Research Programme
Operations
The cost of level crossings - an international
benchmarking exercise
The Cost of Level
Crossings
International
Benchmark Report
Final Report to
Rail Safety and Standards Board
June 2006
Notice
This report was commissioned by Rail Safety and Standards
Board on terms specifically limiting the liability of Arthur D.
Little Limited. Our conclusions are the results of the exercise
of our best professional judgement, based in part upon
materials and information provided to us by Rail Safety and
Standards Board and others. Use of this report by any third
party for whatever purpose should not, and does not, absolve
such third party from using due diligence in verifying the
report's contents.
Any use which a third party makes of this document, or any
reliance on it, or decisions to be made based on it, are the
responsibility of such third party. Arthur D. Little Limited
accepts no duty of care or liability of any kind whatsoever to
any such third party, and no responsibility for damages, if any,
suffered by any third party as a result of decisions made, or
not made, or actions taken, or not taken, based on this
document.
Arthur D. Little Limited
Science Park, Milton Road
Cambridge CB4 0XL
United Kingdom
Telephone +44 (0)1223 392090
Fax +44 (0)1223 420021
www.adlittle. uk.com
© Copyright 2006 Rail Safety and Standards Board.
This publication may be reproduced free of charge for research, private study or for
internal circulation within an organisation. This is subject to it being reproduced and
referenced accurately and not being used in a misleading context. The material must be
acknowledged as the copyright of Rail Safety and Standards Board and the title of the
publication specified accordingly. For any other use of the material please apply to
RSSB's Head of Research and Development for permission. Any additional queries can
be directed to research@rssb.co.uk. This publication can be accessed via the RSSB
website www.rssb.co.uk
2
Table of Contents
Acknowledgements ....................................................................................................... 3
Executive Summary ...................................................................................................... 4
1. Introduction............................................................................................................. 9
2. Cost Benchmarking.............................................................................................. 11
3. Cost Drivers.......................................................................................................... 16
4. Comparative Safety Performance....................................................................... 23
5. Funding Mechanisms........................................................................................... 27
6. Impact of Cost Reduction - Case Study............................................................. 30
7. Conclusions.......................................................................................................... 35
8. Good Practice in Controlling Costs.................................................................... 39
Appendix: Participating Countries....................................................................... 46
1
List of Figures
Figure 1: Cost Comparison - Automatic Barrier Crossings......................................... 11
Figure 2: Variation in Cost............................................................................................ 12
Figure 3: Comparative Costs of Different Levels of Protection.................................... 14
Figure 4: Comparison of Costs and Cost Driver between different Countries ............ 17
Figure 5: Cost versus Safety Risk Index...................................................................... 24
Figure 6: Influence of Reducing Automatic Barrier to Manually Controlled Full
Barrier MCB Cost on Risk Reduction and Total Spend............................... 32
Figure 7: Influence of Reducing Automatic Open Crossings to Automatic Barrier Cost
on Risk Reduction and Total Spend.............................................................. 33
2
Acknowledgements
The authors would like to acknowledge the kind assistance of all those who participated
in this study, without whom this research would not have been possible, (in alphabetical
order):
•
AMEY Rail
•
Banverket
•
Danish State Railway
•
Department of Transport - Iowa
•
Deusche Bahn
•
EC Harris
•
GE Transportation
•
Her Majesty's Inspectorate of Railways
•
Irish Rail
•
Mitretek
•
National Railway Supplies
•
Network Rail
•
SPX Corporation
•
Swiss Federal Railways
•
Track Warning
•
Transport Canada
•
Vale of Rheidol Ry Railway
•
Volpe Institute
•
Westinghouse Rail Systems
3
Executive Summary
Background
Rail Safety and Standards Board (UK) commissioned research to understand level
crossing upgrade and conversion costs, and to find ways of improving overall safety
performance at all level crossings by identifying more cost-effective means of upgrade.
This research focused mainly on data for the UK railway, but significantly included a
review of costs across eight railways in other countries and interviews with
representatives from each of these railways.
This Report
This report provides a summary 'benchmark' of the costs reported by those countries
that participated in the study, and explores the factors that determine the cost of level
crossings. The aim of this benchmarking report is to share this knowledge to the benefit
of all railways that might be interested in managing the costs of their level crossing
investment programmes.
To preserve confidentiality, all data and findings presented in this report are presented
anonymously.
Findings - Overall Costs
•
The cost of automatic barrier crossings varies significantly between countries -
ranging from a national average of just over €100,000 to over €900,000.
•
There is significant variation in the technical scope and complexity of crossings
between different countries, which is a key factor in determining the cost.
•
The recorded costs of level crossings are also highly dependant on the scope of
work recorded within the project costs - there is no such thing as a 'standard' level
crossing upgrade or renewal.
•
Only three participating countries provide the most comprehensive type of level
crossing protection - manually controlled crossings - which are between 30% and
90% more expensive than automatic crossings.
•
The cost of 'non-materials' costs (design, installation, overheads, testing and
commissioning and project management) dominates the cost of level crossings,
rather than materials costs which only contribute between 20% and 30% of total
cost.
4
•
Countries with relatively high automatic crossing upgrade costs (and complex
technical scopes) are delivering comparatively good safety performance. However,
'good' safety performance is also being delivered by countries with relatively lower
cost automatic crossings; perhaps because of comparatively low levels of usage of
the crossing by road users, and low rail traffic densities.
•
It is unclear the extent to which the technical complexity of crossings in some
countries contributes to strong comparative safety performance, and therefore what
safety benefit is being delivered for the significant additional cost.
•
Many different funding models exist across the participating countries. These range
from the work entirely funded by the railway, to other models in which costs are
shared between the railway, the local community, road authorities and central
government.
•
Analysis suggests that reducing the unit costs of level crossing upgrades could
mean that more upgrades could be justified on the basis of reasonable practicability.
If cost of upgrades could be reduced significantly, it may be possible to deliver an
increased overall safety performance for a reduced total spend.
Conclusions - Key Cost Contributors
•
A key factor that determines the cost is the level crossing design complexity. The
consequence of higher complexity is increased cost of design, equipment,
installation, testing and commissioning.
•
Where designs are highly bespoke, design costs can be high (the average is 7% of
total project costs, but they can be as high as 20%). In comparison, countries which
employ more 'off the shelf' solutions, for example those associated with designs for
crossings that use predictors, have lower design costs.
•
'In-house' design teams are reported to help reduce design overheads, increase
retention of skills and knowledge, move towards more standard designs and to
create a more predictable flow of work to the industry.
•
Two forms of independent train detection, which requires both track circuit and
treadle, is not a requirement in most countries, but does add significantly to the cost
where it is used.
•
Remote monitoring of the status of crossings is also not a requirement in most
countries, but where it is required can add significantly to the cost (particularly for
geographically remote crossings as this is done by cable). Lower cost alternatives
include wireless (GSM) transmission of status, or the removal of the requirement to
monitor status altogether, (such as providing a free-phone service for the public to
report problems).
5
•
Buildings to house equipment that have large bases, where used, can add
significantly to the cost (in excess of €150K in some cases for a fully fitted and
installed building). 'Good practice' would suggest a greater use of lower cost
alternatives (smaller 'location cases') which have a substantially reduced footprint
and significantly lower cost.
•
On busy networks, weaknesses in the planning of possessions, and in the
management of procurement of equipment and testers can, in isolated cases, lead to
significant delays and associated increases in cost.
•
There is a common perception that where there are numerous contractual layers
associated with delivering a level crossing, overheads and profit add significantly to
costs. Some countries deliver work in larger quantities with fewer contractual
layers, and whilst we have been unable to compare the costs directly there is an
impression that this does lead to reduced overall cost.
•
Project management and overheads can contribute up to 20% of total cost; the
highest being associated with delays to the work schedule (for various reasons).
•
Assembly of level crossing components on site can lead to an increase in cost,
because of the relative difficulty of working near to a running railway and the
requirement for testing onsite rather than at component level in the factory.
•
A possible factor influencing cost is the extent to which level crossings are
delivered on a programme basis using more generic designs and controls. Although
the study has not uncovered significant supporting data, there is a widespread view
that delivering crossings in programmes with use of more generic designs helps to
reduce costs.
•
The volume of work in upgrading and renewing level crossings is reported to have
some impact on the unit cost of delivery. 'Turn-key' approaches to delivery are
thought to reduce cost by up to 20%.
6
Lessons Learnt - Controlling Costs
In overview, the research identified four main areas in which 'good practice' exists for
reducing cost:
•
Simplify Design
−
Modular components
−
Providing train detection by 'predictors'
−
Carry out design work within the infrastructure company
•
Risk Based Review of Current Requirements and Practices
−
Review the way in which level crossing equipment is housed
−
Review size and form of bases provided
−
Review need for monitoring crossing status by cable
−
Review need for bi-directional controls
•
Good Planning
−
Earlier planning of work to optimise track access requirements
−
Optimise procurement of scare resources (e.g. testing contractors)
•
Supplier Efficiency
−
Multi-skilled smaller teams
−
Economies of scale
7
Document Map
Chapter
Key Contents
1. Introduction
Background to study
2. Cost
Benchmarking
Overall comparison of costs
Cost of different levels of level crossing protection
3. Cost Drivers
Differences in technical scope of level crossings
Costs of project management and overheads
Equipment housing, cabling, barriers and lights
Accessing the railway to do work
Labour and installation
Testing and commissioning
4. Comparative
Safety
Performance
Relationship between cost of level crossings and safety
performance delivered
5. Funding
Mechanisms
Different mechanisms for funding level crossing programmes
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
6. Impact of Cost
Reduction
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
7. Conclusions
Summary of key findings and conclusions from study
8. Good Practice
In Controlling
Costs
Suggestions for controlling costs focusing on key cost drivers
identified
Suggestions for work breakdown structure
8
1. Introduction
1.1 Background
Rail Safety and Standards Board commissioned research to understand level crossing
upgrade and conversion costs, and to find ways of improving overall safety
performance at all level crossings by identifying more cost-effective means of upgrade.
This research focused mainly on data for the British railway, but significantly included
a review of costs across eight railways in other countries. The work involved
interviews with key stakeholders, including representatives of railways in each country,
collection of data on the cost of level crossings, and analysis of the data, in particular to
enable comparison of costs between different countries.
1.2 This Report
This report provides a summary 'benchmark' of the costs reported by those countries
that participated in the study, and explores the factors that determine the cost of level
crossings. The aim of this benchmarking report is to share knowledge on these aspects,
to the benefit of all railways that might be interested in managing the costs of their level
crossing investment programmes.
To preserve confidentiality, all data and findings presented in this report are presented
anonymously.
9
Document Map
Chapter
Key Contents
1. Introduction
Background to study
2. Cost
Benchmarking
Overall comparison of costs
Cost of different levels of level crossing protection
3. Cost Drivers
Differences in technical scope of level crossings
Costs of project management and overheads
Equipment housing, cabling, barriers and lights
Accessing the railway to do work
Labour and installation
Testing and commissioning
4. Comparative
Safety
Performance
Relationship between cost of level crossings and safety
performance delivered
5. Funding
Mechanisms
Different mechanisms for funding level crossing programmes
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
6. Impact of Cost
Reduction
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
7. Conclusions
Summary of key findings and conclusions from study
8. Good Practice
In Controlling
Costs
Suggestions for controlling costs focusing on key cost drivers
identified
Suggestions for work breakdown structure
10
2. Cost Benchmarking
The cost benchmarking reveals significant differences between participating
countries, with a near ten-fold difference between the highest and lowest costs.
True like-for-like comparison is, however, not possible due to differences in
technical scope and rail network characteristics.
This Chapter presents a 'benchmark' of the cost of level crossing upgrades of the nine
participating railways (hereafter, the nine countries are labelled 'A' to 'I', ordered by
highest cost to lowest cost)1.
2.1 Overall Cost Comparison
Automatic barrier crossings are provided on all participating railways2, and therefore
provide the best basis for a benchmark comparison of overall costs. The data (Figure 1)
shows that there is a significant variation in the cost between countries; the cost for
country 'A' is 50% higher than the next most costly, 'B', and nine-times higher than the
country with the lowest cost automatic barrier crossing ('I').
Figure 1: Cost Comparison - Automatic Barrier Crossings
0
100
200
300
400
500
600
700
800
900
1000
A
B
CDE
F
G
H
I
Country
A
u
t
o
B
a
r
r
i
e
r
A
ver
ag
e
C
o
s
t
(0
00s
E
u
r
o
)
Source: Arthur D Little - data provided for this study by participants
Note: all reported costs have been converted to Euros at the relevant exchange rate
1
The data is based on that provided by the representatives of the participating railways.
2
There are different philosophies in the design of automatic barrier crossings; railways in some countries permit automatic full-barrier
crossings ('D', 'E' and 'I'), whereas in other countries automatic crossings are restricted to half-barriers (to provide a clear exit route for a
vehicle that would otherwise be 'trapped' on the crossing) .
11
•
The 'average' cost across all nine participating countries is €350,000.
•
Based on the data provided, the lowest cost automatic barrier crossings (countries
'G', 'H' and 'I') cost marginally over €100,000. This compares to the cost of
crossings in country 'A' which average over €900,000.
•
Significantly, the three countries with the highest costs ('A', 'B' and 'C') share the
most complex technical scope (for a more detailed analysis of technical scope and
other differences see section 2.3).
•
Those countries reporting amongst the lowest costs share broadly similar technical
scopes and similar traffic densities between these countries. They also share the use
of predictor technology to detect the approaching train as an alternative to more
traditional track circuits (see Chapter 3 for an in-depth review of cost drivers).
The costs shown here are 'average' or typical costs, although in reality the cost in any
single country varies very considerably (Figure 2). This variation is substantially due to
the differences in the technical scope of each upgrade project; the costs are highly
dependent on the nature and extent of work carried out (see section 2.3 for more detail).
As such, there is overlap in the costs of crossings in countries 'A' and 'C'; the highest
cost crossings in country 'C' are about the same as the average cost of crossings in
country 'A'.
Figure 2: Variation in Cost
0
200
400
600
800
1000
1200
1400
1600
AC
F
H
I
Country
C
o
st
(
0
00
s E
u
r
o
)
Importantly, there are key differences between countries in design philosophy, technical
scope and network characteristics, that need to be considered in a comparison of cost
information. Indeed, a like-for-like comparison is not possible. (See Chapter 3 for an
in-depth review of the 'cost drivers').
12
Network characteristics vary considerably between the participating countries, which
has a bearing on the cost of level crossings. For example, the rail network of Britain is
on average more heavily utilised than those of Australia, Ireland, Sweden and much of
the USA. Also more of these networks are double-track with moderate/high line speed.
Representatives from some countries reported that costs depend significantly on
whether the crossing was installed on single or double track, and if there were
particularly complicated signalling arrangements. What is provided in Figure 1 is a
comparison of a 'typical' crossing rather than the least or most costly of a particular
type. Figure 2 shows a comparison of the costs of automatic barrier crossings, showing
the average from Figure 1 and also the upper and lower costs where these were
provided.
The study also reveals that the management of project cost information reflects a
significant diversity of contract strategy and reporting mechanisms, both across
different countries, and across individual projects within a single country. As such,
detailed comparisons of the breakdown of costs is not possible since there is no
common approach to cost reporting.
2.2 Cost of Different Forms of Protection
The data shown in section 2.1 above is for automatic barrier crossings. Fewer data are
available for other types of protection but general comparisons are possible (Figure 3)
highlighting how cost increases with increasing levels of protection.
•
The 'simplest' form of protection is the footpath with warning lights . The cost of
such crossings is typically some 50% lower than the automatic barrier crossing
(although it is worth noting that cost variation is considerable) . This is an
interesting data point, since this type of crossing lacks much of the equipment
provided at automatic barrier crossings, but importantly has similar technical
requirements (train detection equipment and associated cabling). The fact that
footpath crossings with lights average about half the cost of automatic barrier
crossings indicates the significance of the signalling related components of the cost
of level crossings.
•
Automatic 'lights and bells' crossings are similar to automatic barrier crossings,
but lack the barrier and related equipment (barrier machines and associated control
and power systems). These cost anywhere between 10 and 50% less than automatic
barrier crossings. This suggests that in some cases, the additional cost of providing
the barriers and associated systems is fairly trivial, whilst in other cases these add
significantly to the cost.
13
•
The highest form of protection available is the manually controlled barrier
crossing , which is only used in three of the participating countries. At such
crossings, full barriers are provided (i.e. those that close fully across the road), and
the crossing is only cleared for the train to proceed once it has been checked
manually (either locally or via CCTV). The cost of manually controlled barrier
crossings are anywhere between 30% and 90% higher than automatic barrier
crossings.
Figure 3: Comparative Costs of Different Levels of Protection
-100%-80%-60%-40%-20%0%20%40%60%80%100%
FP with lights
Auto lights and bells
Auto barrier
Manually controlled
barrier
Cost Compared to Auto Barrier (% lower or higher)
Increasing
p
r
otec
tion
Note: each bar represents the % higher or lower cost relative to automatic barrier crossing costs, in the same country.
14
Document Map
Chapter
Key Contents
1. Introduction
Background to study
2. Cost
Benchmarking
Overall comparison of costs
Cost of different levels of level crossing protection
3. Cost Drivers
Differences in technical scope of level crossings
Costs of project management and overheads
Equipment housing, cabling, barriers and lights
Accessing the railway to do work
Labour and installation
Testing and commissioning
4. Comparative
Safety
Performance
Relationship between cost of level crossings and safety
performance delivered
5. Funding
Mechanisms
Different mechanisms for funding level crossing programmes
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
6. Impact of Cost
Reduction
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
7. Conclusions
Summary of key findings and conclusions from study
8. Good Practice
In Controlling
Costs
Suggestions for controlling costs focusing on key cost drivers
identified
Suggestions for work breakdown structure
15
3. Cost Drivers
A number of important factors drive higher level crossing costs. These range
from highly bespoke designs, requirements for monitoring crossing status and
use of traditional train detection systems compared with more modular solutions
that utilise predictor technology. Overall cost can also be significantly affected
by less direct items such as overheads and project management, and use of
buildings to house equipment rather than smaller location cases.
3.1 Overview
The research determined a number of key factors that have an influence on the costs of
level crossings:
•
The technical scope (i.e. the equipment provided)
•
Project management and overheads
•
Requirements for buildings to house equipment, cabling, barriers and lights
•
Accessing the worksite (frequency of rail traffic - amount of 'white space')
•
Usage of contractors versus completing work in-house
•
Supply chain differences
•
Labour rates
•
Installation (i.e. the number of man-hours)
•
Testing and commissioning
But how closely do these factors correlate to the differences in costs between different
countries?
The relative costs of these elements have been ranked to give a 'cost element score' in
comparison to the country with the highest costs (country 'A'). For example, if country
'D' has technical scope that is less complex than country 'A', then it receives a score of
-1 for this particular element. Conversely, if the labour rate in country 'D' is higher
than country 'A' then it receives a score of +1. For 'significant' differences, a score of
-2 or +2 is given, and where there is little or no difference, a score of 0 is given. The
overall 'score' is the sum of all cost elements; a large negative score indicates that costs
should be significantly less than in country 'A'. Note that it has generally not been
possible to base this on actual data, as a breakdown in each of these categories was not
widely available. The analysis is therefore based partly on the opinions of those
participating in the study, and indicates those areas in which costs may be higher, lower,
or similar.
16
The results (Figure 4) show that in all participating countries, the overall 'score' is
negative, which is consistent with country 'A' reporting the highest level crossing costs
of all participating countries. There appears to be a reasonable trend, which suggests
that these cost elements are indeed an important determinant in the cost of level
crossings:
•
Countries 'F', 'G' and 'H', have amongst the lowest actual costs, and also have the
largest negative 'score'.
•
Conversely, the costs in countries 'B' and 'C', which are closer to those in country
'A', have the smallest negative 'score'.
•
The comparatively low costs in country 'I' (and to a lesser extent 'D' and 'E'),
however, cannot fully be explained by this analysis. This suggests either that there
are some other important factors that are not included in this analysis, or that
weightings would need to be applied to the components of the cost element score to
produce a more accurate model.
Figure 4: Comparison of Costs and Cost Driver between different Countries 3
0
100
200
300
400
500
600
700
800
900
1000
ABCDEF GHI
Country
A
u
t
o
B
a
r
r
i
er
A
v
er
a
g
e
C
o
st
(00
0s
Eu
r
o
)
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
C
o
s
t
E
l
em
en
t
S
c
o
r
e
Automatic barrier cost
Cost elements score
Source: Arthur D. Little analysis, based on interviews and data provided by representatives of participating countries
17
3.2 Technical Scope
Perhaps most significantly of all cost drivers, there are some notable differences in the
type and level of equipment provided to deliver the level crossing functionality - i.e. the
technical scope. In general, those countries that report highest costs also have the
greatest 'technical scope'. Specific aspects of technical scope relate to train detection,
monitoring of the crossing status, the highest form of protection provided and overall
design complexity:
•
Train detection : The costliest train detection function is provided by using both
track circuits and independently cabled treadles. The cost of the treadle and cabling
can be in excess of €200K - which is over and above the cost of the train detection
function in those countries where a treadle is not generally provided (either where
predictors are used or where conventional track circuits are used).
•
Monitoring crossing status : In some countries there is a requirement to monitor
remotely the status of protected level crossings. Such monitoring is potentially
costly, as the status of the crossing must be sent to the signalbox via dedicated
cabling. Where there is no available cable route this can potentially have a high
impact on the cost. Again, those countries where such monitoring is a requirement
can be associated with high overall costs (i.e. the left-hand end of Figure 1). An
alterative philosophy, used in some countries, is to provide a phone number to call
in the event of an emergency or apparent crossing failure.
•
Highest form of protection: The highest form of level crossing protection is the
manually controlled full barrier crossing - with an average cost between 30% and
90% higher than an automatic crossing, in the three countries where these are
provided. The six other countries consider either a full or half-barrier automatic
crossing to be the highest form of protection before grade separation. The opinion
of one person interviewed for this research was "…. the only affordable type of
crossing is one that is fully automatic and not monitored".
A lower cost option for controlling the operation of full-barrier crossings (in
comparison to the CCTV function which requires manning) is radar detection. The
costs of this system are reported to be around €100K per installation (which
represents a whole-life cost saving in comparison to manning the crossing).
•
Design complexity: Many of the aspects of technical scope are reflected in the
'design complexity' of the crossing; where technical scope is greater, there is an
associated higher complexity in design. This means that design can be highly
bespoke and costly (average of 7% of total project costs, but as high as 20%), in
comparison to countries where the technical scope is less complex, and design is
more modular and 'off-the-shelf'. Use of predictors, in particular, is cited as a
particular aspect of the system which is conducive to lowering the cost of design (as
well as of equipment).
18
Also, there is a widespread view that delivering level crossings in larger
programmes using more generic designs helps to reduce cost, through both lower
design cost and reduced project management and overheads (see also section 3.3
below).
3.3 Project Management and Overheads
Generally, the cost of project management in all participating countries is of the order
of 20% of the total costs - where this was provided.
The costs of project management and overheads will depend on how the work is
managed, and the extent to which contractors are used to carry out the work. Many
countries use a main contractor to carry out the bulk of the work. Some, in contrast,
complete the work in-house, or use a separate division of the railway to complete the
design in-house and carry out the project management of the work which is given to a
contractor.
Work may be outsourced and a 'turn-key' approach used; although comparative data
has not been available for this study, this approach is thought to have reduced project
management costs by around 20%.
3.4 Requirements for Buildings to House Equipment, Cabling, Barriers and
Lights
There are no specific data available for comparing the detailed costs of components of
level crossings, but the following observations are made:
•
Cabling costs (where traditional train detection systems are used) average 5% of the
total project costs, although variation is considerable with it ranging from 2% to
15% of total project cost. Cabling costs will be lower where predictor technology is
used, and where there is no requirement for independently cabled treadles for train
detection or for monitoring the crossing remotely by cable. As mentioned above,
the cost of the treadle and cabling can be around €200K, so this can have a
significant impact on the overall cost.
•
The cost of supplying and installing barrier machines are significantly different
between countries. This may reflect to some extent the volume of equipment
provided (in other words, large volumes of supply might make lower prices
possible) . The cost of barrier machines themselves will have some impact on the
total cost (the highest costs reported were €14K per barrier and machine, although
lower costs are reported in some countries). However, the installation rather than
the equipment cost is a more significant cost factor - total installed costs of well
19
over €100K have been reported for a four barrier crossing, with a large proportion
of the cost related to the use of large bases, which can be difficult to fit in the
available space. (For difference in costs between crossings with and without
barriers see comparison of automatic barrier crossings to automatic lights and bells
crossings - Figure 3).
•
The building used to house the equipment is a highly variable cost driver, and can
have a significant impact on the cost (in some cases in excess of €150K). The cost
is very dependant on the size of the building, the furniture and contents, and also the
civil engineering costs of the base, which can be significant. 'Good practice' might
be seen as avoiding buildings altogether; many countries house equipment in
weather-proof equipment cases which are lower cost in comparison.
3.5 Access ('white space')
On networks that carry a higher frequency of rail traffic it is more difficult (and also
potentially more costly) to gain access to the track to do work. This means that good
planning of work is essential. Problems are exacerbated where large numbers of
companies are used to each carry out specialist activities, all of whom require access to
the work site. Missed possession slots and delays can be a cause of significant increases
over the original budget. Even where no significant delays are incurred, work is less
efficient on busy lines, simply because the time available between trains is less. One
solution in such cases is to carry out other major works at the same time and close the
line for longer periods.
3.6 Labour Rates and Installation
Many interviews confirmed that non-material 'labour' costs dominate the overall cost of
level cross ings, compared with the materials costs. Non-materials costs in this context
include project management costs, feasibility and design, installation, testing and
commissioning. Detailed cost breakdown information was not available for most
countries, but what was available suggests that the ratio of non-materials to materials
costs varies from 4:1 and 2:1. The higher ratio is more typical of countries that have
more complex technical scopes (meaning that associated design, installation, testing, etc
costs are greater).
Another issue is that in many cases the split between level crossing related labour (i.e.
time) and equipment is not accurately recorded on a job. This is particularly true on
large worksites where there may be a number of projects (e.g. a resignalling scheme
which includes work on level crossings).
20
3.7 Testing and Commissioning
Testing and commissioning costs are influenced primarily by the design of the crossing,
the extent to which the design is bespoke and how much testing can take place away
from the lineside. Whilst detailed costs for testing and commissioning in participating
countries were generally not available for this research, the figures that were available
show a significant variation. At the lower end, the testing and commissioning cost was
less than €15,000, while the country with the highest testing and commissioning costs
had costs about six times higher than this. At the upper end of the cost range, crossing
design and construction is highly bespoke and all testing is done on site, as it is only
possible to do this when the crossing has been integrated into the signalling system on
the live railway. In comparison, in countries that employ predictor technology and
design is more 'off-the-shelf' much of the testing can be done prior to construction at
the site, so the costs of testing are comparatively lower.
In some cases, poor availability (and planning) of testers has been given as a reason for
escalating cost, usually because unavailability of this scarce resource leads to a delay in
the overall project with consequent general cost increases.
21
Document Map
Chapter
Key Contents
1. Introduction
Background to study
2. Cost
Benchmarking
Overall comparison of costs
Cost of different levels of level crossing protection
3. Cost Drivers
Differences in technical scope of level crossings
Costs of project management and overheads
Equipment housing, cabling, barriers and lights
Accessing the railway to do work
Labour and installation
Testing and commissioning
4. Comparative
Safety
Performance
Relationship between cost of level crossings and safety
performance delivered
5. Funding
Different mechanisms for funding level crossing programmes
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
6. Conclusions
Summary of key findings and conclusions from study
7. Good Practice
In Controlling
Costs
Suggestions for controlling costs focusing on key cost drivers
identified
Suggestions for work breakdown structure
22
4. Comparative Safety Performance
There is some evidence that countries with comparatively costly crossings are
delivering relatively good safety performance, but it is not clear the extent to
which the higher costs represent good value for money.
Chapters 2 and 3 compare the costs of level crossings between countries and examine
the factors that influence costs. This Chapter explores whether or not those countries
that report higher level crossing costs demonstrate a better safety performance.
Figure 5 compares safety performance with cost for countries participating in this study
for which relevant data was available. The green bars indicate safety performance as
"Safety Risk Index"4. The red-line shows the average cost of automatic crossings (from
Figure 1). This comparison suggests some link between safety performance and cost:
•
Countries with relatively high costs ('A' and 'B') are delivering good safety
performance.
•
Good safety performance, however, is not only delivered by countries with
relatively high cost automatic crossings; the other country with good safety
performance (country 'I') actually has the lowest cost automatic crossings of all.
This shows that other factors are very significant in determining the actual level of
safety loss; clearly, the comparison takes no account, for example, of the level of
usage of the crossing by road users, nor the rail traffic density or type. Country 'I'
has both a comparatively very lightly used rail network and low road traffic
volumes.
•
The comparison also suggests that in countries where costs are relatively high,
further improvements in safety performance will be very difficult to deliver cost-
effectively; for countries 'A' and 'B' costs are high and residual safety performance
is good so it will be harder to justify more expenditure.
•
The reverse is true for countries 'F', 'C', and 'G' where costs appear lower
compared to the safety performance. In principle, there is greater potential in these
countries to achieve further improvements in safety performance cost-effectively.
4
"Safety Risk Index" is used as a benchmark of safety performance expressed as fatalities per year normalised by the number of
crossings. This has been normalised again to unity for country 'F' which has the largest safety loss of participating countries.
23
Figure 5: Cost versus Safety Risk Index
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
FC
G
A
I
B
Country
S
a
f
e
t
y
R
i
sk
I
n
d
ex (
N
o
r
m
a
lised
)
Low
er
i
s
bett
e
r
0
100
200
300
400
500
600
700
800
900
1000
A
u
t
o
B
a
r
r
i
er
A
ver
ag
e C
o
st
(
000s
E
u
r
o
)
Safety Risk Index
Average cost
Source: Safety Risk Index derived from normalisation of data reported in "Road Vehicle Level Crossings Special Topic Report", RSSB,
January 2004
It is important to bear in mind that the relative safety performance shown in Figure 5
does not account for the differences in the average traffic moments (number of trains
multiplied by number of vehicles) between each country. Traffic moment has been
strongly linked to the occurrence of accidents at level crossings, and as such is a key
feature for most level crossing risk models. Countries with higher traffic moments will
have a greater risk exposure, all other aspects being equal. The average traffic moment
at automatic barrier crossings in country 'A' is over three times greater than for similar
crossings in country 'B', which may go towards explaining the apparent difference in
safety performance. Average traffic moments for other countries are not known in
detail, but country 'I' for example is known to have a traffic moment which is relatively
low.
Also, the overall level crossing safety performance demonstrated by a railway is likely
to be less influenced by the technical standard of a given type of level crossing than
other key factors:
•
the mix of level crossings types (protected crossings versus passive crossings) in a
country will have a significant influence on the overall level of safety. The mix will
depend on what was provided historically, and the basis for upgrade between
countries; in some places upgrades are based more on an assessment of risk, and in
others technical standards or traffic volume is used. It might reasonably be
expected that where the criteria for upgrades are more stringent or, for example
there are fewer heavily used passive crossings, that overall safety performance will
be better (all other aspects being equal).
24
•
the general road safety performance in each country is likely to have an impact on
safety performance, given that the vast majority of level crossing accidents are
caused by road vehicle driver error or misuse as opposed to level crossing failures.
The study has been unable to find evidence to show that higher technical standards and
additional functionality for a given level crossing type justifies the higher associated
costs.
25
Document Map
Chapter
Key Contents
1. Introduction
Background to study
2. Cost
Benchmarking
Overall comparison of costs
Cost of different levels of level crossing protection
3. Cost Drivers
Differences in technical scope of level crossings
Costs of project management and overheads
Equipment housing, cabling, barriers and lights
Accessing the railway to do work
Labour and installation
Testing and commissioning
4. Comparative
Safety
Performance
Relationship between cost of level crossings and safety
performance delivered
5. Funding
Mechanisms
Different mechanisms for funding level crossing programmes
6. Impact of Cost
Reduction
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
7. Conclusions
Summary of key findings and conclusions from study
8. Good Practice
In Controlling
Costs
Suggestions for controlling costs focusing on key cost drivers
identified
Suggestions for work breakdown structure
26
5. Funding Mechanisms
There is a wide variety of means for funding level crossing upgrade programmes
and in most cases there is a mechanism for sharing of costs between the
railway, local authorities (roads) and government.
The research identified a number of funding models used in different countries.
•
In one participating country, level crossing upgrades and renewals are funded by the
railway (infrastructure controller) with little or no costs paid for by other parties,
regardless of the reason for the upgrade.
•
In many countries costs are shared, although there are different mechanisms for
doing so, for example in different countries:
−
Life expired crossing renewals are funded by the state. For crossings that are
upgraded on the basis of safety, funding is often split between the state, road
and railway
−
Maintenance costs are shared according to a fixed quota: 12.5% road, 7.5%
railway, 80% government. Funding for the upgrade of level crossings to
improve safety depends on the reason for the increase in risk; the railway will
pay if the train services are being increased, if road traffic levels increase due
to, for example, construction of commercial or residential areas then the road
will pay for work
−
In one country, there has apparently been ongoing debate about how road
authorities contribute to the costs of level crossings. If the crossing is on a
public road, then the road authority may fund half the cost of the work. There
is, however, apparently no prescriptive mechanism in place so the funding
varies from project to project. Level crossings that are on private roads will be
funded by the railway
−
The allocation of funding depends on the reasons for the work. The railway
and road authority share the costs (both of whom are governmental
organisations) and local authorities will often pay for a proportion of costs, if
they agree that they will receive benefit from the work. Again, there is not
actual legislation to divide the provision of funding and waiting for funding
approval from local authorities can apparently delay the work in some cases
−
Funding for a significant programme of work for upgrading level crossings was
approved on the basis of a risk model applied to all level crossings on the
network. Costs are not shared between road and railways explicitly, although
the railway is publicly owned
−
Generally the cost of work is split between the railway infrastructure company,
and the road authority. Typically, the cost of road related work will only by
27
around 10% of the total cost and sometimes may be even less, meaning that the
railway fund the majority of the cost
28
Document Map
Chapter
Key Contents
1. Introduction
Background to study
2. Cost
Benchmarking
Overall comparison of costs
Cost of different levels of level crossing protection
3. Cost Drivers
Differences in technical scope of level crossings
Costs of project management and overheads
Equipment housing, cabling, barriers and lights
Accessing the railway to do work
Labour and installation
Testing and commissioning
4. Comparative
Safety
Performance
Relationship between cost of level crossings and safety
performance delivered
5. Funding
Mechanisms
Different mechanisms for funding level crossing
programmes
6. Impact of Cost
Reduction
Case study looking at possible impact of reducing costs on
overall safety performance and overall spend
7. Conclusions
Summary of key findings and conclusions from study
8. Good Practice
In Controlling
Costs
Suggestions for controlling costs focusing on key cost drivers
identified
Suggestions for work breakdown structure
29
6. Impact of Cost Reduction - Case Study
Analysis of the level crossing risk profile indicates that if significant savings
could be made in the cost of upgrades this could result not only in reduced
overall safety risk but also in reduced overall programme spend.
This section presents a case study showing the potential impact on overall spend and
overall safety performance associated with reducing the unit cost of level crossing
upgrades.
Although the costs and safety risk profiles will differ by country, the general
conclusions of the case study should be transferable. The underlying principle of risk
management here is that crossings are upgraded only if the upgrade reduces risk in a
cost effective way and to a level that is ALARP (As Low As Reasonably Practicable).
Figure 6 shows how reducing the average costs of upgrades could, in theory, provide
greater potential for overall risk reduction.
•
The blue line shows the percentage of automatic half-barrier (AHB) crossings that
could be justified for upgrade on a cost-benefit basis, as a function of a reduced unit
cost of conversion. As the reduction in upgrade cost increases (from 0 to 90%
along the x-axis) the number of crossings upgrades that could be justified also
increases. Based on current upgrade costs, about 10% of automatic barrier
crossings can be upgraded to manually controlled barriers. This increases to 20%
of crossings if the cost of upgrade could be reduced by 50% and to 50% of
crossings if costs could be reduced by 90%.
•
The green line shows the level of overall risk reduction that could be achieved if all
the cost-effective upgrades were carried out. With current costs of upgrade, this
shows that some 60% risk reduction could be achieved with only 10% of automatic
crossings upgraded to full-barrier (this is because the highest risk level crossings
would be upgraded, so each would have a significant impact on the overall risk
profile). The line broadly follows the number of crossings that can be upgraded,
reaching about a 90% risk reduction if costs could be reduced by 90%.
•
The red line shows the total spend on upgrading, assuming that funds were
available to carry out all cost effective upgrades. If the unit costs of upgrade were
reduced by up to 25%, the total spend on upgrading would stay the same. However,
if the unit costs could be reduced beyond 50%, the total spend would actually
reduce, even though a greater risk reduction could be achieved. This trend relates
to the fact that the current cost of upgrades targets only those crossings towards the
top of the risk profile (where the curve is comparatively flat).
30
Figure 7 shows a similar graph for locally monitored automatic open crossings upgrades
to automatic half-barriers. This shows similar trends to those described above for
AHBs.
In summary, the analyses in Figures 6 and 7 show that if costs of upgrading level
crossings can be reduced beyond a certain point, then not only will it be affordable to
upgrade more crossings with an associated reduction in the overall level of risk but the
total spend should over time also reduce.
31
Figure 6: Influence of Reducing Automatic Barrier to Manually Controlled Full Barrier MCB Cost
on Risk Reduction and Total Spend
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0%
25%
50%
75%
90%
Unit Upgrade Cost Reduction
-100%
-80%
-60%
-40%
-20%
0%
20%
C
h
a
n
g
e
i
n
to
ta
l
s
p
e
n
d
% Crossings
Upgraded that
Satisfy ALARP
% Risk Reduction
Change in total
spend (%)
5
1. At the current average upgrade cost, about 10% of automatic barrier crossings could be upgraded on
a cost-benefit basis, i.e. upgrades would proceed at crossings where the safety benefits would equal
or exceed the costs of the upgrade. Risk would be reduced by about 60% across all automatic barrier
crossings.
2. If the costs of upgrade were reduced by 25%, a further 3% of automatic barrier crossings could be
justified on a cost-benefit basis. The total risk reduction would increase to around 66%. The total
spend would stay the same as (1).
3. A 50% cost reduction would mean that twice the number of level crossings could be upgraded for a
marginally higher spend. The overall risk reduction would increase to approximately 77%.
4. If costs were reduced by 75% the total spend would actually decrease. 31% of AHBs could be
upgraded giving a risk reduction of around 86%, for a 20% reduction in total spend.
5. An even more significant cost reduction of 90% would mean that the number of crossings that could
be upgraded would increase to almost 50%, with an overall spend that is reduced by over 40%. The
risk reduction would be almost 90%, as only the very lowest risk crossings would not be upgraded.
Source: Arthur D. Little analysis
3
2
1
4
32
Figure 7: Influence of Reducing Automatic Open Crossings to Automatic Barrier Cost on Risk
Reduction and Total Spend
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0%
25%
50%
75%
90%
Unit Upgrade Cost Reduction
-100%
-80%
-60%
-40%
-20%
0%
20%
C
h
an
g
e
i
n
t
o
t
a
l
sp
en
d
% Crossings
Upgraded that
Satisfy ALARP
% Risk Reduction
Change in total
spend (%)
5
4
3
2
1
1. At the current average cost of upgrade from automatic open crossings to half-barrier, only 5% of
automatic open crossings could be upgraded on a cost-benefit basis - i. e. upgrades would proceed at
crossings where the safety benefits would equal or exceed the costs of the upgrade. This would give
a risk reduction of around 38% across all automatic open crossings.
2. If the costs of upgrade were reduced by 25%, then a further 2% of crossings could be justified on a
cost-benefit basis. The total risk reduction would increase to just over 40%. The total spend would
increase by around 5-10%.
3. If a 50% cost reduction could be achieved the total spend would not increase, but it would be possible
to deliver a risk reduction of nearly 50%.
4. If costs were further reduced by 75% the total spend would decrease by 20% and the risk reduction
improve to 54%.
5. An even more significant cost reduction of 90% would give an increase in the number of crossing
upgrades and in the overall risk reduction for a 30% reduction in overall spend.
Source: Arthur D. Little analysis
33
Document Map
Chapter
Key Contents
1. Introduction
Background to study
2. Cost
Benchmarking
Overall comparison of costs
Cost of different levels of level crossing protection
3. Cost Drivers
Differences in technical scope of level crossings
Costs of project management and overheads
Equipment housing, cabling, barriers and lights
Accessing the railway to do work
Labour and installation
Testing and commissioning
4. Comparative
Safety
Performance
Relationship between cost of level crossings and safety
performance delivered
5. Funding
Mechanisms
Different mechanisms for funding level crossing programmes
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
6. Impact of Cost
Reduction
Case study looking at possible impact of reducing costs on overall
safety performance and overall spend
7. Conclusions
Summary of key findings and conclusions from study
8. Good Practice
In Controlling
Costs
Suggestions for controlling costs focusing on key cost drivers
identified
Suggestions for work breakdown structure
34
7. Conclusions
The study shows that there are a number of reasons behind the significant
variation in the cost of level crossing upgrades between different countries.
7.1 Overall Conclusions
The study draws the following overall conclusions:
•
The cost of automatic barrier crossings varies significantly between countries -
ranging from a national average of just over €100,000 to over €900,000.
(ref: Figure 1)
•
There is significant variation in the technical scope and complexity of crossings
between different countries, which is a key factor in determining the cost.
(ref: section 3.2)
•
The recorded costs of level crossings are also highly dependant on the scope of
work recorded within the project costs - there is no such thing as a 'standard' level
crossing upgrade or renewal.
(ref: section 2.1)
•
Only three participating countries provide the most comprehensive type of level
crossing protection - manually controlled crossings - which are between 30% and
90% higher cost than automatic crossings.
(ref: section 2.2)
•
The cost of 'non-materials' costs (design, installation, overheads, testing and
commissioning and project management) dominates the cost of level crossings
rather than materials costs which only contribute between 20% and 30% of total
cost.
(ref: section 3.6)
•
Countries with relatively high automatic crossing upgrade costs (and complex
technical scopes) are delivering comparatively good safety performance. However,
'good' safety performance is also being delivered by countries with relatively lower
cost automatic crossings; perhaps because of comparatively low levels of usage of
the crossing by road users, and low rail traffic densities.
(ref: Chapter 4)
•
It is unclear the extent to which the technical complexity of crossings in some
countries contributes to strong comparative safety performance, and therefore what
safety benefit is being delivered for the significant additional cost.
(ref: Chapter 4)
35
•
Many different funding models exist across the participating countries. These range
from the work entirely funded by the railway, to other models in which costs are
shared between the railway, the local community, road authorities and central
government.
(ref: Chapter 5)
•
Analysis suggests that reducing the unit costs of level crossing upgrades could
mean that more upgrades could be justified on the basis of reasonable practicability.
If cost of upgrades could be reduced significantly, it may be possible to deliver an
increased overall safety performance for a reduced total spend.
(ref: Chapter 6)
7.2 Summary of Cost Drivers
In summary, the main drivers of costs of level crossings can be categorised as follows:
•
A key factor that determines the cost is the level crossing design complexity. The
consequence of higher complexity is increased cost of design, equipment,
installation, testing and commissioning.
(ref: section 3.2)
•
Where designs are highly bespoke, design costs can be high (the average is 7% of
total project costs, but they can be as high as 20%). In comparison, countries which
employ more 'off the shelf' solutions, for example those associated with designs for
crossings that use predictors, have lower design costs.
(ref: section 3.2)
•
'In-house' design teams are reported to help reduce design overheads, increase
retention of skills and knowledge, move towards more standard designs and to
create a more predictable flow of work to the industry.
(ref: section 3.3)
•
Two forms of independent train detection, which requires both track circuit and
treadle, is not a requirement in most countries, but does add significantly to the cost
where it is used.
(ref: section 3.2)
•
Remote monitoring of the status of crossings is also not a requirement in most
countries, but where it is required can add significantly to the cost (particularly for
geographically remote crossings as this is done by cable). Lower cost alternative
include wireless (GSM) transmission of status, or the removal of the requirement to
monitor status altogether, (such as providing a free-phone service for the public to
report problems).
(ref: section 3.2)
36
•
Buildings to house equipment that have large bases, where used, can add
significantly to the cost (in excess of €150K in some cases for a fully fitted and
installed building). 'Good practice' would suggest a greater use of lower cost
alternatives (smaller 'location cases') which have a substantially reduced footprint
and significantly lower cost.
(ref: section 3.4)
•
On busy networks, weaknesses in the planning of possessions, and in the
management of procurement of equipment and testers can, in isolated cases, lead to
significant delays and associated increases in cost.
(ref: section 3.5)
•
There is a common perception that where there are numerous contractual layers
associated with delivering a level crossing, overheads and profit add significantly to
costs. Some countries deliver work in larger quantities with fewer contractual
layers, and whilst we have been unable to compare the costs directly there is an
impression that this does lead to reduced overall cost.
(ref: section 3.3)
•
Project management and overheads can contribute up to 20% of total cost; the
highest being as