions of trains Derailments of trains Level-crossing Accidents to persons
accidents caused by rolling
stock in motion
Investigated Significant accidents not investigated

Figure 36: Serious railway accidents investigated by NIBs together with the resulting casualties (EU-27)

18
( ) Collision or derailment of trains resulting in the death of at least one person or serious injuries to five or more persons
or damage above EUR 2 million and any other similar accidents – see Background information for RSD definition.

Page | 32
The overview of the number of investigations carried out by NIBs in Europe is showed in Figure 37. There was
a drop in the number of investigated occurrences that occurred in 2009; since then the number of
occurrences investigated by NIBs across Europe has been rising. The figure also shows that over the past four
years, the percentage share of investigation that were closed during the year following the occurrence has
been rather stable, at about 70 %.
250
Status of investigation
by year of occurrence:

29
200 6 4 77
57 2012 Open
13 12 31
44 2012 Completed
11 7
13 12 52 48
28 23 2011 Open
150 42
77 54 2011 Completed

2010 Open
180 182
106
169 2010 Completed
100 195 197 2009 Open
190
136 176 177 146
168 123 2009 Completed
158 159 162 166 162 161
153
143 148
136 140 2008 Open
128
116
50 103 2008 Completed

2007 Open
56
2007 Completed
28 30 27
22 23
15 2006 Open
0
Status end of 2007 08 09 10 11 12 2007 08 09 10 11 12 2008 09 10 11 12 2009 10 11 12 2010 11 12 2011 12 2012 2006 Completed

Occurrences during 2006 2007 2008 2009 2010 2011 2012

Figure 37: NIB investigation carried out since 2006 with the status of investigation (ERAIL database)

The average number of days between the accident occurrence and the notification on starting an
19
investigation to the Agency has been decreasing over time: It now stands at less than 50 days ( ) in average.
Despite an improvement recorded over time, an important proportion of started investigation is not notified
20
to the Agency within 10 days after the decision has been taken on the investigation to start ( ).

The final investigation reports on the investigations carried out by NIBs should be made public as soon as
possible, and normally not later than one year after the date of the occurrence. The average number of
months before the final report is submitted to the Agency has also been decreasing over time: from more
than 15 months for accidents occurring in 2008 to around 10 months for accidents occurring in 2011.

Year of occurrence / average number of

2008 2009 2010 2011 2012
days between occurrence and
Notification 91 60 49 50 35*
Final investigation report 460 402 377 313 -
Table 11: Average time span between occurrence and accident notification and between occurrence and
the submission of the final investigation report to the Agency (in days)

The list of serious accidents (collisions and derailments) that occurred in 2012 is shown together with basic
information in Annex 2. The status of the investigation and the reference in ERAIL database of investigated
occurrences is provided.

19
( ) Estimate based on the assumption that the time-span between the occurrence and the decision to start an
independent investigation by NIB is usually short (single days).
20
( ) RSD, Art. 24(1): “Within one week after the decision to open an investigation the investigating body shall inform the
Agency thereof.”

Page | 33
Background information
The report Railway safety performance in the European Union summarises information on the development
of railway safety in Europe. The primary purpose is to provide safety intelligence and information on risks to
EU policy-making bodies, NSAs and NIBs, and to the general public. The report reviews the performance
levels achieved during 2011 across a number of topic areas. It includes basic statistical analyses on a wide
range of safety performance indicators and highlights significant findings.

The report is based on the common safety indicators (CSIs) data reported to the ERA by 5 November 2012.
Any changes after that date have not been taken into account. Information presented on serious accidents
and their investigations is based on reports available to the ERA on 4 March 2013. Any event occurring after
that day is not covered by this report. This report covers the railways in 25 of the 27 EU countries; Cyprus and
Malta do not have railway systems that are covered by EU legislation. These 25 Member States are referred
to as ‘Member States‘, ‘EU’, or ‘EU countries’ in the report. The Channel Tunnel (CT) is a separate reporting
entity, so that relevant data are given separately to the French and UK data. The data are also reported by
Norway. Therefore, there were a total of 27 reporting entities in 2011; the term ‘Europe’ was sometimes
used for this complete group in the report.

European legislation requires Member States to report to the ERA on significant accidents and serious
accidents occurring on their territory. The NSAs must report all significant accidents. The NIBs must
investigate all serious accidents, notify the ERA of these investigations and, when closed, send the
investigation report to the ERA. The term significant accident covers a wider range of events than serious
accidents. The legislation provides the following definitions for these two groups of accident:

Significant accident Serious accident

Directive 2004/49/EC, Commission Directive Directive 2004/49/EC
2009/149/EC and Regulation (EC) No 91/2003
‘significant accident’ means any accident involving 'serious accident' means any train collision or
at least one rail vehicle in motion, resulting in at derailment of trains, resulting in the death of at
least one killed or seriously injured person, or in least one person or serious injuries to five or more
significant damage to stock, track, other persons or extensive damage to rolling stock, the
installations or environment, or extensive infrastructure or the environment, and any other
disruptions to traffic. Accidents in workshops, similar accident with an obvious impact on railway
21
warehouses and depots are excluded ( ). safety regulation or the management of safety;
Significant damage is damage that is equivalent to 'extensive damage' means damage that can
EUR 150 000 or more. immediately be assessed by the investigating body
22
to cost at least EUR 2 million in total ( ).
Reporting of CSIs by NSAs Accident investigation by NIBs
Each year the safety authority shall publish an Within one week after the decision to open an
annual report concerning its activities in the investigation the investigating body shall inform the
preceding year and send it to the Agency by 30 Agency thereof. The investigating body shall send
September at the latest. The report shall contain the Agency a copy of the final report normally not
information on: later than 12 months after the date of the
24
the development of railway safety, including an occurrence ( ).
aggregation at Member State level of the CSIs laid
23
down in Annex I ( )
Table 12: Accidents reported to the ERA according to the EU legislation

21
( ) Appendix to Annex I to the RSD, Article 1.1.
22
( ) Article 3(l) of the RSD.
23
( ) Article 18 of the RSD.
24
( ) Article 24 of the RSD.

Page | 34
The current legislative framework does not require Member States to collect information on all railway
accidents. The reporting is often limited to significant accidents and a selection of other events. Data on
incidents are not necessarily collected by RUs/IMs and the NSAs do usually rely on accident data when
planning their supervision activities.

SERIOUS
ACCIDENTS SIGNIFICANT
RSD Art.19 ACCIDENTS
RSD Annex I
ACCIDENTS

NON-SIGNIFICANT
ACCIDENTS

INCIDENTS
NIB Investigations
CSI REPORTING
(EU level)
(HAZARDOUS) EVENTS

Figure 38: Pyramidal model for railway safety management

Moreover, the information about less serious accidents and incidents are not systematically collected at the
EU level; some Member States do not have such a database as well. This may represent an obstacle to
efficient learning and early identification of recurring safety issues in EU railway system.

There are certain limitations in respect to the current EU railway safety monitoring approach. It relies
exclusively on outcome indicators such as number of accidents and resulting casualties. With these indicators
moving ultimately towards zero at country level, the approach has severe limitations in terms of reactivity
and capability to capture underlying raising safety issues.

Figure 39: Pyramidal model for railway safety management

Not all safety performance indicators are covered by the CSIs; so new indicators may be introduced in the
future. Similarly, there is no common approach towards the measurement of a risk regulation regime at
Member State level. Indicators reflecting corresponding levels of the pyramid may be developed in the
future. Not all of them are traditional quantitative indicators such as CSIs or key performance indicators
(KPIs). They may lead to a new approach to the assessment of railway safety management systems at both
Member State and EU levels.

Page | 35
Annexes
Annex 1 — Common safety indicators

List of CSI data tables

Table Nr Name
1 Fatalities by category of person
2 Serious injuries by category of person
3 Number of accidents by type of accidents
4 Number of accidents involving at least one railway vehicle transporting dangerous goods
5 Number of suicide
6 Traffic and infrastructure data

For full set of CSI data, please go to http://erail.era.europa.eu

Legend
Natural variation

Natural variation due to a single accident

Change of definition or reporting procedure

Unknown reason for variation

Further detailed explanation available

Table 1 Fatalities by category of persons
Year AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK Total EU
Passengers 2008 2 2 12 0 13 1 0 0 1 5 0 10 10 0 4 0 0 1 0 8 3 15 0 0 2 0 89 89
2009 1 2 1 0 1 3 0 0 0 2 0 7 0 0 5 0 0 1 0 0 8 0 4 0 0 2 0 37 37
2010 0 18 0 0 2 0 0 0 1 15 0 2 3 0 7 0 0 0 0 0 7 1 4 2 0 0 0 62 62
2011 0 0 1 0 5 9 0 0 0 2 0 7 3 0 0 0 0 0 0 0 10 0 0 0 0 1 0 38 38
Employees 2008 2 1 1 0 4 8 0 0 2 1 0 2 1 0 5 2 2 0 0 1 1 4 0 0 0 1 38 38
2009 0 1 1 0 0 4 1 2 1 0 1 1 1 0 5 1 1 1 1 0 1 1 3 0 1 0 1 29 29
2010 0 1 2 0 5 8 0 1 0 3 1 1 2 0 5 0 0 0 0 0 6 1 4 2 0 2 0 44 44
2011 2 2 1 0 3 10 0 0 0 0 1 2 1 0 1 0 0 0 0 0 2 0 2 2 0 0 0 29 29
Level crossing users 2008 17 10 4 0 24 50 3 1 6 15 8 38 42 1 6 6 6 18 0 39 15 38 4 4 11 14 380 380
2009 12 8 4 0 21 41 3 3 13 16 11 36 28 0 5 8 1 2 13 2 72 17 40 6 7 25 13 407 405
2010 13 9 8 0 34 44 4 2 12 9 8 29 30 2 11 5 0 5 8 3 54 11 35 7 6 9 4 362 359
2011 21 8 2 0 17 28 0 3 5 8 2 29 27 0 15 6 0 2 10 1 60 4 22 7 1 11 6 295 294
Unauthorised
persons 2008 18 8 27 0 3 78 8 7 8 23 13 40 62 2 49 32 15 1 1 260 23 151 9 9 41 41 929 928
2009 19 5 22 0 4 103 11 5 8 13 2 31 63 1 36 24 0 8 0 3 284 14 103 13 3 44 36 855 852
2010 17 5 6 0 6 80 6 9 16 10 4 37 47 1 48 26 0 13 0 3 216 9 96 31 7 44 16 753 750
2011 12 15 33 0 4 82 6 6 8 15 2 50 53 1 49 20 0 8 3 4 244 10 76 15 3 37 45 801 797
Other persons 2008 0 0 0 0 0 27 1 0 0 2 0 4 0 0 0 0 6 0 0 0 0 0 0 0 2 2 44 44
2009 2 0 0 0 0 19 0 0 0 0 0 1 0 0 30 0 1 5 0 1 0 0 0 0 0 1 3 63 62
2010 0 11 0 0 1 14 0 0 0 0 0 0 0 0 0 0 0 4 2 3 0 0 0 0 1 3 5 44 41
2011 0 2 0 0 0 11 0 0 0 0 0 0 0 0 0 0 0 3 1 0 4 0 0 0 0 0 4 25 25
Total persons 2008 39 21 44 0 44 164 12 8 17 46 21 94 115 3 64 40 29 20 1 308 42 208 13 13 56 58 1480 1479
2009 34 16 28 0 26 170 15 10 22 31 14 76 92 1 81 33 3 17 14 6 365 32 150 19 11 72 53 1391 1385
2010 30 44 16 0 48 146 10 12 29 37 13 69 82 3 71 31 0 22 10 9 283 22 139 42 14 58 25 1265 1256
2011 35 27 37 0 29 140 6 9 13 25 5 88 84 1 65 26 0 13 14 5 320 14 100 24 4 49 55 1188 1183
Table 2 Serious injuries by category of persons
Years AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK Total EU
Passengers 2008 6 36 8 1 40 30 3 0 9 3 0 14 28 0 5 0 2 0 1 44 6 26 3 11 5 1 282 281
2009 9 10 10 1 9 13 5 0 0 2 0 14 43 1 35 0 0 2 1 1 49 4 20 2 2 0 2 235 234
2010 4 171 0 0 14 8 1 0 3 14 0 12 29 0 6 0 0 0 5 0 35 3 14 10 1 17 7 354 354
2011 5 3 14 0 13 33 3 0 0 5 3 14 32 1 4 0 0 0 0 1 58 2 15 2 5 4 1 218 217
Employees 2008 12 28 2 0 4 33 2 0 2 1 3 4 1 0 4 1 2 1 0 5 2 7 1 10 1 5 131 131
2009 9 1 0 0 2 18 2 0 3 0 0 3 0 0 7 0 0 2 1 1 9 2 2 4 2 0 3 71 70
2010 14 4 1 0 3 18 0 0 3 2 0 4 1 0 5 0 0 0 0 0 10 2 12 5 0 3 6 93 93
2011 6 1 2 0 4 28 3 0 0 0 0 5 1 0 0 0 0 1 2 3 11 0 5 2 1 1 1 77 74
Level crossing users 2008 23 16 6 0 42 32 2 0 12 2 1 14 16 0 8 4 7 5 0 113 10 74 1 0 15 5 408 408
2009 27 6 1 0 33 22 0 1 16 4 3 22 11 0 0 3 0 4 4 0 50 5 50 7 7 14 2 292 292
2010 23 4 9 0 45 32 4 10 8 3 3 17 22 0 3 2 0 5 1 1 52 3 65 5 7 2 2 328 327
2011 25 9 8 0 21 25 3 7 6 3 3 9 21 0 1 2 0 5 3 1 46 3 50 3 5 14 1 274 273

Unauthorised 2008 12 2 22 0 52 38 2 5 6 13 2 6 15 0 21 8 15 0 0 111 20 126 1 20 15 6 518 518

persons 2009 9 1 11 0 48 39 7 6 3 5 7 21 30 0 16 9 0 1 0 1 89 7 115 2 3 20 7 457 456
2010 8 1 12 0 45 29 3 0 6 8 5 11 18 0 18 13 0 7 3 0 91 8 91 5 4 17 6 409 409
2011 7 4 18 0 36 41 3 0 8 2 2 23 22 0 29 9 1 13 3 0 93 5 81 8 1 20 4 433 433
Other persons 2008 0 1 0 0 1 23 0 0 0 0 0 2 0 1 0 0 5 0 0 4 1 0 0 0 2 4 44 44
2009 6 0 0 0 0 26 1 0 0 1 0 1 0 0 13 0 0 3 3 1 2 0 0 0 0 1 1 59 58
2010 0 6 0 0 0 29 0 4 0 0 0 2 0 0 0 0 0 3 1 4 0 0 0 0 0 6 2 57 53
2011 2 5 0 0 0 20 1 0 0 0 0 2 0 0 0 0 0 2 0 0 1 0 0 0 0 0 2 35 35
Total persons 2008 53 83 38 1 139 156 9 5 29 19 6 40 60 1 38 13 31 6 1 277 39 233 6 41 38 21 1383 1382
2009 60 18 22 1 92 118 15 7 22 12 10 61 84 1 71 12 0 12 9 4 199 18 187 15 14 35 15 1114 1110
2010 49 186 22 0 107 116 8 14 20 27 8 46 70 0 32 15 0 15 10 5 188 16 182 25 12 45 23 1241 1236
2011 45 22 42 0 74 147 13 7 14 10 8 53 76 1 34 11 1 21 8 5 209 10 151 15 12 39 9 1037 1032
Table 3 Accidents by type of accidents
Accident types Years AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK Total EU
Collisions of trains 2008 3 94 3 0 5 13 0 0 1 4 0 97 1 0 2 1 1 2 6 8 0 0 4 4 12 8 269 263
2009 5 34 3 0 5 16 1 0 2 1 0 7 0 1 7 4 0 1 2 6 18 0 2 1 1 6 17 140 134
2010 3 5 2 0 3 13 1 1 4 2 0 15 1 0 2 0 1 1 5 9 4 2 10 3 0 13 8 108 99
2011 2 0 0 0 6 18 1 0 1 4 2 12 0 0 6 0 0 0 4 15 8 1 1 2 1 11 3 98 83
Derailments of trains 2008 7 21 0 0 2 12 0 2 2 15 1 97 1 1 10 1 0 1 3 105 3 1 14 0 6 14 319 316
2009 1 41 0 0 3 7 0 0 2 7 2 14 1 0 6 1 0 0 2 3 63 1 1 7 0 3 12 177 174
2010 2 2 1 0 3 19 1 0 2 7 1 14 1 0 3 1 0 0 3 4 17 3 0 7 0 2 6 99 95
2011 2 3 1 0 5 14 1 0 0 7 0 11 1 0 4 0 0 1 1 4 23 2 1 7 0 7 6 101 97
Level-crossing accidents 2008 36 56 9 0 53 76 5 12 17 18 9 115 44 1 16 19 10 21 0 278 20 86 6 41 63 23 1034 1034
2009 36 31 5 0 42 64 2 7 26 19 12 49 39 1 7 14 5 8 13 2 288 15 57 13 11 51 16 833 831
2010 33 17 10 0 57 73 9 17 16 11 9 36 42 2 15 6 2 10 9 3 86 14 58 14 16 50 7 622 619
2011 43 16 7 0 34 56 2 15 8 8 5 40 38 0 18 6 0 8 14 2 86 7 43 7 6 50 11 530 528
Accidents to persons
2008 35 25 52 1 72 193 13 12 19 43 14 57 79 3 83 42 45 1 2 397 49 314 13 14 78 57 1713 1711
caused by rolling stock
in motion 2009 37 34 40 1 62 201 21 12 11 22 10 64 136 2 83 33 2 19 4 4 400 27 235 20 2 74 49 1605 1601
2010 29 15 20 0 61 166 10 13 17 24 10 64 96 1 80 37 0 27 5 3 341 22 190 38 4 116 34 1423 1420
2011 30 32 65 2 51 175 14 13 15 23 7 76 104 2 78 27 1 26 3 6 366 17 166 28 3 102 54 1486 1480
Fires in rolling stock 2008 0 24 1 1 1 6 0 0 1 0 0 24 0 0 2 5 0 0 3 9 0 0 3 0 8 0 88 85
2009 1 6 0 0 1 4 0 0 0 0 0 16 1 0 9 3 0 0 0 1 3 0 0 1 0 14 6 66 65
2010 0 0 0 1 0 2 0 0 0 0 0 6 0 0 0 0 0 0 1 1 0 0 2 0 0 9 2 24 23
2011 0 0 1 0 1 4 1 0 0 0 0 2 0 0 0 0 0 0 1 1 0 0 1 2 1 9 2 26 25
Other accidents 2008 16 0 0 0 0 29 4 0 0 0 3 63 30 0 3 0 5 1 0 92 1 10 6 6 50 2 321 321
2009 8 0 0 0 0 18 5 0 1 2 2 21 3 1 7 0 0 2 1 0 71 0 9 4 5 88 4 252 252
2010 12 1 6 0 1 24 1 0 0 1 3 20 2 0 3 0 0 3 1 0 1 1 11 7 1 41 5 145 145
2011 7 0 0 0 2 18 1 0 0 0 0 13 4 0 2 0 0 0 6 7 5 0 5 8 0 56 2 136 129
Total nr accidents 2008 97 220 65 2 133 329 22 26 40 80 27 453 155 5 116 68 61 26 14 889 73 411 46 65 217 104 3744 3730
2009 88 146 48 1 113 310 29 19 42 51 26 171 180 5 119 55 7 22 16 843 43 304 46 19 236 104 3043 3027
2010 79 40 39 1 125 297 22 31 39 45 23 155 142 3 103 44 3 41 24 20 449 42 271 69 21 231 62 2421 2401
2011 84 51 74 2 99 285 20 28 24 42 14 154 147 2 108 33 1 35 29 35 488 27 217 54 11 235 78 2377 2342
Table 4 Dangerous goods accidents
Year AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK Total EU
Accidents involving at 2010 0 2 0 0 0 4 2 0 0 1 0 0 0 0 5 3 0 2 0 0 32 1 2 0 0 0 0 54 54
least one railway vehicle
2011 0 0 0 1 0 3 2 0 0 4 0 2 0 0 0 3 0 2 3 0 6 0 1 0 0 0 1 28 28
transporting DG
Accidents involving at 2010 0 0 0 0 0 2 1 0 0 1 0 0 0 0 3 0 0 0 0 8 1 1 0 0 0 0 17 17
least one railway vehicle
transporting DG in which 2011 0 0 0 0 0 1 2 0 0 4 0 0 0 0 0 2 0 1 2 0 5 0 1 0 0 0 1 19 19
DG are NOT released
Accidents involving at 2010 0 2 0 0 0 2 1 0 0 0 0 0 0 5 0 0 2 0 0 24 0 1 0 0 0 0 37 37
least one railway vehicle
transporting DG in which 2011 0 0 0 1 0 2 0 0 0 0 0 2 0 0 0 1 0 1 1 0 1 0 0 0 0 0 0 9 9
DG ARE released

Table 5 Suicide fatalities

Years AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK Total EU
Total nr of suicides 2008 93 27 0 160 714 24 1 1 174 52 289 111 7 137 0 9 164 7 29 50 29 71 20 58 202 2429 2422
2009 101 69 19 0 185 875 32 0 3 163 62 337 139 2 111 2 4 10 197 8 25 69 25 67 10 56 210 2781 2773
2010 90 84 18 0 198 899 23 0 2 124 44 328 121 6 109 4 3 13 201 7 47 51 23 68 15 48 224 2750 2743
2011 87 98 27 0 235 853 26 0 4 128 64 332 155 6 140 5 7 10 215 11 28 42 76 62 25 40 203 2879 2868
Table 6 Traffic volumes and infrastructure
Year AT BE BG CT CZ DE DK EE EL ES FI FR HU IE IT LT LU LV NL NO PL PT RO SE SI SK UK EU
Train 2008 158 93 35 6 175 1044 82 7 21 193 53 541 109 20 367 16 20 139 47 224 42 96 138 20 49 549 4197
km (million)
2009 152 92 31 6 163 1003 82 7 20 188 50 504 106 18 351 14 8 19 132 43 209 41 89 143 18 45 569 4059
2010 156 98 31 6 160 1032 83 9 17 187 51 485 97 18 324 14 8 17 146 46 219 40 94 141 19 48 520 4019
2011 152 101 31 6 161 1063 85 7 13 191 51 502 110 18 317 15 9 18 161 46 227 37 104 140 20 45 528 4114
Passenger 2008 10.60 10.40 2.33 0.00 6.66 82.5 6.47 0.27 1.66 22.07 4.05 87.00 8.29 1.98 49.41 0.40 0.00 0.95 16.50 2.86 20.14 4.15 6.96 10.84 0.83 2.28 53.00 409.75
km (billion)
2009 10.50 10.49 2.14 0.00 6.47 81.6 6.39 0.23 1.41 21.73 3.88 83.26 7.95 1.68 46.43 0.36 0.33 0.75 16.80 3.00 18.58 4.15 6.18 11.22 0.84 2.25 52.76 398.38
2010 10.70 10.49 2.10 0.50 6.55 83.7 6.59 0.46 1.14 20.98 3.96 81.75 7.67 1.68 43.47 0.37 0.35 0.74 16.62 3.15 17.80 4.11 5.50 11.04 0.81 2.29 55.83 397.20
2011 10.90 9.49 2.07 0.51 6.75 85.0 6.89 0.39 0.96 21.40 3.88 82.75 7.80 1.64 41.33 0.39 0.35 0.73 16.89 3.04 18.05 4.14 5.14 11.43 0.77 2.43 56.06 398.18
Passenger 2010 105 81 23 1 123 777 79 5 16 158 35 410 85 17 280 5 7 6 134 35 146 32 71 94 11 33 485 3221
train km
(million) 2011 100 86 23 1 123 790 80 3 12 162 36 426 84 17 274 5 8 6 147 35 143 31 79 97 11 31 491 3266
Freight train 2010 45 16 7 0 37 255 4 4 1 26 16 75 12 0 44 9 1 10 12 9 73 8 23 41 8 15 35 775
km (million)
2011 44 15 7 0 38 273 4 4 1 26 15 76 26 0 43 10 1 12 13 9 80 7 25 43 9 14 38 825
Other train 2010 7 1 1 5 0 0 0 0 3 0 0 1 0 0 0 0 3 0 6 0 0 0 23
km (million)
2011 8 0 1 5 0 0 3 0 1 1 0 0 0 0 2 5 0 0 0 0 0 23
Freight 2010 22 0 0 0 34 107 3 14 0 0 10 33 0 0 20 13 1 17 6 4 49 3 11 23 3 0 19 389
tonne km
(billion) 2011 22 0 0 0 34 113 3 12 0 9 9 0 28 0 21 15 1 21 6 4 53 2 18 23 4 18 23 436
Line 2010 5807 3540 3973 108 9628 33803 3613 900 2552 13853 5919 42039 8657 1683 16794 1768 275 1897 3016 4114 20045 2842 17263 11066 1228 3622 15777 231668
kilometres
2011 5188 3587 3946 159 9614 33736 2650 918 2523 13965 5944 29297 7915 1683 16789 1768 275 1865 3035 4114 20066 2794 17220 11206 1209 3624 16187 217164
Track 2008 8197 6282 5116 159 11554 51851 3800 2133 3062 17960 8848 45951 10577 2110 25720 2180 4731 6700 4080 28673 3528 20348 16075 2192 4638 31534 323917
kilometres
2009 8154 6426 5154 159 11554 51780 3687 2166 3070 17972 8847 46007 10577 2141 26174 2182 614 3396 6868 28836 3528 15349 2187 4638 31571 303037
2010 8049 6344 5154 159 11554 63839 2167 3070 18967 8862 42039 10577 2165 24370 2148 614 3395 6830 4341 28743 3531 20171 15347 2187 4638 31631 326551

2011 7201 6344 5154 200 11554 63067 4094 2164 3041 19372 8885 42088 10577 2165 24377 2184 614 3998 7000 4341 28730 3483 20129 15601 2177 4641 31108 329948
Annex 2 — List of serious accidents – collisions and derailments - occurring in 2012 and notified to ERA

Date of Serious Available casualties

Title Occurrence type Fatalities Reference ERAIL
occurrence injuries details

Trains collision with an obstacle, 13/01/2012, Langenhorn Schl.; Trains collision with an 1 passenger fatality, 2
13/01/2012 1 2 DE-1329
Strecke Westerland - Elmshorn (Germany) obstacle passenger serious injury

Train derailment, 15.02.2012, On Vestfoldbanen, between 2 passenger serious

15/02/2012 Train derailment 0 5 NO-1359
Nykirke and Holmestrand station (Km 92,000) (Norway) injuries

Trains collision, 2012-03-03, Line 64 section Sprowa-Starzyny km

03/03/2012 Trains collision 16 2 13 passenger fatalities PL-1378
point. 21,150 electrified double-track line; track No 1 (Poland)

Trains collision with an obstacle, 06/04/2012, CH 29+500 Trains collision with an

06/04/2012 4 0 4 other fatalities EL-1434
Thessaloniki-Athens at Kryoneri (Greece) obstacle

Trains collision, 13/04/2012, Mühlheim (Main) - Hanau

13/04/2012 Trains collision 3 6 6 serious passenger injury DE-1395
(Germany)
Trains collision, 21-4-2012, Amsterdam Singelgracht Aansluiting -
Heavily used 6-way track between Amsterdam Central Station 1 passenger fatality, 22
21/04/2012 Trains collision 1 23 NL-1413
and Amsterdam Sloterdijk Station, equipped with switches and passenger serious injury
crossings. (The Netherlands)

Trains collision with an obstacle, 14/06/2012, Line 25 track A at Trains collision with an 1 other fatality, 1 other
14/06/2012 1 1 BE-1436
the end of a platform in the railway station of Duffel (Belgium) obstacle serious injury

Trains collision, 26.07.2012, Hosena (Germany) 26/07/2012 Trains collision 1 1 1 other fatality DE-0131
Annex 3 — List of national safety authorities and national investigation bodies

Code Country National safety authority National investigation body

BE Belgium Federale Overheidsdienst Mobiliteit en Vervoer Federale Overheidsdienst Mobiliteit en Vervoer

Directoraat-generaal vervoerte Land Onderzoeksorgaanvoor Ongevallen en Incidenten op het Spoor
Service Public federal Mobilité et Transports Service Public federal Mobilité et Transports
Direction générale Transport terrestre Organisme d’enquête sur les accidents et les incidents ferroviaires
http://www.mobilit.fgov.be http://www.mobilit.fgov.be/
BG Bulgaria Ministry of Transport — Railway Administration Executive Agency Ministry of Transport — Railway Accident Investigation Unit
www.iaja.government.bg http://www.mtitc.government.bg
CZ Czech Republic Drážní Úřad (DU) — Rail Authority Drážní inspekce (DI) — Rail Safety Inspection Office
http://www.ducr.cz http://www.dicr.cz/
DK Denmark Trafikstyrelsen Havarikommissonen for Civil Luftfart og Jernbane
http://www.trafikstyrelsen.dk http://www.havarikommissionen.dk
DE Germany Eisenbahn-Bundesamt (EBA) Bundesministerium für Verkehr, Bau und Stadtentwicklung
http://www.eba.bund.de Eisenbahn-Unfalluntersuchungsstelle
http://www.bmvbs.de
EE Estonia Tehnilise Järelevalve Amet Ohutusjuurdluse Keskus (OJK) — Safety Investigation Bureau
http://www.tja.ee http://www.ojk.ee
IE Ireland Railway Safety Commission Railway Accident Investigation Unit
http://www.rsc.ie http://www.raiu.ie
EI Greece Hellenic Ministry of Infrastructure, Transport and Networks Hellenic Ministry of Infrastructure, Transport and Networks
Department of Railway Safety Committee for Accident Investigation
http://www.yme.gr http://www.yme.gr
ES Spain Dirección General de Infraestructuras Ferroviarias Ministerio de Fomento
http://www.fomento.es Comision de Investigación de Accidentes ferroviarios
http://www.fomento.es
FR France Établissement public de sécurité ferroviaire (EPSF) Bureau d'Enquêtes sur les Accidents de Transport Terrestre
http://www.securite-ferroviaire.fr http://www.bea-tt.equipement.gouv.fr
IT Italy Agenzia Nazionale per la Sicurezza delle Ferrovie Railway Safety Commission
http://www.ansf.it http://www.mit.gov.it
LV Latvia Valsts dzelzceļa tehniskā inspekcija — State Railway Technical Inspectorate Transporta nelaimes gadījumu un incidentu izmeklēšanas birojs — Transport
http://www.vdzti.gov.lv Accident and Incident Investigation Bureau (TAIIB)
http://www.taiib.gov.lv
LT Lithuania Valstybinė geležinkelio inspekcija Katastrofų tyrimųvadovas
State Railway Inspectorate National Investigation Body
http://www.vgi.lt http://www.sumin.lt
LU Luxembourg Ministère du Développement durable et des Infrastructures Administration des Enquêtes Techniques
Administration des Chemins de Fer (ACF) http://www.mt.public.lu/transports/AET/
http://www.gouvernement.lu
HU Hungary Nemzeti Közlekedési Hatóság — National Transport Authority Közlekedésbiztonsági Szervezet — Transportation Safety Bureau
http://www.nkh.gov.hu http://www.kbsz.hu
NL Netherlands Inspectie Leefomgeving en Transport (ILT) The Dutch Safety Board
http://www.ilent.nl http://www.safetyboard.nl
AT Austria Bundesministerium für Verkehr, Innovation und Technologie Bundesanstalt für Verkehr (VERSA)
Oberste Eisenbahnbehörde Unfalluntersuchungstelle des Bundes, Fachbereich Schiene
http://www.bmvit.gv.at http://versa.bmvit.gv.at
PL Poland Urząd Transportu Kolejowego Państwowa Komisja Badania Wypadków Kolejowych (NIB)
http://www.utk.gov.pl http://www.mi.gov.pl
PT Portugal Instituto da Mobilidade e dos Transportes Terrestres Gabinete de Investigação de Segurança e de AcidentesFerroviários (GISAF)
http://www.imtt.pt http://www.iot.gov.pt (site under construction)
RO Romania Autoritatea Feroviară Română (AFER) — Romanian Railway Safety Authority Organismul de Invesigare Feroviar Român (OIFR) — Romanian Railway
http://www.afer.ro Investigating Body
http://www.afer.ro
SI Slovenia Javna agencija za železniški promet Republike Slovenije (AŽP) — Public Ministry of Transport
Agency of the Republic of Slovenia for Railway Transport Railway Accident and Incident Investigation Division
http://www.azp.si http://www.mzp.gov.si
SK Slovakia Úrad pre reguláciu železničnej dopravy (URZD) — Railway Regulatory Ministry of Transport Posts and Telecommunication
Authority http://www.telecom.gov.sk
http://www.urzd.sk
FI Finland Liikenteen turvallisuusvirasto — Finnish Transport Safety Agency (TraFi) Onnettomuustutkintakeskus — Accident Investigation Board of Finland
http://www.trafi.fi http://www.onnettomuustutkinta.fi
SE Sweden Transportstyrelsen — Swedish Transport Agency Statens haverikommission
http://www.transportstyrelsen.se http://www.havkom.se
UK United Kingdom Office of Rail Regulation (ORR) Rail Accident Investigation Branch
http://www.rail-reg.gov.uk http://www.raib.gov.uk/

NO Norway Statens Jernbanetilsyn (SJT) — Norwegian Railway Authority Statens havarikommisjon for Transport — Accident Investigation Board Norway
http://www.sjt.no (AIBN)
http://www.aibn.no
CT Channel Tunnel Channel Tunnel Intergovernmental Commission (IGC) See the relevant authority or body in France or United Kingdom for the respective
Commission intergouvernementale Tunnel sous la Manche part of the Channel Tunnel
http://www.channeltunneligc.co.uk
www.cigtunnelmanche.fr