Riser 2006 D6

D06: European Best Practice for Roadside Design:
Guidelines for Roadside Infrastructure on New and

Existing Roads
Project ACRONYM: RISER
TITLE: Roadside Infrastructure for Safer European Roads
PROJECT START DATE: 1/01/2003 DURATION: 36 months
Original Date of issue of this report: 28/02/2006
Submitted by: Chalmers University of Technology

Main authors:
Robert Thomson, Helen Fagerlind, Chalmers, SE
Angel V. Martinez, Antonio Amenguel, HIASA, ES
Claire Naing, Julian Hill, VSRC, UK
Heinz Hoschopf, TUG, AT
Guy Dupré, Olivier Bisson, CETE, FR
Marko Kelkka, HUT, FI
Richard van der Horst, TNO, NL
Juan Garcia, CIDAUT, ES
and the RISER Consortium
Project funded by the

European Community under
the ‘Competitive and
Sustainable Growth’
Programme (1998-2002)
European Best Practice for Roadside Design
Guidelines for Roadside Infrastructure on New and Existing Roads
FOREWORD
The European Commission Directorate General for Transportation and Energy (DG-
TREN) sponsored a research project to investigate the best practice guidelines for
roadside infrastructure. The RISER consortium has compiled the following document
which is a synthesis of existing practice in Europe with additional information
collected from accident and human behaviour studies. Several technical reports
developed in the RISER project were the basis for this document and can provide
more technical information.
The following information is presented as a template for future users. There are
national and regional issues that arise when it comes to the implementation of
European norms or guidelines into the member states. This document should be
considered as a starting point for national policies that must be adapted to the local
geographical, economic, and demographic conditions. Through the use of a common
starting point, commonly accepted best practice procedures will be spread
throughout the EU member states and facilitate improved roadside safety design,
and – most importantly – safety levels throughout the EU.
In view of the current EU focus on road safety it is important to recognize the 3-pillar
concept for road safety being:
Infrastructure design
Vehicle design
Driver (education)
It is evident that the following information addresses the infrastructure aspect for road
safety. It should be recognized that the RISER project has included driver and
vehicle aspects to not lose sight of the integrated approach that is required to reduce
road traffic casualties.
The information contained in this document should be used in conjunction with the
document “European Best Practice for Roadside Design: Guidelines for
Maintenance and Operations of Roadside Infrastructure”.
i
RISER CONSORTIUM
Organisation Acronym Country

CHALMERS Sweden
Chalmers University of Technology (Coordinator)
Robert Thomson, Helen Fagerlind, Gunnar Lannér
CETE France
Centre d’Etudes Techniques de l’Equipement Normandie Centre
Guy Dupré, Olivier Bisson
CIDAUT Spain
Fundación para la Investigación y Desarrollo en Transporte y
Energía
Juan M. García, Francisco López
ERF Belgium
European Union Road Federation
Brendan Halleman, Jose Papí
HIASA Spain
HIASA Grupo Gonvarri
Angel V. Martínez, Antonio Amengual
HUT Finland
Helsinki University of Technology
Jarkko Valtonen, Marko Kelkka, Ute Gosse
TNO The Netherlands
Netherlands Organisation for Applied Scientific Research
Cees W. Klootwijk, Boudewijn Hoogvelt,
Richard van der Horst, Selma de Ridder,
TUG Austria
Graz University of Technology
Heinz Hoschopf
VSRC United Kingdom
Vehicle Safety Research Centre, Loughborough University
Claire Naing, Julian Hill
V&R Germany
Volkmann & Rossbach GMBH & CO.KG
Wolfgang Wink
ii
TABLE OF CONTENTS
FOREWORD..........................................................................................................................................................I
TABLE OF CONTENTS....................................................................................................................................III
LIST OF FIGURES ..............................................................................................................................................V
LIST OF TABLES ........................................................................................................................................... VIII
INTRODUCTION................................................................................................................................................. 1
PRINCIPLES FOR SAFE ROAD NETWORKS ............................................................................................................ 1
THE SELF EXPLAINING ROAD AND FORGIVING ROADSIDE PHILOSOPHIES .......................................................... 2
FRAMEWORK FOR ROADSIDE INFRASTRUCTURE DESIGN .................................................................................... 4
CHAPTER 1: DEFINITION AND IDENTIFICATION OF ROADSIDE AND MEDIAN HAZARDS ... 6
1.1 INTRODUCTION ..................................................................................................................................... 6
1.2 DEFINITIONS OF ROADSIDE AND MEDIAN OBJECTS (POINT AND DISTRIBUTED HAZARDS)................... 6
1.2.1 Point Objects................................................................................................................................... 6
1.2.2 Distributed Objects ......................................................................................................................... 8
1.3 OVERVIEW OF RESEARCH ..................................................................................................................... 9
1.3.1 Previous Studies .............................................................................................................................. 9
1.4 RISER ANALYSIS ................................................................................................................................10
1.4.1 RISER Statistical Database ............................................................................................................10
1.4.2 RISER Detailed Database ..............................................................................................................11
1.4.3 Current European Guidelines ........................................................................................................14
1.5 SUMMARY ...........................................................................................................................................17
1.5.1 Proposed Measurements for the Definition of Hazards .................................................................17
REFERENCES ......................................................................................................................................................19
CHAPTER 2: SPECIFICATIONS FOR SAFETY ZONES.........................................................................20
2.1 INTRODUCTION ....................................................................................................................................20
2.2 DEFINITION OF SAFETY ZONE ..............................................................................................................20
2.3 OVERVIEW OF RESEARCH ....................................................................................................................20
2.3.1 The Safety Zone Concept................................................................................................................20
2.4 RISER ANALYSIS ................................................................................................................................21
2.4.1 Criteria to Dimension the Safety Zone ...........................................................................................21
2.4.2 Exit Conditions for Run-Off-Road Accidents .................................................................................22
2.4.3 Strategy for Determining a Safety Zone Width...............................................................................29
2.5 SUMMARY ...........................................................................................................................................35
2.5.1 Safety Zone Requirements ..............................................................................................................35
REFERENCES ......................................................................................................................................................36
CHAPTER 3: THE RECOVERY ZONE ......................................................................................................37
3.1 INTRODUCTION ....................................................................................................................................37
3.2 DEFINITION OF THE RECOVERY ZONE ..................................................................................................37
3.2.1 The Main Functions of a Recovery Zone........................................................................................37
3.3 OVERVIEW OF RESEARCH ....................................................................................................................38
3.3.1 Hard Shoulder Positive Effects on Accident Rates.........................................................................38
3.3.2 Soft Shoulder Negative Effects on Run-off-road Accidents ............................................................39
3.3.3 Hard Shoulder Positive Effect on Accidents against Pedestrians, Bikes and Mopeds ...................41
3.3.4 Hard Shoulder Positive Effect on Other Accident Types................................................................41
3.3.5 Accident Locations and Road Types ..............................................................................................41
3.3.6 The Width of a Recovery Zone .......................................................................................................42
3.3.7 The Roadside Surface Characteristics ...........................................................................................43
3.4 SUMMARY ...........................................................................................................................................44
3.4.1 Good Condition, Good Surface for Paved Shoulders.....................................................................45
3.4.2 No Grass, No Gravel, No Loose Material on Paved Shoulders .....................................................45
3.4.3 Paved Shoulders opposite T-Junctions and Private Accesses ........................................................45
3.4.4 Audible Road Markings as Additional Corrective Actions.............................................................45
3.4.5 Paved Shoulders on the Outside of Curves ....................................................................................45
iii
3.4.6 Discussion ......................................................................................................................................45

REFERENCES ......................................................................................................................................................46
CHAPTER 4: TECHNICAL DESIGN STRATEGIES WITH CASE EXAMPLES .................................48
4.1 INTRODUCTION ....................................................................................................................................48
4.2 OVERVIEW OF HAZARD PROTECTION ..................................................................................................49
4.2.1 Point Hazards.................................................................................................................................49
4.2.2 Distributed Hazards .......................................................................................................................51
4.3 ANALYSIS AND RECOMMENDATIONS OF SPECIFIC HAZARD PROTECTION............................................52
4.3.1 Point hazards Case Examples ........................................................................................................52
4.3.2 Distributed Hazards Case Examples..............................................................................................59
4.3.3 Roadside and Median Road Restraint Systems Considered as Hazards ........................................63
REFERENCES ......................................................................................................................................................65
CHAPTER 5: PASSIVE SAFETY ROAD EQUIPMENT ...........................................................................66
5.1 INTRODUCTION ....................................................................................................................................66
5.2 CLASSIFICATION AND FUNCTION OF PASSIVE SAFETY ROAD EQUIPMENT ...........................................66
5.2.1 Safety Barriers ...............................................................................................................................66
5.2.2 Crash Cushions ..............................................................................................................................67
5.2.3 Terminals/End Treatments .............................................................................................................67
5.2.4 Transitions .....................................................................................................................................68
5.2.5 Arrester Beds..................................................................................................................................69
5.2.6 Break-Away or Energy Absorbing Structures ................................................................................69
5.3 EUROPEAN STANDARDS APPLICABLE TO ROADSIDE INFRASTRUCTURE...............................................69
5.3.1 Requirements and Test Methods.....................................................................................................70
5.3.2 Performance Criteria .....................................................................................................................71
5.4 INFLUENCE OF PASSIVE SAFETY ROAD EQUIPMENT ON ROAD SAFETY ...............................................76
5.4.1 Cost Benefit ....................................................................................................................................76
5.5 INSTALLATION RECOMMENDATIONS FOR PASSIVE SAFETY ROAD EQUIPMENT ...................................77
5.5.1 Identification of Hazards................................................................................................................78
5.5.2 Selection of Containment Level......................................................................................................79
5.5.3 Lateral Placement of Road Equipment...........................................................................................82
5.5.4 Installation Length of Need ............................................................................................................83
5.6 SUMMARY ...........................................................................................................................................91
REFERENCES ......................................................................................................................................................93
CHAPTER 6: ACCIDENT DATA COLLECTION......................................................................................94
6.1 INTRODUCTION ....................................................................................................................................94
6.2 OVERVIEW OF ACCIDENT DATA COLLECTION .....................................................................................94
6.2.1 Accident Prevention .......................................................................................................................95
6.2.2 Severity Reduction..........................................................................................................................95
6.2.3 Post-Crash Treatment ....................................................................................................................95
6.3 RISER RESULTS...................................................................................................................................95
6.3.1 RISER Databases ...........................................................................................................................96
6.4 CONCLUSIONS......................................................................................................................................97
6.4.1 Police .............................................................................................................................................97
6.4.2 Maintenance and Operation Organisation ....................................................................................98
6.4.3 Hospital and Rescue Services.........................................................................................................98
6.4.4 In-depth Investigations...................................................................................................................98
6.5 VARIABLES ..........................................................................................................................................99
REFERENCE ......................................................................................................................................................101
GLOSSARY .......................................................................................................................................................102
APPENDIX A .....................................................................................................................................................113
APPENDIX B .....................................................................................................................................................131
APPENDIX C .....................................................................................................................................................132
APPENDIX D .....................................................................................................................................................137
iv
LIST OF FIGURES
Figure I.1. Design and Operation Knowledge Base ................................................2
Figure I.2. Self Explaining Road with Forgiving Roadside.......................................3
Figure I.3. Procedure for Roadside Infrastructure Design .......................................4
Figure 1.1. Examples of Safety Barrier Terminations, Blunt End, Ramped not
Flared End..............................................................................................8
Figure 2.1. Relationship between the size of the clear zone and the accident
reduction [5] .........................................................................................21
Figure 2.2. Distribution of fatal accidents against fixed obstacles according to the
distance to the edge of the carriageway [6]..........................................21
Figure 2.3. Exit angle on a straight road.................................................................23
Figure 2.4. Relationship between the trajectory of a vehicle running off the
carriageway and the travelled speed. Friction coefficient 0.7 ...............23
Figure 2.5. Exit Angle for a Curve (no steering) .....................................................24
Figure 2.6. Relationship Between Curve Radius and Exit Angle............................24
Figure 2.7. Relationship between Exit Speed and Exit Angle (theoretical f=0.7,
d=2m)...................................................................................................25
Figure 2.8. Definition of Vehicle Exit Angles...........................................................25
Figure 2.9. Cumulative distribution of initial departure angle from road .................26
Figure 2.10. Magnitude of lateral sliding ..................................................................27
Figure 2.11. Relationship between Exit Angle And Exit Speed (Reconstructed)......27
Figure 2.12. Distribution of exit speed for initial run-off-road event ..........................29
Figure 2.13 Vehicle Course on a Slope...................................................................32
Figure 2.14. Set-back Information for Obstacles recorded in RISER Detailed
Database..............................................................................................34
Figure 2.15 Cumulative Distribution of Struck Obstacles ........................................34
Figure 3.1. Example of a Recovery Zone/Hard Shoulder in Spain .........................37
Figure 4.1. Strategy for Hazard Countermeasures.................................................48
Figure 4.2 Risk Information for Various Struck Objects.........................................49
Figure 4.3. Passive Safety Infrastructure ...............................................................50
Figure 4.4. Rigid Lighting Columns protected with Guardrail .................................50
Figure 4.5. The culvert is protected with a short guardrail but the gantry pole and
the old rigid lighting column are unprotected (Photo, HUT)..................51
Figure 4.6. Examples of the distributed hazards. Left: unprotected rock cut very
close to the carriageway. Right: untested guardrail with concrete posts
and too low positioning (Photo, HUT)...................................................51
Figure 4.7. Example of a Clear Safety Zone when trees have been removed
(Photo, HUT) ........................................................................................52
Figure 4.8. Example of hazardous, unshielded bridge pier and bridge abutment...53
Figure 4.9. Hazardous underpass on motorway, notice too short safety barriers...53
Figure 4.10. Underpass opening on motorway median protected with safety barriers.
(Photo, HUT) ........................................................................................54
Figure 4.11. Examples of culverts in minor road junctions: traditionally the ends are
vertical and the side slopes of the minor roads steep which makes them
very hazardous objects in collisions (Photo, HUT) ...............................54
Figure 4.12. Errant vehicle is prevented to impact the end of the culvert by adequate
set back of the culvert. Safe set back of the culvert depends on the
width of the ditch and width of the safety zone (figure: Finnish Road
Administration; TYLT 6800-6870). .......................................................55
v
Figure 4.13. Bevelled culvert ends in minor road junctions (Photo, HUT) ................55
Figure 4.14. Bevelled end of a concrete culvert: safer design than traditional vertical
end shape (Photo, HUT) ......................................................................56
Figure 4.15. Left: Break-away wooden column; glued and laminated. Right: Yielding,
energy absorbing steel column (Photo, HUT) ......................................57
Figure 4.16. Left: old wooden lighting column. Right: old steel lighting column, both
are modified to break away at impact.(Photo, HUT).............................57
Figure 4.17. Unprotected gantry poles in the safety zone are fatal hazards (Photo,
VALT)...................................................................................................58
Figure 4.18. Gantry pole shielded with safety barrier, note that the use of unflared,
turned down barriers is not recommended (Photo, HUT) .....................59
Figure 4.19. Small break-away sign support ............................................................59
Figure 4.20. Fill slope with a safety barrier, height exceeds 3.0 m...........................60
Figure 4.21. Recoverable ditch shapes: 1:4 fore slope, rounded bottom and 1:2 back
slope. (Photo, HUT) .............................................................................61
Figure 4.22. Recoverable fill slope; bottom of the ditch is rounded, no hazardous
obstacles in fill slope. (Photo, HUT) .....................................................61
Figure 4.23. There is evident need for better roadside safety because of trees and
steep slope. In this road section the only solution for better roadside
safety is safety barrier – by choice the guardrail with aesthetical value62
Figure 4.24. Rock face cutting removed from the safety zone (Photo, HUT) ...........62
Figure 4.25. Rock face cutting shielded with safety barrier (Photo, HUT) ................62
Figure 4.26. Left: too short guardrail, Right: lengthened guardrail (new section with
weaker posts). (Photo, HUT)................................................................63
Figure 4.27. Left: good solution of transition structure, the guardrail is overlapped
with the concrete barrier and fastened with steel bolt. (Photo, HUT)
Right: poor solution of the transition from steel guardrail to the concrete
pillar of the bridge parapet. (Photo, VALT) ...........................................63
Figure 4.28. Energy absorbing guardrail terminal and flared guardrail terminal
(Photo, HUT) ........................................................................................64
Figure 5.1. Example of Crash Cushions.................................................................67
Figure 5.2. Safety Barrier Terminal ........................................................................68
Figure 5.3. Transition .............................................................................................68
Figure 5.4. Arrester Bed .........................................................................................69
Figure 5.5. Break-Away Utilities .............................................................................69
Figure 5.6 Impact Configuration ............................................................................71
Figure 5.7 Comparison of Lateral Kinetic Energy for the Different Containment
Levels...................................................................................................73
Figure 5.8. Barrier and Vehicle Deflection Parameters .........................................75
Figure 5.9 Vehicle Redirection Requirements .......................................................76
Figure 5.10. The culvert is protected with a short guardrail but the gantry pole and
the old rigid lighting column are unprotected (Photo, HUT)..................79
Figure 5.11. Selection Criteria for Safety Barrier Containment Level in Germany
(RPS 2003) ..........................................................................................80
Figure 5.12 Lateral Distance Behind a Road Restraint System ..............................82
Figure 5.13 Hazard Length and Dimensions for Road Restraint Systems ..............84
Figure 5.14 Approach Length to Shield Hazards.....................................................84
Figure 5.15 Review of Multiple Hazards..................................................................85
Figure 5.16 Possible Vehicle Motions Behind Safety Barrier ..................................85
Figure 5.17 Design Solutions ..................................................................................86
vi
Figure 5.18 Length of Need Prior to a Hazard.........................................................86

Figure 5.19 German Dimensions for Road Restraint Systems................................87
Figure 5.20 Spanish Dimensions for Road Restraint Systems................................88
Figure 5.21 Failure in end sections .........................................................................90
Figure 5.22 Comparison between the behaviour of safety barriers with (green) and
without (red) end terminal fixation ........................................................91
vii
LIST OF TABLES
Table 1.1. Distribution of Struck Infrastructure Types in RISER Statistical
Database..............................................................................................11
Table 1.2. Distribution of Struck Infrastructure Types in RISER Detailed Database12
Table 1.3. Type and Characteristics of Post/Pole Impacted (RISER Detailed
Database).............................................................................................12
Table 1.4. Characteristics of Impacted Sloping Ground (RISER Detailed
Database), measures in brackets are for accidents involving vehicle
rollovers only ........................................................................................13
Table 1.5. Existing National Guidelines, figures when trees are considered
hazardous (Y=Yes) ..............................................................................15
Table 1.6. Existing National Guidelines, when posts/poles are considered
hazardous (Y=Yes) ..............................................................................15
Table 1.7. Existing National Guidelines, when rock and concrete objects are
considered hazardous (Y=Yes)............................................................16
Table 1.8. Existing National Guidelines, when sloping ground are considered
hazardous (Y=Yes) ..............................................................................16
Table 1.9. Point Hazard Characteristics for Serious or Fatal Injuries in the RISER
Detailed Database................................................................................17
Table 1.10. Distributed Hazard Characteristics Identified in the RISER Detailed
Database..............................................................................................18
Table 2.1. National Criteria for Dimensioning the Safety Zone .............................22
Table 2.2. Rolling resistance coefficient................................................................31
Table 2.3. Theoretical Safety Zone Widths ...........................................................33
Table 3.1. Recovery Zone Width on Non-Motorway Roads ..................................44
Table 3.2. Recovery Zone on Motorways..............................................................45
Table 5.1. Vehicle Impact Test Criteria .................................................................72
Table 5.2 Containment Levels for Longitudinal Barriers.......................................72
Table 5.3. Impact Severity Levels [1] ....................................................................74
Table 5.4. Benefit of some RRS Measures of Road Safety Programme in the USA
(FHWA, 1993) ......................................................................................76
Table 5.5. Benefit/Cost of some Infrastructural Safety Remedial Measures in
Spain....................................................................................................77
Table 5.6. Effects on accidents of guardrails along the roadside. Percentage
change in the number of accidents [5] .................................................77
Table 5.7 RISER Review of Installation Criteria for Road Equipment ..................77
Table 5.8 Critical Distances between Carriageway and Hazards [9]....................81
Table 5.9 Safety Barrier Lengths (L2) Before and After Hazards .........................87
Table 5.10 Barrier Approach Lengths for Spain (Figure 5.20a)..............................88
Table 5.11 Angled Barrier Approach Lengths for Spain (Figure 5.20´b) ................88
Table 6.1. Variable Matrix for Accident Data Collection ......................................100
viii
INTRODUCTION
The relevance of single vehicle accidents (SVA) (also known as run-off-road – ROR
or roadside accidents) as a traffic safety issue can be demonstrated in a review of
European accident data. In 1998, 33.8 % of all fatalities in the European Union were
the result of single vehicle collisions (Eurostat).The data collected on single vehicle
accidents in the "Roadside Infrastructure for Safer European Roads" (RISER) project
represented about 10 % of the total road accidents reported for the respective
countries. If the data is restricted to comprise only fatal accidents, then 45 % of all
fatal accidents are SVA. Worse yet, the cases collected in RISER do not represent
100 % of single vehicle accidents and indicate that even higher fatality figures may
occur. The over-representation of SVA fatalities (10 % of accidents producing 45 %
of all fatalities) cannot be ignored when developing road safety plans.
The objective of the RISER project was to develop best practice guidelines that can
be a foundation for national policy and guidelines. The project has synthesized the
data and expertise from nine European countries: Austria, Belgium, Finland, France,
Germany, the Netherlands, Spain, Sweden, and the United Kingdom. The following
information does not reflect one national practice for structuring the roadside area.
The documents outlining the technical functions and requirements for roadside will
need to reflect the conditions of the local road network. Climatic, geographical, and
demographic characteristics for the road sections of interest must be developed by
local authorities.
Principles for Safe Road Networks

The goal of any road authority, road operator and road designer is to provide the best
service for the travelling public. The capacity of the system must allow the public to
reach their destination in a timely manner without creating a safety risk for the vehicle
occupants. The challenge for road infrastructure designers is to provide the
appropriate road and roadside constructions knowing that the driver is not a perfect
operator but is susceptible to mental and physical shortcomings that will create
situations leading to accidents.
Figure I.1 illustrates the first requirement for the safe road environment – road safety
infrastructure systems must be designed using the existing experience from existing
roads and monitored to ensure the system supplies the functions demanded by the
existing traffic conditions. A life-cycle cost analysis comprises the installation,
maintenance, and accident (societal) costs. Without a complete documentation of
system performance as depicted in Figure I.1, life-cycle cost analyses are not
possible.
1
Historical Continual
knowledge from Road monitoring of road
similar road Infrastructure operational and
environments safety conditions
Figure I.1. Design and Operation Knowledge Base
The Self Explaining Road and Forgiving Roadside Philosophies

The physical layout of the road and roadside environment provides visual clues and
signals to the vehicle operators. The type of road – a motorway or a small forest road
– should be clear to the driver without explicit signs. The road width, types of lane
markings, roadside geometry, etc. should provide clues to the driver that signal
appropriate driving speeds and lane positioning for the type of road and indicate what
type of other road users to expect. As an example, the Netherlands advocate the
principle of Sustainable Safety with a road network structure of a limited number of
clearly distinguishable road categories, each with its own function (flow function,
distributor function, and access function). This explicit function determines the way
the road and roadside environment should be designed to induce the appropriate
road user behaviour (including speed choice and lateral positioning) on the one hand,
and the required safety level of roadside elements on the other. The process of
implicitly informing the driver of appropriate driving conditions is necessary to achieve
a "Self Explaining Road". The Self Explaining Road concept advocates a road and
traffic environment that elicits safe driving behaviour simply by its design. By
maintaining consistent and uniform road and roadside design procedures, the road
meets drivers’ expectancy and drivers can anticipate changes in operating conditions
(expecting at-grade intersections, pedestrians, etc.) even if they have not observed a
sign stating that, for example, the motorway has ended. The physical layout of the
road environment explains the driving context.
There are aspects of the roadside infrastructure that provide visual clues to the driver
that can help "explain" appropriate driving conditions for a road section, but also the
opposite may be true (for example: a misleading line of guidance by trees that
deviate from the road). The human factors that cause roadside elements to influence
traffic conditions should not be ignored in the design of safe roadsides. The two
critical traffic elements that can be influenced by roadside objects are the traffic
(vehicle) speed and lateral positioning. An overview of Human Factor aspects are
presented in Appendix A. Relevant Human Factors design aspects are also provided
throughout this document.
2
A second design philosophy that must be highlighted is the "Forgiving Roadside".

This is simply the requirement that the roadside environment should not contain
dangerous elements that will seriously injure or kill vehicle occupants that have
unplanned trajectories off the carriageway. A fundamental component of this
philosophy is the definition of an obstacle-free safety zone beside the carriageway.
Since this is economically and functionally not always achievable, the introduction of
passive safety equipment like road restraint systems (safety barriers), crash
cushions, and energy absorbing (or break-away) posts to protect vehicles and
minimize the consequences from dangerous impact hazards. It is important to
recognize that all objects placed near a travel lane are potential impact hazards. The
proper engineering design of passive safety infrastructure ensures that any
subsequent impact with a safety device is much less severe than the resulting impact
if the safety device was not in place.
It is difficult to find examples of roadways which possess both good self explaining
characteristics and good forgiving roadsides, As an illustrative example, Figure I.2
shows a road section with separated travel lanes, a pedestrian overpass, wide hard
shoulders with gentle side slopes and metal safety barriers protect the overpass
supports. The road environment is clear to a western European driver – higher travel
speeds (over 80 km/h) are permitted, no pedestrians will appear on the road and no
left turning traffic will cross the carriageway. The Forgiving Roadside is partially
achieved by the protection of objects (concrete pillars) close to the road with a safety
barrier. The shallow ditch leading up to the pillars is blocked by the end treatment of
the barrier and "closes the window" where a vehicle can move behind the barrier.
The wide hard shoulder provides an area for modest vehicle manoeuvres in case the
driver is not attentive. Terrain beyond the paved surface is smooth and free of fixed
objects for a distance of about 10 m (to the electrical transmission tower).
Figure I.2. Self Explaining Road with Forgiving Roadside
What should also be noted in Figure I.2 is that the median barriers are close to the
lighting columns. If there is not enough space between the barrier and the columns,
the safety performance of the median barrier can be degraded during a crash. The
3
buried terminal in the ditch may also present a hazard. Buried terminal ends can
cause vehicle rollovers and they should not be placed too close to the travel lanes.
Through the application of the knowledge from RISER and other ongoing research
activities, the technical requirements for achieving Self Explaining Roads and
Forgiving Roadsides will become clearer. Thus each road section, like Figure I.2, can
be analysed to improve safety.
The information in this document should assist road designers, safety auditors and
reviewers, maintenance personnel, and government officials to build and maintain
safe road sections. The goal of this document, and a complementary document for
Maintenance and Operations Guidelines, is to provide an overview of identifying
potential safety issues as well as provide suggested countermeasures to improve the
road traffic environment.
Framework for Roadside Infrastructure Design

The information in this document is structured to follow the analysis procedures for
best practice guidelines. This structure may also be useful when conducting road
safety audits for roadside safety. The design and analysis process is presented in
Figure I.3. The first step in evaluating roadside infrastructure safety needs is to
identify the types of fixed obstacles (or objects) adjacent to the road. These objects
must be identified and inventoried in terms of type of obstacle and the lateral position
to the road. From this information, the appropriate clear and recovery zone criteria
should be defined for the road section to identify the critical areas in the roadside
environment. From this information the first engineering judgements begin –
identifying which obstacles are located in sensitive (affecting safety) areas and
determining if these obstacles are a safety hazard.
Obstacle
Chapter 1 Identification
Safety Zone Recovery Zone

Chapter 2&3
Criteria Criteria
Hazard No
Evaluation
Yes
Chapter 4 Hazard Yes

Removal
No
Hazard Yes Final

Monitoring
Modification Solution
Chapter 5 No Chapter 6
Hazard Yes
Protection
Figure I.3. Procedure for Roadside Infrastructure Design
4
Chapter 1 describes the special types of roadside obstacles and provides criteria that
can be used to determine if these objects are hazardous or not.
Chapters 2 and 3 provide the best practice information regarding the dimensions of
suitable safety and recovery zones.
Chapter 4 provides practical information important for the technical (engineering)

design guidelines for different roadside environments. These guidelines are intended
to provide best practice for determining the suitable type of roadside infrastructure.
Specifications for type, placement, and the influence of the features on traffic are
provided in this and the supporting technical documents.
Chapter 5 lists information outlining the performance features of different types of

safety equipment which can be installed.
Finally the need for monitoring the performance of the road network is described in
Chapter 6. The importance of collecting, storing, analysing, and reporting the
accident data are highlighted.
5
CHAPTER 1: DEFINITION AND IDENTIFICATION OF

ROADSIDE AND MEDIAN HAZARDS
1.1 Introduction
The general nature of a run-off-road (ROR) or roadside accident is that the vehicle
will run-off the road into the roadside and has at least one collision with either
roadside equipment or the roadside itself. Therefore, one of the main factors which
determine the severity of these types of accidents is the layout of roadside and the
type of objects present which potentially could become collision hazards.
Even in the design of new roads, the placement of certain objects (such as lighting
and utility poles, sign posts, boundary fences, bridge piers etc.) in the road-
side/median can often not be avoided. There are also the natural roadside objects
which can not be moved from the roadside (for example, trees with aesthetic/cultural
value, water courses etc.). The steepness of the roadside slopes can also be a
contributing factor to roadside safety.
Due to the poor energy-absorbing qualities of many roadside objects, an impact

would result in serious damage to the vehicle and more severe injuries to occupants.
Therefore, the aim of this chapter is to identify which objects often found on the
roadside are potentially the most hazardous, and what part of their characteristics
makes them hazardous such as size and frequency etc.
1.2 Definitions of Roadside and Median Objects (Point and

Distributed Hazards)
This section includes a list of definitions of roadside and median objects frequently
found within the roadside. They include objects which were identified in the RISER
detailed and statistical databases as being hazardous objects which, when impacted,
can lead to serious occupant injuries (see section 1.4). They are also referred to in
many current guidelines across Europe as being roadside hazards. They have been
divided into 'point' and ' distributed' objects (see glossary for further details and
photos).
Point objects includes narrow items in the roadside that could be struck in a collision
for example trees, all types of bridge supports, lighting poles, utility poles, sign posts,
terminations of barriers, etc.
Distributed objects, also known as ' continuous obstacles'

, are potential hazards
which extend along a length of the roadside, such as all types of embankments,
ditches, rock face cuttings, retaining walls, safety barriers not meeting current
standards, forest and closely spaced trees.
1.2.1 Point Objects
Trees and Tree Stumps

Trees are prevalent in roadsides, particularly in rural locations, and can be
very unforgiving during an impact, absorbing very little of the energy created
by the impact. Although individually they are point hazards, they can also
6
arguably be distributed hazards, especially in the cases where there are

rows of trees along the roadside. Often, older established trees are
protected by preservation orders, so they cannot be removed or relocated
from the roadside. Therefore, methods of protecting the vehicle and
occupants from the tree are required.
Poles and Posts

These include lighting poles, utility poles, gantry poles, high mast lighting
columns and sign supports, and all posts of road signs which are not tested
according to EN12767. All 'passively safe'posts and poles are deemed not
to be roadside hazards.
Rocks and Boulders

Rocks and boulders are large rounded masses of rock lying on the surface
of the ground or embedded in the soil in the roadside, normally detached
from their place of origin.
Bridge and Overpass Structures

A bridge pier or bridge pillar is an upright structure, often a series of
columns, which supports a bridge or an overpass. It can be located in the
central reserve or in the roadside.
An abutment is the end of a bridge or tunnel wall.
Culverts and Culvert Ends

A culvert is a structure to channel a water course. It can be made of
concrete, steel or plastic.
A culvert end is the end of the channel or conduit, also normally a concrete,
steel or plastic structure.
Underpass
A roadway or other path passing under the main roadway.
Other Point Hazards

Rivers, other water hazards and railway lines passing under the roadway,
drainage pipes, headwalls.
Hazardous Safety Barrier Ends

Poorly designed or positioned barrier ends, including end treatments which
do not fulfil the requirements of EN 1317. In RISER Deliverable D05 [9]
examples of barriers with blunt ends and terminations which do not flare
away from the carriageway are shown.
7
Figure 1.1. Examples of Safety Barrier Terminations, Blunt End, Ramped not Flared End
1.2.2 Distributed Objects
Slopes and Embankments

A' slope'is a general term used for embankments. It can also be used as a
measure of the relative steepness of the terrain expressed as a ratio or
percentage. Slopes may be categorized as negative (fore slopes) or positive
(back slopes) and as parallel or cross slopes in relation to the direction of
traffic.
An embankment is a general term for all sloping roadsides, including cut
slopes (upward) and fill slopes (downward).
A cut slope is an earth embankment created when a road is excavated
through a hill, slopes up from the roadway
A fill slope is an earth embankment created when extra material is packed to
create the road bed, slopes down from the roadway.
Drainage Features
Ditches are drainage features that run parallel to the road. Excavated
ditches are distinguished by a fore slope (between the road and the ditch
bottom) and a back slope (beyond the ditch bottom and extending above the
ditch bottom).
A drainage gully is a structure to collect water running off the roadway.
Rock and Concrete Hazards

Rock face cuttings are created for roads constructed through hard rock
outcroppings or hills.
A retaining wall is a wall that is built to resist lateral pressure (especially a
wall built to support or prevent the advance of a mass of earth).
Other distributed rock and concrete hazards include buildings, rock fences.
Non-safety Fences
Fences on the roadside mainly used to identify boundary edges. Include
wooden, metal and wire (wildlife) fences.
Old Design/Poorly Installed Safety Barriers

Safety barriers which do not fulfil the requirements of EN1317, designs
which have been documented with poor performance in real world crashes,
and those which have been incorrectly installed are hazards.
8
Other Distributed Hazards

Water hazards, such as lakes, reservoirs, the sea, rivers running parallel to
the road.
Parallel roads, railways, rows of trees and forests.
1.3 Overview of Research

1.3.1 Previous Studies
The following list is an overview of a number of previous research studies which
investigated collisions involving roadside objects. This search has been restricted to
an investigation of published reports from a variety of sources regarding collisions
with roadside objects. The summaries reflect the range of information taken from
these sources:
The Highway Perspective of Side Impacts [1]

Fixed object crashes in the USA account for 21 % of all fatalities and 15 %
of all injuries. Over 160,000 people are involved in accidents where the side
of the passenger car impacts a fixed roadside object such as a tree, utility
pole or light support and around 1 in 100 occupants are killed in these
collisions. The annual cost is around $3 billion.
Forgiving Roadsides [2]
This report reviews the extent of the problem involving collisions with street
furniture for a number of European countries. In Finland, collisions with road
equipment accounted for 24 % of all fatal accidents, with trees and utility
poles most frequently struck. In France, collisions with road equipment
accounted for 31 % of all fatal accidents, with trees most frequently struck.
In Germany, collisions with road equipment accounted for 18 % personal
injury accidents, and 42 % road deaths, again with trees being the most
frequently struck object. In Great Britain, collisions with road equipment
accounted for 18 % of all fatal accidents, with trees and lamp columns most
frequently struck. In the Netherlands, collisions with road equipment
accounted for 22 % of all fatal accidents. In Sweden, collisions with road
equipment accounted for 25 % of all fatal car occupant accidents, with trees,
safety fence, and utility poles most frequently struck.
Fatality Facts 2003: Roadside Hazards [3]
More than 20 % motor vehicle crash deaths result from a vehicle leaving the
road and impact a fixed object such as a tree or utility pole. These crashes
occur in both urban and rural situations but are more common on rural
roads. Around 20 % fixed object crashes also involve rollovers, and 20 %
involve occupant ejection. Trees are the most commonly struck objects
accounting for around half of all fatalities (4,522 fatal crashes in 2003).
Road Environment Safety Update, Fatal Roadside Object Study [4]
Roadside object crashes contributed to 33 % of all fatal crashes in NSW in
2000-01. Trees, utility poles and embankments were the most commonly
struck objects. In secondary impacts water bodies also featured significantly.
In a significant number of fatal crashes the first impact was with a frangible
object such as a fence or guidepost, but then other objects struck as
secondary impacts contributed to the severity of outcome.
9
Strategic Highway Safety Plan [5]

This document identifies 22 goals to pursue in order to significantly reduce
highway crash fatalities. Goal 15 is Keeping Vehicles on the Roadway, and
Goal 16 is Minimizing the Consequences of Leaving the Road. The goals
concentrate on, amongst other factors, crashes with trees in hazardous
locations.
EuroRAP 2005: British Results [6]
The four main types of accidents on single carriageway roads are: head-on
collisions, run-off-road accidents involving collisions with roadside objects
e.g. trees, collisions at junctions and accidents involving vulnerable road
users e.g. pedestrians and cyclists.
Single Vehicle Loss of Control Collisions and Passive Safety [7]
Data for Great Britain show around 20,000 injury accidents each year with
off road objects. The trend is constant over the past 15 years, whilst other
accident types have reduced. The collisions have a high severity index, with
24 % collisions resulting in serious or fatal injuries. Trees are the most
frequently struck objects, followed by crash barrier, and lamp posts. There is
however, a large group of nearly 7,000 collisions where the object struck is
“unknown”. Trees also have the highest severity index, with 33 % of all
collisions resulting in serious or fatal injuries.
Fatal Accidents Against Fixed Objects, CEESAR [8]
CETE Normandie-Centre states that on rural two-lane roads, fatal accidents
against fixed objects account for 30 % of the total fatal accidents. Of these
fatal accidents, 44 % were against fixed objects occur on straight lines and
56 % in curves. Trees are the main impacted objects (60 %), utility poles
accounting for 10 %, and safety fences 1 %. When the distance of the
obstacle from the edge of the road is 2 m, the fatal accidents account for
43 % and 78 % when it is < 4 m. On motorways, fatal accidents against
safety barriers (metallic fences, fence terminations and central concrete
barriers) account for 44 %.
1.4 RISER Analysis

1.4.1 RISER Statistical Database
The RISER statistical database holds nearly 265 000 single vehicle accident cases
from seven European countries (Austria, Finland, France, the Netherlands, Spain,
Sweden and the United Kingdom). In 67 % of these cases, it was known that the
vehicle struck an object. Table 1.1 shows the percentage of cases where an object
was known to be struck in the accident.
10
Table 1.1. Distribution of Struck Infrastructure Types in RISER Statistical Database

% of ALL
% of Each Struck Object
accidents
Struck Object ALL Severities Fatal Serious Slight
Tree 11.1 17 39 44
Post 8.2 9 31 61
Safety Barrier 15.5 6 20 74
Ditch 10.6 8 32 60
Other natural object1 0.9 7 32 61
Other man-made structure2 8.0 11 33 56
Other 12.5 - - -
Unknown/None/NA 33.2 - - -
1
Include rock faces, stones, water submersion etc.
2
Include signs, concrete structures, non-safety fences, culverts, underpasses etc.
Safety barriers appear to be the object most often impacted in the accidents.
However, this does not necessarily mean that barriers are more dangerous than
other roadside objects, because within the safety barrier object type group all safety
barriers are taken into account, even those that not fulfil the current standards. The
level of exposure (i.e. the number of these objects located in the roadside and the
opportunity to come into contact with one) is also not taken into account. Of the
objects more frequently impacted, both trees and ditches were impacted in over 10 %
of all the accidents sampled and posts in more than 8 %.
When injury severity is taken into account, the results show that a tree was impacted
in more fatal accidents than other struck objects (25 % of all fatal accidents). In
addition, when looking at tree accidents only, 17 % were fatal accidents, a greater
proportion than any other object type (see Table 1.1). Nearly three-quarters of
accidents involving safety barriers involved only slight injuries, which show that
although safety barriers were involved in a higher proportion of single vehicle
accidents than other objects, the impacts generally resulted in only minor injuries.
1.4.2 RISER Detailed Database

The RISER detailed database contains 211 in-depth single-vehicle accidents cases.
The main type of objects struck in these cases were trees, post & poles, rock &
concrete objects (rock faces, bridge supports, boulders etc.), sloping ground
(embankments, ditches etc.), non-safety fences and safety barriers. An overview of
the information recorded for the types of roadside infrastructure struck are presented
in Table 1.2
11
Table 1.2. Distribution of Struck Infrastructure Types in RISER Detailed Database

Rock/ Non-
Post/ Sloping Safety
Tree concrete safety
pole ground barriers
object fence
No. of Accidents 38 45 34 71 11 72
No. of impacts/ interactions 50 51 37 88 11 86
1 2
Only impact 30 20 15 44 1 41
Fatal or serious cases 26 15 12 22 0 21
Non-motorways 28 16 12 35 1 18
Single carriageways 25 12 10 30 1 6
Single run-offs 26 18 13 38 1 29
Greatest set-back in fatal 3 4 5 No fatal No fatal 5
6.8 4.5 6.4 1.5
accident [m] (6.8 slight) (7.2 slight)
Smallest diameter in fatal
0.3 0.2 - - - -
accident3 [m]
After impact distances [m] 0 - 44 0 - 39 0.6 - 32 0 - 49 1 0 - 208
1
Cases where specific object was the only impact in the accident
2
Only or '
main'impact in crash (21 cases where tree impact was the only impact)
3
Seatbelt worn
4
Seatbelt not worn but would have made little difference as it was a side impact
5
Seatbelt not worn
Tree Impacts Only

The majority of tree impacts only cases involved fatal or serious injuries. The
narrowest tree diameter involved in a fatal collision being 0.3 m (0.2 m where no
seatbelt was worn) and the largest set-back distance from the edge of the road was
6.8 m (10.8 m no seatbelt worn). All fatal accidents involved impacts speeds of
70 km/h or greater (where speed data was known). When serious accidents were
also included, impact speeds were 40 km/h or greater.
Post/Pole Impacts Only

In accidents involving posts and poles, posts/poles as narrow as 0.2 m were
impacted (0.11 m where no seatbelt worn), and impact speeds in fatal accidents were
70 km/h or above. Impact speeds in serious and fatal accidents were 40 km/h or
above. Side impacts with poles and posts were more often fatal accidents than frontal
impacts.
Table 1.3. Type and Characteristics of Post/Pole Impacted (RISER Detailed Database)
Minimum Accident Maximum
Material Post/pole type No. of cases
diameter [m] Severity Set-back [m]
Concrete Utility 3 0.25 Serious 3.3
Metal Gantry 3 0.35 Fatal -
51 0.201 Fatal1 4.51
Lighting 2 2 2
2 0.22 Fatal -
Sign 5 0.11 Fatal 2.4
2 2 2
Wood Lighting 1 Unknown Fatal -
Utility 1 0.22 Slight 2.1
1
Not passively safe post/pole
2
Passively safe post/pole
Impacts with Rock/Concrete Objects Only

The majority of the 15 accidents involving impacts with only rock or concrete objects
in the database were fatal or serious. The greatest set-back in a fatal accident
12
involving a rock face/boulder was 6.4 m. The greatest set-back in a fatal accident
involving a bridge pier or abutment was 1.5 m.
There were only 5 collision speeds recorded for impacts with rock/concrete objects.
The lowest impact speed which resulted in a fatal or serious accident was 50 km/h.
Impacts with Sloping Ground Only (Embankments, Ditches etc.)

Table 1.4 shows minimum slope measurements and maximum set-back distances for
serious or fatal accidents involving sloping ground. The measures in brackets are for
accidents involving vehicle rollovers only. Impact speeds were above 40 km/h in all
serious and fatal cases (where impact speeds were recorded). Of all sloping ground
impacts 50 % were fatal or serious.
Table 1.4. Characteristics of Impacted Sloping Ground (RISER Detailed Database), measures
in brackets are for accidents involving vehicle rollovers only
Slope type Height [m] Gradient Set-back1 [m]
Upward from road level 0.4 (0.6) 1:3 6.1 (5.2)
Downward from road level 2.0 (5.0) 1:7 4.6 (4.6)
Back slope of ditch/gully 1.0 (1.0) 1:2 5.1 (5.1)
Fore slope of ditch/gully 1.0 (0.7)2 1:3 6.8 (6.8)
1
Minimum measure from slight accident. No set-back measurements available for fatal or serious cases.
2
Non-injury case
All Impacts with Culverts

There were 10 cases in the database which involved an impact with a culvert,
including cases where other objects were impacted. Of these cases, 8 were fatal and
1 was serious. In addition, the culvert was the final impact in 8 of the cases. The
culvert was also the main impact in the majority of these cases. Only 2 cases had
set-back measurements, 3.5 m (slight) and 10.6 m (fatal) and only 2 cases had
calculated impact velocities, 77 km/h (slight) and 100 km/h (fatal).
All Impacts with Non-Safety Fences

An impact with a non-safety fence was involved in 11 cases including cases where
other objects were impacted. In only 1 of these cases, the impact with the non-safety
fence was the main cause of the injuries. In this case, a broken rail from the wooden
fence penetrated the windscreen and caused serious facial and head injuries to the
driver. No set-back measurements are available for this case. However, this
highlights the potential for the posts of wooden fencing to be as hazardous as
wooden utility poles during an impact.
Although there are very few impacts with metal fences in the database, metal fence
posts also have the potential to be as hazardous as non-break-away or non-energy
absorbing posts/poles.
All impacts with wire boundary/wildlife fences did not directly result in serious injuries.
It was previous or subsequent impacts with other types of objects that caused the
injuries.
Impacts with Safety Barriers Only

There are 41 cases where only a safety barrier was impacted in the accident, of
which 21 resulted in fatal or serious injuries. Although it could be concluded from this
13
that safety barriers themselves are hazards, it should be noted that the detailed
database is biased towards fatal accidents since many cases were obtained from
fatal accident databases. There were also other contributory factors which led to the
severity of many of the accidents with safety barriers.
For example, in a number of cases:
The vehicle was a truck or motorcycle. Therefore the barrier installed may
not have been appropriate to contain these types of vehicles.
The barrier impact led to a vehicle rollover or spin and a resulting secondary
impact which caused the occupant injuries.
However, it is inevitable that barriers will often be impacted in accidents because of

their quantity and frequency in the roadside. A number of issues were also identified
in all accidents involving safety barriers, for example:
The length of barrier (vehicle travelling behind barrier and impacting hazard
or rear of the barrier).
Poorly installed or maintained barriers, often resulting in vehicles going over
or breaking through the barrier.
Barrier termination – Poorly designed or poorly positioned barrier termina-
tions can become a roadside hazard, resulting in launches off the termina-
tion or impacts with blunt barrier ends.
Older design barriers can be hazards, particularly those which do not comply
with EN1317.
Barriers can be hazardous to occupants of large trucks or riders of
motorcycles if the barriers are not designed for motorcycle or truck impacts.
1.4.3 Current European Guidelines

In addition to examining single vehicle accident data, it is also important to take into
consideration guidelines covering roadside safety and road restraint systems which
are currently used across Europe. By reviewing current guidelines, it can be
determined what roadside and median hazards are already identified as hazards
across Europe and under what circumstances. Forthcoming UK Standards
concerning passive safety, IRRRS and Risk assessment is presented in Appendix B.
Review of Current Guidelines in Europe

As part of RISER, a review of current European Design Guidelines for Roadside
Infrastructure was carried out for seven countries and the findings are reported in
Deliverable 5 [9]. In this report, the main documents reviewed which identify hazards
and objects on the roadside are listed. Table 1.5 to Table 1.8 show an overview of
the type of roadside objects which are listed as being hazards in the countries
reviewed.
Trees
In all seven RISER participating countries, trees were considered as roadside
hazards in the design guidelines. However, the minimum diameter when a tree was
considered hazardous varied from 0.1 m to 0.3 m.
14
Table 1.5. Existing National Guidelines, figures when trees are considered hazardous
(Y=Yes)
Hazard FI FR DE GB1 NL ES SE
>0.1 2
Trees/ tree stumps [m] >0.1 >0.07 >0.3 >0.08 Y >0.1
stumps >0.2
Rows of trees/forests
Y Y Y Y Y
hazardous
1
< 4.5 m from carriageway edge line
2
Measured at a height of 0.3 m above ground level
Posts and Poles

Post and poles on the roadside of varying types were also considered as hazards in
a number of the countries. In Great Britain and Finland, traffic sign supports with a
minimum diameter of 0.09 m and 0.11 m respectively are considered as hazards. In
Spain, trees and poles over 0.15 m diameter are hazards depending on the distance
to the carriageway edge line.
Table 1.6. Existing National Guidelines, when posts/poles are considered hazardous (Y=Yes)
Hazard FI FR DE GB1 NL ES2 SE
Utility poles [m] Y Y Y >0.15 Y
Vertical, Overhead & Luminaire sign
Y Y Y Y >0.15
supports [m]
3
Sign gantry legs [m] Y Y >0.09 Y
>0.09
3
Posts of large signs [m] Y <1.5 m above Y
ground
<10 m from
High Mast Lighting columns [m] Y Y3
road edge
Steel columns/High voltage
Y Y Y3
electricity columns
3
Non-break-away poles [m] Y Y >0.09 Y Y
3
Traffic sign supports [m] 0.11 Y Y >0.09 Y
CCTV Masts Y
1
Unless stated otherwise, < 4.5 m from carriageway edge line
2
If the distance between the hazard and the carriageway edge line is less than a minimum value which varies between 4.5 and
14 m depending on the road layout.
3
If the post is not equipped with a passively safe feature that allows it to detach or bend upon impact.
Rock and Concrete Objects

The following rock and concrete objects found on the roadside are specified in
guidelines across Europe as being hazardous.
15
Table 1.7. Existing National Guidelines, when rock and concrete objects are considered
hazardous (Y=Yes)
Hazard FI FR DE GB1 NL ES SE
Rocks and boulders Y Y Y Y
Bridge piers/pillars/abutments Y Y Y Y Y Y
Tunnel entrances Y Y Y Y
Culvert ends/headwalls Y Y Y Y
Culverts and drainage pipes Y Y Y Y Y Y
<1.5 m
Rock face cuttings/rock fences Y Y Y Y Y
above ground
Retaining walls Y Y
<0.7 m set-back
Buildings/walls Y Y Y
<1:40
1
Unless stated otherwise, < 4.5 m from carriageway edge line
Sloping Ground
Roadside slopes and ditches of varying inclinations are deemed to be hazardous in
all of the reviewed countries. Minimum slope gradient varies from 1:3 to 1:8 and
minimum slope height from 0.5 to 6 m.
Table 1.8. Existing National Guidelines, when sloping ground are considered hazardous
(Y=Yes)
Hazard FI FR DE GB NL ES SE
>0.5 m >0.5 m 1
Ditch and drainage gullies Y Y >1:3 Y Y
>1:3 >1:4
Cut (upward) slopes and >1.0 m
1
embankments >1:1
Fill (downward) slopes and All >6.0 m or
embankments >1.0 m; >1:1
>2.0 m >4.0 m
All slopes and embankments Y Y >1:8 Y
>1:3 >2:3
1
See cut and fill slopes
Other Hazards
Many other hazards are also identified in guidelines across Europe, including:
Water hazards (rivers, lakes, canals, reservoirs, stilling ponds) (FI, DE, GB,
NL, ES, SE)
Underpasses (pedestrian, agricultural) (GB)
Property fences (FR, NL, ES)
Other roads and railway lines (FI, DE, GB*, ES, SE) (*Roads <10 m from
carriageway edge)
Electricity transformers (FR, GB, NL, ES)
Control cabinets (GB) and traffic counting stations (FR)
Hazardous storage installations (GB)
Road references points (FR, DE, NL, ES)
Old barriers and barrier terminations (FI, FR, NL, ES)
Central reserves with no safety barriers (FI, GB <10 m wide)
Curves in road (GB - Radius <850 m (with varying roadside slope gradient
and height ); ES - Radius <1500 m))
16
1.5 Summary
Using the findings from the statistical and detailed database analyses and the review
of current guidelines in Europe, definitions of roadside and median hazards have
been produced. This includes minimum measures, impact speeds and set-backs that
cause serious or fatal injuries for the accidents studied in RISER.
1.5.1 Proposed Measurements for the Definition of Hazards

Table 1.9 and Table 1.10 outline each roadside object and the characteristics which
define it as a roadside hazard (for example, size, location, frequency) according to
the detail database analysis and the review of the current guidelines.
NOTE: These MINIMUM measures are conclusions from the RISER analysis. If
guidelines within individual countries already include a greater margin of safety than
those stated here (for example, smaller diameters, smaller heights, less severe
gradients, slower speeds), then those national guidelines should also still apply.
Where possible, dangerous impact speeds have been identified from reconstructed
cases in RISER detailed database. The dangerous impact speed is the minimum
speed at which a hazard can be impacted and still cause serious injuries to the
occupants.
Given the limited cases (211) available in the detailed database, this data may not
fully identify the range of impact conditions that cause serious or fatal injuries.
Therefore the existing guidelines (see Table 1.5 to Table 1.8) should be consulted to
identify specific values for a country or region.
Table 1.9. Point Hazard Characteristics for Serious or Fatal Injuries in the RISER Detailed
Database
Dangerous
Diameter
Hazard impact speed Additional comments
[m]
[km/h]
Typically >0.1 in many
Trees and tree stumps >0.2 40
national guidelines
2
The following poles/posts
- Utility poles
- Standard lighting poles >0.2 40
(wood, metal and concrete)
- Posts of roadside signs >0.1 40
- Gantry/large traffic signs
- Supports/CCTV masts/High mast
lighting columns >0.1 40
- Supports/other high mast
posts/poles.
Rocks and boulders - -
Bridge piers/pillars/abutments 50
Culvert ends/ headwalls/drainage
-
pipes
Underpasses and other point hazards Including those at the foot of
-
(rivers, railway) an embankment
Blunt barrier terminations
and ramped ends which do
Safety barrier terminations -
not bend towards the
roadside (see Chapter 4)
1
Does not include '
passively safe'posts and poles.
17
In Table 1.10, for slopes (cut and fill), ditches and drainage gullies minimum slope
heights & gradients are given to identify when slopes becomes a hazard.
Table 1.10. Distributed Hazard Characteristics Identified in the RISER Detailed Database
Dangerous
Height/
Gradient impact
Hazard Depth Additional comments
[m] speed
[m]
[km/h]
Cut (upward) slopes >1.0 >1:1 40
In addition, ALL
Fill (downward) embankments 6 m high
>1.0 >1:1 40
slopes/embankments or more (i.e. ALL set-
backs).
Ditches and drainage gullies
>0.75 >1:3 40
(fore & back slope)
Any exposed rock face
Rock face cuttings/rock
50 cutting slopes <1.5 m
fences
above carriageway level.
Less than 1.5 m above
Retaining walls -
carriageway level.
Buildings/walls -
Wire wildlife/boundary
Non-safety fences - fences not considered as
hazards.
Barriers not compliant
with EN1317 and with
Old design safety barriers -
poor performance
records.
Same measures as for
Rows of trees/forests 40
individual trees.
Adjacent roads, railways,
-
water hazards
Median Hazards
Any of the previously defined point or distributed roadside hazards should
also be treated as hazards if present within a central reserve.
A central reserve on any roads with a speed limit above 70 km/h which has
a width of less than 10 m between opposing edges of the carriageway
should itself be considered a hazard.
Hazards at Curves
Although few conclusions can be made in this study regarding curves in roads and
their potential as hazards, it is apparent from the detailed database analysis and from
previous studies that curves in roads do impose an increased risk of a run-off-road
accident. It is more difficult for a driver to recover from a run-off in a curve than on a
straight section of road. Therefore, roadside hazards should be considered an even
greater risk when located near curved sections of roads.
Unfortunately, no conclusions can be made about the relationship between the radius
of curves and injury risk from the RISER accident data. Previous studies indicate the
higher risk of accidents in curves [10, 11, 12], therefore measures quoted in current
guidelines [9] are a useful starting point.
18
References
1 Reagan. J.A., The Highway Perspective of Side Impacts, Society for Automotive
Engineers International Congress and Exposition, Detroit, Michigan, USA, March 1-5,
1993
2 European Transport Safety Council (ETSC), Forgiving Roadsides - Research Paper
prepared for the ETSC Road Infrastructure Working Party, Brussels, 1998
3 Insurance Institute for Highway Safety USA, Fatality Facts 2003: Roadside Hazards,
www.iihs.org/safety_facts/fatality_facts/ roadside_hazards.htm
4 Roads and Traffic Department New South Wales, Fatal Roadside Object Study, Road
Environment Safety Update 20, Australia, 9 March 2003
5 American Association of State Highway and Transportation Officials (AASHTO), Run-off
Road Collisions, Executive Summary of the Strategic Highway Safety Plan, December,
2004
6 AA Motoring Trust, EuroRAP 2005, British Results, Tracking collisions and road
improvements, Monitoring motorcycle involvement, Benchmarking across Europe, March
2005
7 Proctor, S., Single Vehicle Loss of Control Collisions and Passive Safety, Traffic Signs,
Signals, and Lighting, the Passive Revolution Conference, Institution of Civil Engineers,
London, 22 June 2005
8 Les accidents mortels contre obstacles (fatal accidents against fixed obstacles). CETE,
CEESAR, France, 1995
9 D05: Summary Of European Design Guidelines For Roadside Infrastructure, European
Community R&TD Project, 5th Framework Programme “Growth”, Project "RISER"
GRD2/2001/50088, 2003
10 Caratéristiques routières et sécurité (road characteristics and safety), INRETS, 1986.
11 Accidents en virage sur route bidirectionnelle inter-urbaine, modélisation (Accidents in
curves on rural two-lane roads, modelling) CETE, SETRA, 1998.
12 SAFESTAR - Safety standards for road design and redesign, European Community
R&TD Project, 4th Framework Programme, Project reference RO-95-SC.0204, 1998
19
CHAPTER 2: SPECIFICATIONS FOR SAFETY ZONES
2.1 Introduction
Roadside safety addresses the area outside of the roadway and is an important
component of total road design. There are numerous reasons why a vehicle leaves
the roadway. Regardless of the reason, a forgiving roadside can reduce the
consequences of leaving the roadway. The ideal road has roadsides and median
areas that are flat and unobstructed by hazards.
2.2 Definition of Safety Zone

The safety zone (often called clear zone) is defined as the total roadside border area,
starting at the edge line of the carriageway, available for safe use by errant vehicles.
This area may consist of a shoulder, a recoverable slope, a non-recoverable slope,
and/or a clear run-out area. The desired width is dependent upon the traffic volumes,
speeds and on the roadside geometry.
Although the safety zone dimension includes the recovery zone, some functions of
the recovery and safety zones are different. Thus the recovery zone is treated
separately in Chapter 3.

2.3.1 The Safety Zone Concept
The safety zone concept is an area free of fixed objects or dangerous slopes,
adjacent to the roadway. It provides an area for drivers to control or stop their
vehicles if they have had an unplanned departure from the carriageway. The ground
should be relatively flat and gently graded. Impacts with hazards or rollovers due to
terrain conditions should be eliminated. The desired minimum width is dependent
upon traffic volumes and speeds and on the roadside geometry. RISER Deliverable 5
[1] provides the current dimensions recommended in reviewed European countries.
USA research [2] indicates that on high-speed roads (>70 km/h) a safety zone of 9 m
from the edge of the carriageway permits about 80 % of vehicles leaving the roadway
out of control to recover. Australian research [3] suggests that a clear zone of at least
2 m and preferably 3 m back from the kerb edge will significantly reduce the
consequences of vehicles leaving the carriageway. In the USA, the Insurance
Institute for Highway Safety (IIHS) [4] suggests that if the 10 foot (3 m) clear zones
were increased to 35 feet (10.5 m), collisions with street furniture would be reduced
by a further 10 %. A similar study in the UK [5] was performed where the relationship
between the width of the safety zone and the reduction of related accident types was
made (see Figure 2.1).
20
Figure 2.1. Relationship between the size of the clear zone and the accident reduction [5]
In a study on the Routes Nationales in France [6], one third (32 per cent) of fatal
collisions with street furniture take place within two metres of the edge of the
carriageway, and more than two thirds (70 per cent) occur within four metres (see
Figure 2.2).
Figure 2.2. Distribution of fatal accidents against fixed obstacles according to the distance
to the edge of the carriageway [6]
2.4 RISER Analysis

The risk of injury during a run-off-road event depends on the motions of the vehicle
and the type of objects contacted in the roadside area. Chapter 1 provided
information on the types and characteristics of obstacles and hazards found on
European roadsides. The following analyses will discuss the manner in which
vehicles leave the road and the resulting requirements for a safety zone.
There are two main topics of interest for run-off-road events; the conditions under
which a vehicle leaves the road (speed, angle, etc.) and the distance they travel in
the roadside. These issues are discussed and related to the design criteria necessary
for defining the safety zone. Both a theoretical basis for designing the safety zone
and accident based information, gained in the RISER project, will be presented.
2.4.1 Criteria to Dimension the Safety Zone

The dimensions of the safety zone must be based on the local terrain, weather
conditions and traffic environments, etc. A review of several European countries in
RISER provided seven main criteria used to specify the dimensions for the safety
zone. The main parameters of interest are:
21
1. Road type – the class of road (motorway, national road, divided or non
divided traffic lanes, etc.).
2. Traffic – the volume and mix of traffic observed on the road usually
expressed in Annual Average Daily Traffic (AADT) and percentage of heavy
vehicles.
3. Speed – the design speed is usually the most common speed used for
designing the road, but redesign of existing roads should use observed
speeds unless they are less than the design speed.
4. Side slope – the characteristics of the slopes adjacent to the roadway,
typically the gradient and height of the slope.
5. Horizontal alignment – separate criteria may be considered for straight and
curved sections.
6. Driving lane width – lateral width of the travel lane(s). Note that this is often
associated with the road type.
7. Other – many modifications of the safety zone width may result from the
location of bodies of water, industrial areas, residential areas and railway
lines, etc.
Table 2.1 provides the application of these criteria for the countries reviewed. For
specific details of the national standards, the reader is referred to RISER Deliverable
5 [1]. Most countries specify a safety zone based on the road characteristics
(identified above) but Spain uses this information implicitly by evaluating the risk of
contacting an obstacle and determining the severity of injuries.
Table 2.1. National Criteria for Dimensioning the Safety Zone
Criteria FI FR DE UK NL ES SE
Road type No Yes Yes Yes Yes Yes No
Traffic Yes Yes Yes No Yes Yes Yes
Speed Yes Yes Yes Yes Yes Yes Yes
Side slope Yes Yes Yes No Yes Yes Yes
Horizontal alignment Yes No Yes No Yes No No
Driving lane width No No No No No No Yes
Others Yes No Yes No Yes No Yes
2.4.2 Exit Conditions for Run-Off-Road Accidents

The speed and angle under which a vehicle leaves the road is dependent on many
factors. If we focus on the physical (non-driver) parameters then the exit angle is a
useful basis for discussion. The angle that a vehicle leaves the road depends on
road-tyre friction, travel speed, lateral position of the vehicle to the carriageway, and
geometrical road properties (vertical and horizontal alignment). As seen in Table 2.1,
many parameters related to the road type become design criteria of interest.
Theoretical Analysis for exit angles

If the behaviour of a vehicle is determined by its steering system, the trajectory of its
centre of mass running off the road on a straight road section can be calculated as a
function of the distance between its initial straight trajectory and the carriageway
edge line (see Figure 2.3). The exit angle depends on the travelled speed and the
maximum lateral acceleration for a vehicle, which can be related to the friction
coefficient between the road and the vehicle’s tyres.
22
The theoretical maximum exit angle can be expressed from this simplified analysis of
the vehicle handling ability. If the maximum lateral friction coefficient between the
tyres and the road surface is given as f, and the distance to the carriageway edge
line is d, the principal of centripetal acceleration will yield the following relationship
between exit angle (NY) and the travelled speed v (given in m/s):
NY = cos −1 1 − fgd / v 2 [ ] Equation 2-1
g is acceleration due to gravity = 9.81 m/s2
Figure 2.3. Exit angle on a straight road
The results of Equation 2.1 can be plotted for different lateral distances and travel
speeds and is shown in Figure 2.4.
40
Travel Speed 70 90
(km/h)
35 110 130
30
Vehicle Angle [deg]
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10
Lateral Distance [m]
Figure 2.4. Relationship between the trajectory of a vehicle running off the carriageway and
the travelled speed. Friction coefficient 0.7
23
On a straight 2 lane rural roads, the maximum trajectory of a vehicle should not be
more than 20 degrees if it leaves the offside of the roadway. This assumes a travel
speed greater than 90 km/h and a friction coefficient of 0.7.
In a similar way, the trajectory of a vehicle running off the road on a curved road
section can be calculated as a function of the distance between its initial straight
trajectory and the carriageway edge line, and the radius of the curve (Figure 2.5).
The exit angle does not depend on the travelled speed in this case. It has to be noted
that its value can be overestimated in this calculation due to the fact that the curve
radius is not constant, but decreasing as the vehicle travels through the curve.
Figure 2.5. Exit Angle for a Curve (no steering)
Theoretical values for the exit angles for curves (as depicted in Figure 2.5) are shown
in Figure 2.6.
16
Curve Radii 350 700
(m)
14 1000 1500
12
Vehicle Angle [deg]
10
0
0 1 2 3 4 5 6 7 8 9 10
Lateral Distance [m]
Figure 2.6. Relationship Between Curve Radius and Exit Angle
24
Again, the exit angle is not generally expected to be over 20 degrees for the usual
ranges of curve radii in current roads.
It is important to recognize the inverse relationship between exit speed and exit
angle. With increasing speed, the magnitude of the exit angle decreases. The
relationship between the exit speed and the exit angle calculated using Equation 2.1
is illustrated in Figure 2.7. This is a conservative relationship since the actual turning
manoeuvre creates extra drag forces on the tyres and slows the vehicle. Thus, exit
speeds are lower than presented in Figure 2.7
20
Vehicle Angle [deg]
15
10
5
70 80 90 100 110
Vehicle Speed [km/h]
Figure 2.7. Relationship between Exit Speed and Exit Angle (theoretical f=0.7, d=2m)
An important assumption made in the previous analyses is that the vehicle turns due
to a large steering input (by the driver) and that the vehicle turns in a stable, non
sliding manner. In reality many run-off-road events are due to abrupt steering
manoeuvres that result in vehicle rotations (yaw) so that the vehicle leaves the road
in a non-tracking manner. This behaviour is presented in Figure 2.8 where the
trajectory of the vehicle centre of mass moves at an angle NY to the carriageway
edge line while the vehicle heading (orientation) is defined by an angle PSI. These
angles are identical if the vehicle is not experiencing any sliding or rotational
behaviour (normal tracking).
Figure 2.8. Definition of Vehicle Exit Angles
25
Accident Data from RISER Detailed Database – Exit angles

From the accident data collected in RISER, 82 accidents were reconstructed and
tabulated for analysis concerning the first departure from the road. The following
information summarizes all road types in RISER and is not broken down by specific
road type, posted speed, etc. since the number of cases is too small to conduct more
specific analyses.
If the vehicle's first departure from the road is considered, the most critical angle to
evaluate is the vehicle trajectory angle, NY (Figure 2.8). This implies the momentum
vector for the vehicle and is a strong indicator how far into the roadside the vehicle
will travel. The cumulative distribution of exit angles from the accident data is
presented in Figure 2.9. It is important to identify the most common exit angles (NY)
and Figure 2.9 shows that 90 % of the exit angles are below 20 degrees which
identifies a correlation to test standards in Europe (Chapter 5).
The exit angles plotted in Figure 2.9 have not been divided into initial departures to
the left or right side of the carriageway. The possible exit angle is increased by a
longer lateral distance available for the vehicle to travel. Cases where departures to
the offside of the carriageway or departures to the nearside following a lane change
tended to have higher exit angles than those where the vehicle left the nearside of
the road.
100%
90%
80%
70%
Cumulative Percent
60%
50%
40%
30%
20%
10%
0%
0 5 10 15 20 25 30 35
Initial Exit Angle [Degrees)
Figure 2.9. Cumulative distribution of initial departure angle from road
Figure 2.10 identifies the amount of spinning or sliding of the vehicle as it left the
road. The values presented on the X-axis represent the difference between the
vehicle's trajectory (NY) and the vehicle's heading (PSI) angles. A normally operating
vehicle has very little difference between vehicle trajectory and heading. The results
plotted in Figure 2.10 show that only about half of the vehicles (47%) had less than a
5 degree difference between these angles and thus were tracking in a reasonably
stable manner. The remainder of the exits (53%) had significant yaw motions (NY-
PSI values over 5 degrees) and are important to consider if there will be subsequent
driver steering inputs which lead to a higher rollover risk (Appendix C).
26
50%
45%
40%
35%
30%
Percentage 25%
20%
15%
10%
5%
0%
0-5 5-1
0
10-
15
15-
20
20-
25
25-
30
30-
35
35-
40
40-
45
Difference between vehicle trajectory and heading (NY-PSI) [Degrees]
Figure 2.10. Magnitude of lateral sliding
The last issue to consider is the relationship between exit angle and exit speed.
Figure 2.7 showed the theoretically relationship between exit angle and exit speed for
a given lateral position of the vehicle before a run-off-road event. The reconstructed
cases from the RISER investigations are presented in Figure 2.11. Even though the
theoretical case represents a low (2 m) lateral motion of the vehicle leaving the road
and the RISER data covers a wide range (up to 10 m) lateral motion, the expected
exit angles are about half of the theoretical angles. This suggests that any theoretical
estimates of exit angle based on maximum cornering capacity are very conservative.
25
Theoretical
(Equation 2.1)
20
Exit Angle (degrees)
15
Curve Fit
to data
10
0
0 20 40 60 80 100 120 140
Exit Speed (km/h)
Figure 2.11. Relationship between Exit Angle And Exit Speed (Reconstructed)
27
Unfortunately, there is a limited amount of information in the detailed database about

the levels of increased risk on roads where there is a bend. No measures are
available of the severity of the bend. However, from looking at the results of the
detailed database analysis, it appears that single vehicle accidents are prevalent at
locations where there was a curve in the road. In the RISER database, 105 out of
211 cases occurred at locations where there was a curved road section. In 26 fatal or
serious injury accidents, the curve was recorded as being a potential risk factor. In
nearly all of these cases (23), the vehicle did not return to the road after its initial run-
off.
Exit Speeds
As already presented for exit angles, the exit speed of the vehicle from the road will
depend on the vehicle speed and position on the travel way prior to its approach to
the carriageway edge line. These factors are linked to the road type as this will affect
the lane width and travel speeds on the roadway. Road surface characteristics are
implicitly connected to road type since renovation and maintenance of road surfaces
are more frequent on higher standards of roads.
It is desirable that roadway designs that result in a more harmonious relationship

between the desired operating speed, the actual operating speed, and the posted
speed limit. The goal is to provide geometric street designs that “look and feel” like
the intended purpose of the roadway (Self Explaining Roads). Such an approach
produces geometric conditions that should result in operating speeds that are
consistent with driver expectations and commensurate with the function of the
roadway. It is envisioned that a complementary relationship would then exist between
design speed, operating speed, and posted speed limits [7].
The most common factors in single vehicle accidents are approaching a curve at too
high speed, over-correcting and then leaving the roadway. The severity of injuries
generally depends on the velocity and the impact configuration. Speed is intimately
related to the risk and severity of a crash. A review of international research on the
relationship between speed, speed limits and accidents came to the conclusion that a
1 km/h change in the mean speed of traffic produces a 3 % change in injury
accidents [8]. Other studies show the contribution of speed variance, vehicles moving
much slower or much faster than the median speed, are over-involved in accidents
[9, 10, 11, 12].
Definition of Speed
Design speed
selected speed used to determine the various geometric design features of
the roadway
Observed speed
operating, running, average, pace, or 85th percentile speeds is the speed at
which drivers are observed operating their vehicles during free flow
conditions. The 85th percentile of the distribution of observed speeds is the
most frequently used measure of the operating speed associated with a
particular location or geometric feature.
28
Posted speed
the maximum or minimum speed for a road as determined by law (usually
shown on speed limit sign)
Initial speed
the speed at which the deceleration in an accident begins. It is the observed
speed without reaction time and brake lag included.
For the determination of the safety zone for NEW roads the design speed and
for EXISTING roads the observed speed should be used.
Accident Data from RISER Detailed Database – Exit Speed

Data from the reconstructed RISER cases were used to develop the distribution of
impact speeds presented in Figure 2.12. Again, the data represents single vehicle
accidents from all road types and posted speed limits for the RISER cases. The data
shows that 90 % of the crashes are below 120 km/h and 80 % are below 110 km/h.
For reference, 110 km/h is the test speed for heavy passenger cars against safety
barriers (Chapter 5).
100%
90%
80%
70%
Cumulative Percent
60%
50%
40%
30%
20%
10%
0%
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
Exit Speed [km/h]
Figure 2.12. Distribution of exit speed for initial run-off-road event
2.4.3 Strategy for Determining a Safety Zone Width

There is a necessity to identify the safety functions when developing the safety zone
designs. The explicit idea is that the vehicle will not contact any hazard which results
in a collision with serious consequences. This is achieved by preventing rollover and
removing fixed objects. The implicit idea is if the vehicle leaves the road and crosses
the entire safety zone, it slows down and reduces the consequences of any resulting
impact beyond the safety zone. Ideally, the vehicle should not ever leave the safety
zone, instead the driver should be able to control and stop the vehicle within the
lateral and longitudinal limits of the safety zone.
From these objectives, it is important to identify the parameters affecting the vehicle
motions in the safety zone so that they can then be used as criteria for dimensioning
the safety zone.
29
Vehicle Protection Limits

Human survival in a crash depends on how well the crash energies are absorbed by
the vehicle. Uncushioned, the human body cannot survive impacts above 40 km/h. If
a crash occurs, in modern vehicles some of the energy will be absorbed by the
vehicle structure and the restraint systems. But the vehicles are optimized for
standard crash tests, and single vehicle accident situations are not well represented
in vehicle design requirements. It is important to note that 40 km/h was the threshold
for fatal injuries observed for many of the hazards documented in the detailed
database (Chapter 1).
It is the energy-absorbing capabilities of the car and the roadside infrastructure taken
together that provide optimum protection from the energies of impact. The car is
designed and crash tested in frontal and side impact configurations (Appendix D).
Modern vehicles are equipped with advanced safety features, but older vehicles
(especially those designed prior to 1990) and some special purpose vehicles do not
possess these advanced devices. Therefore it is important to consider potential
impact speeds which can be considered relevant for setting a desired safety level.
On many continents, vehicles are tested at 50 km/h into a flat, rigid barrier. These
tests are quite severe and may be a useful starting point for designing roadside
safety zones. However, the accident history relevant for the road section should be
reviewed. From the RISER detailed accident database, the most severe accident
configuration was with trees and poles. From this data (reported in Chapter 1) fatal
and serious injuries were reported for impact speeds over 40 km/h. With increasing
financial resources, this impact speed should be reduced further.
Theoretical dimensioning principles for the safety zone

If the goal of the safety zone is to eliminate impacts with objects for impact speeds
above 40 km/h, then the lateral distance of the safety zone must accommodate the
motions of the vehicles until this speed is reached. The approximate braking distance
of a vehicle from the design (or observed) speed of the roadway to 40 km/h
(11.1 m/s) may be determined from the following equation:
v 2 − 11.12 v 2 − 11.12
s = = Equation 2-2
2⋅a 2⋅µ ⋅ g
s = braking distance [m]

v = Design speed or observed speed [m/s]
a = Deceleration [m/s²]
= Coefficient of friction
g = Gravity [9.81 m/s²]
The deceleration possible by the vehicle is dependent on the ground conditions. Most
side slopes and ditches have surfaces of gravel and topsoil, possibly with grass type
vegetation. The rolling resistance on these surfaces are higher than asphalt and tend
to slow vehicles at a rate up to approximately 2 m/s2 (0.2 g) see Table 2.2.
30
Table 2.2. Rolling resistance coefficient

Tyre on Rolling resistance coefficient
Asphalt 0.010 – 0.015
Concrete 0.011 – 0.014
Cobbled pavement 0.015 – 0.020
Cobbled pavement, very bad – 0.033
Gravel, rolled 0.020
Soil way 0.045 – 0.16
Sand 0.15 – 0.30
Field 0.14 – 0.24
Studies documented in the literature show that most drivers decelerate at a rate
greater than 4.5 m/s² during braking. Approximately 90 % of all drivers decelerate at
rates greater than 3.4 m/s². Such decelerations can be handled by most drivers. The
friction levels of the different roadside sections are often not consistent. For a vehicle
running out of the road the worst case is a friction coefficient of 0.3 for wet grass (with
exception of ice). This results in an available deceleration rate of 2.9 m/s².
Roadside geometrics have a great influence on the frequency of serious injury and
fatal crashes; especially the design of the side slopes has influence on the
occurrence of rollover, which is one of the most dangerous events in single vehicle
accidents. In the US the Roadside Design Guide [13] defines recoverable,
traversable and non-traversable slopes. A recoverable slope is a slope on which a
motorist may, to a greater or lesser extent, retain or regain control of a vehicle by
slowing or stopping. Slopes flatter than 1:4 are generally considered recoverable,
where motorists generally can stop their vehicles or slow them down enough to safely
return to the roadway. A non-recoverable slope is a slope which is considered
traversable but on which an errant vehicle will continue to the bottom. Embankment
slopes between 1:3 and 1:4 may be considered traversable but non-recoverable if
they are smooth and free of fixed objects. A clear run-out area is the area at the toe
of a non-recoverable slope available for safe use by an errant vehicle.
For fill slopes the approximate stopping distance of a vehicle at the design or
observed speed of the roadway may be determined from the following equation:
v 2 − 11.12 v 2 − 11.12
s= = Equation 2-3
2⋅a 2 ⋅ g ⋅ (µ ⋅ cos (ϕ ) − sin (ϕ ))
s = braking distance [m]

v = Design speed or observed speed [m/s]
a = Deceleration [m/s²]
= Coefficient of friction
g = Gravity [9.81 m/s²]
= Slope angle [°]
In the formula above the maximum slope angle for a given coefficient of friction is
also contained. If the gradient of the slope is equal to the coefficient of friction, the
limit for a safe deceleration is reached.
tan (ϕ ) = µ Equation 2-4
31
This means for slopes inclinations over 1:3 and coefficient of friction 0.3 there will be
no safe stop.
The absolute inclination of the slope is not relevant for running off the road situations,
but the resulting inclination when driving under a certain course angle as seen in
Figure 2.13. The vehicle' s heading angle changes while on the slope due to driver
input.
Figure 2.13 Vehicle Course on a Slope
The driven inclination can be determined by the side inclination and the speed
direction using the following correlation:
sin (η ) = sin (ϕ ) ⋅ sin (α ) Equation 2-5
= Driven slope angle [°]

= Slope angle [°]
= Vehicle velocity direction [°]
As indicated in Table 2.1, roadside slopes and embankments are a variable in the
safety zone criteria. Both the height and gradient of the slope is important. Testing in
Finland and Sweden [14] indicates that a vehicle can easily traverse a V shaped
ditch only 1 m deep. Similarly for the exit angle and speed, the road type has some
effect on the roadside slopes. Higher standard roads like motorways have wider road
beds and result in shallower side slopes when compared to smaller rural roads.
Accident investigations have shown that the transitions of slopes and the change in
the subsoil are problematic zones. Often the wheels jam into the soil and cause
rollover. Therefore the slopes should be homogenous, transitions should be rounded
and flattened when possible.
Safety Zone Width Suggestions

From the review of European practice, the width of the safety zone has a strong
relationship to speed. As the speed for the road increases, so does the width of the
safety zone.
Using the information presented previously, the width of the safety zone can be
defined as the width necessary to stop a vehicle to avoid serious impact. As an
32
example, the following table lists the recommended safety zone widths if the road,
speed, and slope conditions are:
Coefficient of friction 0.3 (grass)

Initial manoeuvre on the road was abrupt steering
Vehicles decelerate on roadside without manoeuvre
Impact velocity 40 km/h after crossing the safety zone
Flat Ground (ideal conditions)
Table 2.3. Theoretical Safety Zone Widths
Exit
Slope a Exit Speed from Carriageway (km/h)
ang.
(deg) (m/s²) 50 60 70 80 90 100 110 120 130
5 0 0.3 2.9 1 2 4 5 7 10 12 15 17
10 0 0.3 2.9 2 5 8 11 15 19 24 29 35
15 0 0.3 2.9 3 7 11 16 22 29 36 43 52
20 0 0.3 2.9 4 9 15 22 29 38 47 57 69
25 0 0.3 2.9 5 11 18 27 36 47 58 71 85
30 0 0.3 2.9 6 13 22 31 43 55 69 84 100
A comparison of the values in Table 2.3 with actual safety zone dimension in current
guidelines [1] indicates that an exit angle of 5 degrees produces 7 m and 12 m safety
zones for exit speeds of 90 and 110 km/h. These values are consistent with current
practice in many European countries for roads with these posted speed limits. The 5
degree and 90 km/h exit conditions are also the median exit angle and exit speed
observed in the RISER reconstructions (Figure 2.9 and Figure 2.12). Thus the
theoretical safety zone calculation for 5 degrees is a useful starting point for
developing safety zone criteria. The theory produces larger safety zone dimensions
for exit angles above 5 degrees which are larger than in common practice. This is a
practical problem for the road owner/operator and local conditions must be
considered. This approach can be useful for applying local modifications to the safety
zone.
Set-back Measurements in RISER Detailed Database

In the RISER detailed database a general distribution of the minimum and maximum
set-back distances for the different roadside obstacles struck can be seen in Figure
2.14. This figure illustrates that most impacts occur in the first 10 m of the roadside,
measured from the carriageway edge line.
33
Safety barrier - steel (29)

Minimum set-back
Safety barrier - concrete (4)
Saf ety barrier - termination (3)
Maximum set-back
Safety barrier - w ire rope (1)
Bridge parapet (1)
Safety barrier - other (1)

Sign post (8)
Lighting pole (10)
Telegraph pole (8)
Tree (29)
Non-safety fence (6)
Hedge (1)
Wall (9)
Bridge pier (5)

Rock/boulder (3)
Fog pole (1)
Embankment/slope (12)
Ditch (9)
Drainage gully (4)
Foot path (1)
Other (6)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Set-back (m)
Figure 2.14. Set-back Information for Obstacles recorded in RISER Detailed Database
The cumulative distribution of the set-back distance is provided in Figure 2.15. The
85th percentile set-back distance is identified by the dashed lines and is seen to be
7 m. This is consistent with information from other countries like France and the
Netherlands.
100%
80%
Frequency of Impacts [%]
60%
40%
20%
0%
0 2 4 6 8 10 12 14
Setback Distance [m]
Figure 2.15 Cumulative Distribution of Struck Obstacles
34
The information presented in Figure 2.14 and Figure 2.15 are indicative of accident
data on roads generally following the safety zone concept. This means that we would
expect that most impacts with roadside hazards would be beyond the 4.5 m limit
which is the smallest safety zone in the reviewed countries. The fact that 50 % of all
impacts are within the 4.5 m range can partly be explained by the number of
safety/road sign objects listed in Figure 2.14. The data collected in RISER have only
addressed impacts, the maximum distance that a vehicle travelled into the safety
zone during an uncontrolled exit from the road was not measured in the RISER
project.
2.5 Summary
The dimensioning of a safety zone is a difficult process. A theoretical process using
vehicle dynamics and human tolerance information provides results consistent with
current practice if vehicle exit angles from the road are 5 degrees, which is the
median value for the data collected in RISER. An alternative is to use the struck
object set-back distance obtained from the accident data. In this latter approach, the
data coming from RISER appear to support information from France, the US, and the
Netherlands which shows that the risk of contact with an obstacle drops dramatically
after the first few meters and most impacts with roadside obstacles occur in the first
10 m.
Most safety zones in Europe are specified to be between 6-10 m for travel speeds
around 100 km/h. Safety zones are smaller for lower speeds and for 80 km/h roads,
the same countries use 4.5-7 m as a safety zone width.
The RISER analysis provides two alternatives for designing the roadside safety zone.
1. Based on the risk of injury during an impact with a hazard, the safety zone
can be dimensioned for allowable impacts with hazards. In this case the
allowable impact speed for striking a hazard is given in Chapter 1 and the
impact speeds are calculated from the information provided earlier in this
chapter.
2. The safety zone can be dimensioned as the risk for a fatal impact with an
object of a given set-back. Based on the RISER database, the set-back
distances can be grouped into the categories based on the road
characteristics.
2.5.1 Safety Zone Requirements

The requirements for a well designed safety zone are that:
the consequences of a run-off are reduced

the width should be designed that most vehicles that leave the road do not
leave the safety zone
there should only be slopes that do not cause rollovers
the surface should be homogenous and even to prevent rollover
there should be no unprotected fixed objects located within the safety zone
Legislators and authorities should ensure that a safety zone only contains artificial
structures that will collapse or break away on impact without significantly damaging
35
an errant vehicle. Where allocation of the desired safety zone is not practicable, they
should consider erecting an appropriate road restraint system.
Accident data collected in RISER indicates that most vehicle departures from the
road were less than 20 degrees and 110 km/h. These run-off roads events involve a
non-tracking (yawing) vehicle in about half of the cases. Impacts with roadside
obstacles were observed up to 10 m from the road and 85 % of all roadside impacts
happened within the first 7 m of the roadside. A roadside safety zone should be
dimensioned to the local road conditions using the local accident data when possible.
References

GRD2/2001/50088, 2003
2 AASHTO report “Highway Design and Operation Practices Related to Highway Safety”
3 Fox J.C., et al, Collisions with Utility Poles, Department of Transport, Australia, 1979
4 Roadside Hazards, Insurance Institute for Highway Safety, Fatality Facts, 1992
5 Ogden K.W., Safer Roads: A Guide to Road Safety Engineering, ISBN 0 291 39829 4,
Avebury Technical, England, 1996
6 Les Accidents Mortels contre Obstacles, Centre D’etudes Techniques de L’equipement
Normandie Centre, France, April 1995
7 Fitzpatrick K., Carlson P., Brewer M.A., Wooldridge M.D., Miaou S.P., NCHRP REPORT
504, Design Speed, Operating Speed, and Posted Speed Practices, 2003
Transportation Research Board
8 Finch, D.J., Kompfner, P., Lockwood, C.R. & Maycock, G.; Speed, speed limits and
accidents; Project Record S211G/RB /Project Report PR 58; Transport Research
Laboratory TRL, Crowthorne, Berkshire; 1994
9 Munden, J.M.; Relation between driver' s speed and accident rate; Ministry of Transport,
RRL Laboratory Report 88; Road Research Laboratory, Crowthorne, UK; 1967
10 Hauer, E.; Accidents, overtaking and speed control; Accident Analysis and Prevention,
3(1), 1–13; 1971
11 Maycock, G., Brocklebank, P.J., Hall, R.D.; Road layout design standards and driver
behaviour; TRL Report 332; Transport Research Laboratory, Crowthorne, UK; 1998
12 Quimby, A., Maycock, G., Palmer, C., Grayson, G.B.; Drivers’ speed choice: an in-depth
study; TRL Report 326; Transport Research Laboratory, Crowthorne, UK; 1999
13 Roadside Design Guide, Task Force for Roadside Safety of the Standing Committee on
Highways Subcommittee on Design, American Association of State Highway and
Transportation Officials, Washington, D.C. 1996.
14 Thomson R., & Valtonen, J., "Vehicle Impacts in V Ditches", Transportation Research
Record 1797, Pages: pp 82-88, 2002
36
CHAPTER 3: THE RECOVERY ZONE
3.1 Introduction
For some known or unknown reasons road users sometimes leave the running
carriageway. The human factor (drowsiness, drinking and drugs driving, fatigue,
distraction, medication, etc.) is admittedly important in road departure but the
influence of the roadside area should not be underestimated.
There are many opportunities for mitigating roadside accidents, in particular once the
run-off does occur, by providing opportunities for the driver of the vehicle to recover
and return to the road without incident.
This chapter provides road designers with guidance on the design of the roadside
environment, which includes elements to allow for recovery. Road equipment which
is intended to reduce the severity of the roadside collision is addressed in Chapter 4
and Chapter 5.
3.2 Definition of the Recovery Zone

The recovery zone is defined in two different ways according to European policies: in
most countries it is a narrow roadside zone beside the carriageway considered as an
integral part of the total clear, hazard-free safety zone design described in Chapter 2.
In a few countries the recovery zone is a hard shoulder that allows limited vehicle
manoeuvres and is considered as a separate issue.
A hard shoulder is defined as being an asphalt or concrete surface immediately

beyond the carriageway edge line. Shoulder pavement surface and condition as well
as friction properties are intended to be as good as the road surface. An example of
a recovery zone/hard shoulder is illustrated in Figure 3.1.
Figure 3.1. Example of a Recovery Zone/Hard Shoulder in Spain
3.2.1 The Main Functions of a Recovery Zone

On motorways, the hard shoulder, referred to as the emergency lane, provides extra
space for emergency vehicles and unexpected stopping.
37
On dual carriageway or single carriageway roads, paved shoulder is generally used

to accommodate driving errors when the vehicle goes off the road, but it also
provides extra space to:
help car drivers to avoid potential conflicts with errant vehicles veering away
from their normal paths
increase lateral clearance during overtaking manoeuvres
avoid vehicles making a turn towards minor roads and private accesses
accommodate stopped vehicles for emergency use under unexpected
circumstances (mechanical failures, flat tyres, etc.)
be used by cyclists and pedestrians off the carriageway
improve highway capacity and facilitate road maintenance and rural
activities
maintain an access to rescue vehicles in case of emergencies
increase sight distance and clear visibility from minor roads

This subsection is divided into seven parts, each of which treats a different aspect of
the roadside safety research:
1. The first section is dedicated to the positive effect of hard (paved) shoulder
on accident rates
2. The second section discuss the negative effect of soft shoulder on run-offs
3. The third section is dedicated to the positive effect of paved shoulders on
accidents involving vulnerable road users
4. The fourth section points out the positive effect of paved shoulders on other
types of accidents
5. The fifth section gives some information about the best locations to construct
pavement shoulders
6. The sixth section is dedicated to the width of a recovery zone
7. The last section discusses the surface characteristics of a recovery zone
3.3.1 Hard Shoulder Positive Effects on Accident Rates

Several reports have shown that the presence of hard shoulders is recommended
because of their positive safety effects. Some of them refer to the positive influence
of the combination of reducing each traffic lane with widening of the hard shoulder
widths.
In 1998, the SAFESTAR European project [1] highlighted the findings of a great
number of roadside research on rural single carriageway roads from different
countries: In Germany Brannolte [2] detected that two-lane roads with shoulders had
an approximately 10 % lower accident rate compared to similar roads without
shoulders. In the USA Foody and Long [3] found that the mean accident rate for
stabilized shoulder sections was significantly lower that for sections of unstabilised
shoulders. The same research project also quoted several authors [2, 4, 5] and found
decreasing accident rates with increasing paved shoulder widths.
38
Brüde and Larsson [6] studied the safety effects of an increase in lane width (from
3.75 m to 5.5 m) and a simultaneous reduction of shoulder width (from 2.75 m to
1 m): the findings indicated that accident severity increased strongly in connection
with the widening of the road. This confirmed the findings of German research
undertaken in 1993, which concluded that accident rates and accident cost rates
were higher on over wide roads.
In the Handbook of Road Safety Measures [7] researchers have studied the effects of
constructing hard shoulders on the number of accidents. According to the report
roads with hard shoulders which were usually 0.3 to 1 m wide, had an accident rate
around 5 to 10 % lower than roads without hard shoulders (this applied to rural
roads). They also quoted a number of studies evaluating the effect of narrower traffic
lanes combined with wider hard shoulders.
A special report issued in the American Transportation Research Board headed

Designing Safer Roads, Practice for Resurfacing, Restoration and Rehabilitation [8],
states that prior research on lane and shoulder width and shoulder type indicates that
accident rates decrease with increases in lane and shoulder width.
In France, within the framework of the Users’ Safety on Existing Roads process, a
team of researchers carried out a before/after study in order to evaluate the effect of
constructing paved shoulders on 7 national trunk roads all across France. As a result
the conclusion of the report to be published [9], states that the pavement surfacing of
shoulders improved safety on the seven roads and statically representative figures
show that the implementation of paved shoulders has an effect on 43 % of the
number of injury accidents, and shows a decrease by 65 % of the accident severity,
in confirmation of comparable positive findings provided by a number of foreign
studies.
The study “Evaluation of pavement shoulders” [10], was intended to determine the
effect of pavement shoulders on the safety and the structural strength of highways.
The University of Wyoming performed the effectiveness of pavement shoulders on
the safety of highways in Wyoming for a five-year period. It was determined that
significant reductions in accidents could be expected when shoulders widths were
increased. If a 1.8 m shoulder was in place instead of a 0 m shoulder, there would be
a 47 percent reduction in accidents. The first 0.6 m (2 feet) of shoulders is most
effective in reducing accident numbers (a 19 percent reduction in accidents). These
results are consistent with similar studies performed at a national American level.
Recommendation:
A great number of studies show the positive effect on road accident rates of hard or
paved shoulder beside the travel lanes.
3.3.2 Soft Shoulder Negative Effects on Run-off-road Accidents

RISER developed a detailed database fed with 211 in-depth investigations of
accidents. An in-depth analysis of the data in the detailed database was undertaken
and a number of areas were investigated, including the type of infrastructure
impacted, set-backs of hazards and objects, recovery zone width when existing,
accidents involving safety barriers and an overview of accidents where there is an
evidence of an initial run-off with no impact. In the analysis of the recovery zone, the
39
European consortium particularly focussed on single and multiple run-offs. Single

run-off is where the vehicle directly travels on nearside or offside of road with no
return onto the road. Multiple run-off is an accident where the vehicle runs off the
road, then returns to the road, and then runs off the road at least one more time.
There are two descriptions of multiple run-offs: one is where the initial encroachment
is with no impact; the other is with impact.
In the database 37 multiple run-off cases were scrutinized. In 20 cases, the vehicle
had an initial encroachment without impact on roadside furniture, and in the further
17 cases, the vehicle had an initial encroachment with an impact on roadside
furniture.
The study reveals that on single carriageway roads, 11 of the 13 multiple run-offs
without impact occurred on unpaved shoulders (asphalt strips of 0.5 m or less), with 2
having unknown shoulder widths. The study also reveals that no multiple run-off
without impact on roadside furniture during initial encroachment occurred on paved
shoulders.
The research highlights several cases in which roadside environment did play a role
in the accident occurrence. In 18 accidents the vehicle had the initial encroachment
on one side of road before travelling across the road and having the main impact. In
13 of these 18 cases, the vehicle initially left the nearside and had impact in offside.
A step-by-step scrutiny of the cases clearly shows that multiple run-offs generally
occur on roads with unpaved shoulders where grass and gravel shoulders are
prevailing.
Some accurate information on this road design issue is also available from other
sources.
In 2004, CETE Normandie-Centre and CEESAR carried out a study on the influence
of roadside conditions and surface on run-off-road accidents in rural areas [11]. A
total of 56 single and multiple run-offs (corresponding to abovementioned definition)
have been studied. The report concludes that a recovery zone would have played a
key role in the accidents, because when travelling off the road the vehicle was still
controllable in 56 % of the cases. In the multiple run-off cases, 10 to 15 vehicles
travel less than 1.5 m-deep onto the roadside at the initial encroachment. The report
also states that when looking closely at the accident mechanisms, the presence of
rolling stones and gravels is originating a second run-off-road. The second part of the
conclusion is related to road restraint systems and energy absorbing road
equipments (see Chapter 5): regarding final exit speeds 38 values out of 39 are
inferior to 100 km/h, and regarding exit angles, in 74 % of the cases the final exit
angle is inferior to 20°.
In 1997, CETE Normandie-Centre and CEESAR analysed 81 detailed accidents [12].

The main figures show that there were 38 run-offs on straight and 43 on bends, in
which there are 41 multiple run-offs (50 %), this accident mechanism being more
frequent on left bends. The study points out that the initial encroachment angle is
shallow (<4° on average) and it occurs on less than 2 m of the roadside; the second
encroachment angle is much higher than the previous one during multiple run-off
accidents. In almost 50 % of single vehicle run-off accidents (SVA), the driver is still
capable to maintain control of his car when the car is going onto the roadside after
40
leaving the road. Road safety experts assert that the roadside design can be
efficiently improved on this specific area where any slight encroachment of a vehicle
onto a soft shoulder can turn an assumingly ‘controllable’ situation into a real loss of
control. The report concludes that paved shoulders may have had a positive effect in
50 % of the single vehicle accidents.
Another research entitled Exploration of In-Depth Investigations of Accidents [13],

was led in 2000 by INRETS on 84 detailed accidents on bends. In 45 % of the run-
offs, the vehicle can still be controlled while initially encroaching the roadside area. It
also points out that the roadside environment has played a key role in the accident
causation, insisting on that grass or loose shoulder material should be rejected when
designing this part of the road.
3.3.3 Hard Shoulder Positive Effect on Accidents against Pedestrians, Bikes

and Mopeds
A Danish study [14] quoted calculated that a paved shoulder increase from 0.2 to
0.5 m showed a significant reduction of accident risks for vehicle accidents by about
25 % and for pedestrian and cycling accidents by 40 %.
The abovementioned Handbook of Road Safety Measures [7] shows that both
bicycle accidents and accidents involving motor vehicles were reduced when the
width of the hard shoulder was increased from 0.2 m to 0.5 m.
In 1992, Sécurité des Routes et des Rues [15], Safety on Roads and Streets, by
SETRA, France, asserts that pedestrians and pedal cyclists (both 8 % of the fatal
accidents in France) may travel in safer conditions if roadside conditions were
improved. It is shown that in 50 % of the pedestrian accidents in rural areas, people
involved were walking along the driving lanes.
3.3.4 Hard Shoulder Positive Effect on Other Accident Types

In addition to pedestrian and cycling accidents Safety on Roads and Streets [12]
draws up a list of a number of accident types in which there is a relationship between
safety and roadside design: sideswipe and head-on accidents after a loss of control,
multi-vehicle collisions where the roadside environment could not provide extra space
for avoidance (meeting errant vehicles, left-turning vehicles, rear-end collisions in tail-
backs, overtaking accidents). All these accident scenarios have been described in
several international studies [16, 17, 18, 19].
In accidents at junctions, a French experimentation has shown a reduction in

accidents in private accesses and junctions after constructing paved shoulders
alongside particular roads.
Recommendation
Several studies show the positive effect of a paved shoulder on accidents involving
vulnerable road users, but also on other types of accident, such as single vehicle
accidents, head-on collisions and rear-end collisions.
3.3.5 Accident Locations and Road Types

Research carried out by INRETS [20] show that the abovementioned accidents occur
most frequently on:
41
national two-way rural roads

high-traffic roads
straight roads
the outside of curves with radius 200 m
roads with no hard shoulders.
These elements confirm the conclusions provided by RISER’s police database

analysis: most of RISER single vehicle accidents are reported on single carriageway
roads, with high fatality rate on this road type as well. RISER’s report also states
most of the single vehicle accidents occur on straight roads.
RISER’s detailed database analysis of accidents on bends showed that there are a
sizeable number of accidents where the initial run-off occurred in curves where
roadside was not paved (18 cases out of 104, 17 %)
Recommendation
Paved shoulders should be constructed in priority:
on main roads
on the outside of curves
opposite T-junctions
where vulnerable road users may travel
3.3.6 The Width of a Recovery Zone

What is important to know is whether hard shoulder width is related to safety and
whether increasing shoulder width improve or degrade safety. Answers to this crucial
issue are available from different international sources. Most of them highlight the
positive effect of widening the recovery zone - to some extent, though - while fewer
others show that widening shoulders is less effective than widening lanes.
Several authors found decreasing accident rates with increasing paved shoulder
widths, quoting “an optimum value for the shoulder widths was not determined
unambiguously, but a total pavement width of 10 m (3.5-3.7 m wide lanes and paved
or stabilized shoulders of 1.3-1.5 m) is mentioned as the pavement width beyond
which further widening does not improve safety.
American studies [4] have shown the relationship between accident rate and paved
shoulder width. A compilation of additional studies (Zeeger, Deen and Mayes (1981),
Barbaresso and Bair (1983) and Rosbach (1984) shows that increasing the width of
the hard shoulder by around 0.3 m appears to reduce the number of injury accidents
by about 20 %, and both bicycle accidents and accidents involving motor vehicles are
reduced when the width of hard shoulder was increased from 0.2 to 0.5 m.
In the USA, the Traffic Safety Toolbox [21], by the Institute of Transportation
Engineers and Designing Safer Roads shows that accident rates decrease with
increases in lane and shoulder width, and widening shoulders is less effective than
widening lanes.
42
In-depth investigations of accidents have shown the initial encroachments in single

vehicle multiple run-off accidents were about on average 0.7 m-deep on bends [20]
and 1.2 m-deep on straight, with an exit angle of 5° at a speed of 80-90 km/h [22].
From 1992 to 2002, CETE Normandie-Centre has undertaken in-situ studies to

evaluate the effects on driver behaviours of constructing 0.80 m to 1.2 m-wide paved
shoulders along five different two-lane road sections. As a result there were no
variation in traffic speeds and the lateral position of the vehicles was closer to
nearside area (Influence of Paved Shoulders on Driver Behaviour, several cases in
Seine-Maritime, France [23].
In their Handbook of Road Safety Measures [7], Elvik and Vaa conclude that
increasing the width of the hard shoulder by around 0.3 m appears to reduce the
number of injury accidents by about 20 % while increasing the width of the hard
shoulder by around 1 metre does not lead to statistically significant changes in the
total number of accidents (injury and property damage only accidents. It has to be
noticed however that the initial width is unknown and the result referring to injury
accidents comes from just one study.
3.3.7 The Roadside Surface Characteristics
Skid Resistance
In 1998, a report titled Roadside Environment and Light Vehicles Kinetics by UTAC,
France [24] points out that once the vehicle has encroached onto the pavement edge
drop, the driver gives a violent turn of the wheel in such a disproportionate amplitude
that his vehicle goes back onto the road (or onto the roadside) under a far greater
angle. In-depth investigations of accidents also reveal that during the initial
encroachment, the steering manoeuvre is violent and shows the need for skid
resistance on the roadside too.
In 2004, the CETE de Lyon (France) carried out a study of skid resistance on grass
or gravel shoulders [25]; grass and gravel shoulders provide a skid resistance
respectively 30 % and 40 % inferior to asphalted carriageways. The report concludes
that one of the run-off accident causes could lay in insufficient skid resistance of
loose material shoulder where initial encroachment angle are low.
Recommendation
Gravel and grass shoulders do not offer a good level of skid resistance. In case of a
slight encroachment onto the roadside, it is then almost impossible to recover in a
safe way
Shoulder Rumble Strips

According to a number of reports based upon in-depth investigations of accidents
(RISER Deliverable 4 [20, 26], 27]) human factors (mainly alcohol, fatigue and
distraction) was prevailing in accidents where the vehicle was leaving the road at a
low run-off angle but was still controllable.
In 1991, research by CETE Normandie-Centre on French motorways has shown that

the number and the lateral extent of encroachments onto emergency lane was
43
reduced after implementing jiggle-barred shoulder strips [28]. The report concludes
that this type of edge marking would increase the efficiency of a paved shoulder.
In a literature review on recovery zone performed by CETE Normandie-Centre and

the Eastern Paris Regional Laboratory LREP reporting to the French ministry for
Transport, a document [29] quotes two American safety evaluations after installing
shoulder rumble strips: the influence of rumble strips on monotonous main roads is
estimated by 12 % on run-off-road accidents after a migration effect of the accidents.
In another reference, the influence of rumble strips is estimated by 6 %, without
accident migration [30]
In the Safety Evaluation of Rolled-In Continuous Shoulder Rumble Strips Installed on

Freeways, by FHWA in 1999, a before-after field study on 55 locations shows the
positive effect by 13 %.
RISER’s detailed data shows that inappropriate speed or speeding is not the main
factors of accidents. Heavy workload, panic (9 cases), internal or external distraction
(16 cases) and above all fatigue (31 cases out of 189) are other factors of risk. One
can estimate that audible road markings may have played a positive role in some of
these accidents.
3.4 Summary
RISER’s analysis of different criteria for dimensioning the recovery zone has shown
that the design of roadside environments is complex. For a road designer evaluating
alternative designs and choosing among them is difficult because there are many
levels of interaction between different road design components such as the road
itself, speed, traffic volumes and terrain etc.
The information collected among RISER contributing partners’ national policies

clearly show that the width of the recovery zone is different one country from another,
some of the reviewed countries using the general roadside geometry to describe the
safety risks for roadside environment.
From the abovementioned research findings, the recommended width of paved

shoulders on non-motorway roads should be between 1 m and 1.5 m, value beyond
which further widening does not seem to greatly improve safety (see Table 3.1 and
Table 3.2). However these values can be smaller, above all in the outside of curves,
and keep a significant efficiency. With paved shoulder widths greater than 1.5 m
speeds and subsequent accident rates may increase. Further research should be
conducted to confirm this assertion.
Table 3.1. Recovery Zone Width on Non-Motorway Roads
Usage Recommended values
Recovery of errant vehicles 1 m to 1.50 m
Avoidance of overtaking and meeting vehicles 1 m to 1.50 m
Avoidance of vehicles making a turn 1 m to 1.50 m
Travel of vulnerable road users off the driving
0.5 m to 1.5 m
lane
44
Table 3.2. Recovery Zone on Motorways

Usage Recommended values
Emergency lane 3 m to 4 m
3.4.1 Good Condition, Good Surface for Paved Shoulders

Shoulder condition and surface must have the same quality as the road. It must be
constructed so as to bear the static load of heavy trucks.
3.4.2 No Grass, No Gravel, No Loose Material on Paved Shoulders

The material used to construct paved shoulders is very important. The recovery zone
should have the same surface quality as the pavement and skid resistance should be
identical to the carriageway.
Loose materiel or grass shoulders degrade recovery and avoidance manoeuvres. In

addition, bikers and cyclists are reluctant to travel on such uncomfortable material.
3.4.3 Paved Shoulders opposite T-Junctions and Private Accesses

Extra space should be installed opposite T-junctions and private accesses to enable
avoidance of vehicles making a turn towards minor roads and private accesses.
3.4.4 Audible Road Markings as Additional Corrective Actions

In combination with paved shoulders, audible road markings can be implemented
alongside the main roads in order to alert the driver of an errant vehicle.
Rumble strips or edge markings with jiggle bars should be provided to alert a motorist
who is driving in a deteriorated driving situation.
3.4.5 Paved Shoulders on the Outside of Curves

Paved shoulders should be implemented preferably on the outside of curves with
radius greater than 200 m on single carriageway roads.
3.4.6 Discussion
Road designers must recognize that the roadside environment and its design have a
vital role to play in improving roadside safety. A great number of research clearly
show that a recovery zone has a positive effect on both accident rates and driver
behaviour, provided that the abovementioned conditions of implementation are
respected: a smooth and resistant surface made of asphalt or concrete, with no loose
material, wide enough to allow vulnerable road users to make short trips off the
carriageway but not too wide so that car drivers do not understand this roadway
improvement as an extra driving lane.
For economical grounds, further research could be carried out to answer to the
question: what is the optimal paved shoulder and lane widths?
45
References
1 Safety Standards for Road Design and Redesign, SAFESTAR, European project (1998)
2 Brannolte, U., et al, Sicherheitsbewertung von querschnitten ausserörlicher strassen.
Berichte der Bundersansalt für Strassenwesen. Verkehrsteechnik, Helft V5, 1993
3 Foody, T., Long, M., The Identification and Relationships Between Safety and Roadway
Obstructions, Ohio Department of Transportation, Columbus Ohio, 1974
4 Zeeger, C., Deacon, J., Effect of lane width, shoulder width and shoulder type on
highway safety. State of the art report 6: Relationship between safety and key highway
features. A synthesis of prior research, Transportation Research Board, National
Research Council, Washington DC, 1987
5 Zeeger, C., Council, F., Safety relationships associated with cross sectional roadway
elements, Transportation Research Board 1512, TRB , National Research Council,
Washington DC, 1995
6 Brüde and Larsson. Traffic safety effect of wide traffic lanes, Swedish National Road and
Transport Research Institute, VTI meddelande 807,1996
7 Elvik, R., Vaa, T., The Handbook of Road Safety Measures, TØI, Norway, 2004
8 Designing Safer Roads, Practice for Resurfacing, Restoration and Rehabilitation, TRB,
United States, 1987
9 Evaluation des aménagements de sécurité - Safety Evaluation of Corrective Actions,
SETRA, France, to be published
10 Ksaibati, K., Crowe, I., Evaluation of pavement shoulders, University of Wyoming, May
1999
11 Les pertes de contrôle en ligne droite, en rase campagne, quelques éléments
d’information à partir des EDA, Loss of Control in Straight Lines on Rural Roads, a Few
Elements Collected from In-Depth Investigations of Accidents, CETE Normandie-Centre,
Ceesar, France, 2005
12 Les accidents par sortie de chaussée - Run-Off Accidents, CEESAR & CETE
Normandie-Centre, France, 1997
13 Exploitations thématiques des études détaillées d’accidents - Exploration of In-Depth
Investigations of Accidents, INRETS, France, 2000
14 Vejdiretoratet 1984. Kantlinier forbedrer bade bilisters og cyclisters sikkehed, Saertryck
fra Dansk Vejtidsskrift Nr 10/1984, Sekretariatet for Sikkerhedsfremmende
Vejforanstaltninger Nv 1984
15 Sécurité des Routes et des Rues - Safety on Roads and streets, SETRA, France, 1992
16 Yerpes and Ferrandez, F., Caractéristique routière et sécurité INRETS, France, 1986
17 Sourice, O., Thullier, S., Analyse des problèmes de sécurité relatifs au dépassement
ENTPE, 1989
18 Lesot, J.P., Les accidents mortels de pièton en rase campagne, etude de 50 cas SETRA
Non publié, 1990
19 Volken, V., Etude des accidents sans heurt d’obstacle. ENTPE, 1990
20 Michel, Brenac, Magnin, Naude, Perrin, Les pertes de contrôle en courbe. INRETS,
Janvier 2005
21 The Traffic Safety Toolbox, Institute of Transportation Engineers, United States, 1999
22 Les accidents par sorties de chaussée CESAR CETE NC, 1997
23 Influence des accotements revêtus sur les comportement des usagers, cas de la RD982,
de la RD56 en Seine-Maritime, Influence of Paved Shoulders on Driver Behaviour, the
RD982, the RD56 in Seine-Maritime, CETE Normandie-Centre, France, 2001, 2002
24 Accotements et cinématique des véhicules légers - Roadside Environment and Light
Vehicles Kinetics, UTAC/SETRA, France, 1998
46
25 Analyse de l’adhérece des accotements, Analysis of Shoulder Skid Resistance, CETE

de Lyon, France, 2004
26 D04: Identify envelope of vehicle and driver response prior to collisions, European
GRD2/2001/50088, 2005
27 Les accidents par sorties de chaussée CESAR CETE NC, 1997
28 Evaluation des marquages à protubérances, analyse des débords sur BAU d’autoroute
par observation avant-après - Evaluation of Shoulder Rumble Strips, Before/After
Analysis of the Lateral Extent of Encroachments Onto Emergency Lanes, CETE-
Normandie-Centre, France, 1991
29 Viner, J., Griffith, M., Observations on crash studies with continuous shoulder rumble
strips, April 1995
30 Opération Sécurité des Itinéraires, axe 3 = amélioration de la pertinence des études de
sécurité, la zone de récupération, analyse bibliographique, Operation Road Safety on
Main Roads, axe 3, The Recovery Zone, Literature Review
47
CHAPTER 4: TECHNICAL DESIGN STRATEGIES WITH

CASE EXAMPLES
4.1 Introduction
The development of a safe roadside environment depends on the identification of the
hazards in the roadside environment (Chapters 1-3). Once these hazards are
identified and are known to be in the safety zone, suitable strategies (or
countermeasures) are needed to protect traffic from these hazards. The purpose of
this section is to provide guidance for the technical treatment of hazards on both new
and existing roads.
As described in Chapter 1 the hazards that are most common on European roads
can be divided into 2 classifications: point hazards and distributed hazards. These
two classifications create different procedures for selecting a countermeasure for the
hazard. However, the general strategy illustrated in Figure 4.1 can be applied to both
cases.
Hazard No
Evaluation
Yes
Hazard Yes
Removal
No
Hazard Yes Final

Actions
Modification Solution
No
Hazard Yes
Protection
Figure 4.1. Strategy for Hazard Countermeasures
In a review of the accident statistics in the RISER project, a great number of

accidents involved collisions with hazards located in the safety zone. This indicates
that there is a considerable problem at the first decision point in Figure 4.1 –
identifying the hazard. Once the hazard is identified, the selection of appropriate
actions usually is uncomplicated. The initial identification of the hazards is a common
problem throughout the world. When several hazards are identified and priorities
need to be set, it is useful to review the risk of an impact with the type of object and
then the consequences of an impact with that object. Figure 4.2 shows the frequency
48
of impacts versus the consequences of impacts based on the information in the

RISER statistical database. The dashed lines are constant risk curves, the points in
the upper right corner are high frequency, high fatality hazards. Trees are thus an
important priority for improving roadside safety, but the "No object" point indicates
vehicles rolling over in the terrain. As expected, safety barriers have a high frequency
but a lower fatality frequency.
40%
35%
30%
Trees
Severity of Collision
25%
20%
No object
Ditch
15%
Other Object
Other Man Made
10% Object
Safety Barrier
Post
5%
Other Natural
Object
0%
0% 5% 10% 15% 20% 25% 30%
Probability of Collision
Figure 4.2 Risk Information for Various Struck Objects
A brief introduction to the issues of technical design of roadsides will be presented

followed by a more specific description of the different designs.
4.2 Overview of Hazard Protection

4.2.1 Point Hazards
Hazards that are restricted to a small area as identified in Table 1.9 can be
subdivided into man made and natural features. From the RISER statistical database,
the two primary point hazards overlooked in roadside areas are narrow objects like
trees and poles. It is important that the safety zone is properly dimensioned for the
road classification and that any trees or poles are removed from this area.
Action 1 Remove the hazard. Man-made features in the roadside safety zone
should only be there because of a functional requirement (lighting, signs,
bridge support, etc.). If they are not required, then they should be
removed. Trees and poles that are located in the safety zone but cannot
be removed (aesthetic or functional requirements) need to be made less
harmful to vehicles.
49
Action 2 For man-made hazards, modify the hazard becomes the next step. For
lighting and utility columns, energy absorbing and break-away structures
are important structures to incorporate into the roadside area. The two
different structures – deformable and rigid – are shown in Figure 4.3. The
outcome from these two motorway accidents was quite different even
though the impact speeds were quite similar.
Figure 4.3. Passive Safety Infrastructure

(picture to left courtesy Jan Wenäll, VTI Sweden)
Action 3 Natural features like trees are sometimes difficult to remove from the
roadside area due to historic and aesthetic requirements. Therefore the
third task is sometimes more applicable – protect the road user from the
hazard (see Figure 4.4).
Figure 4.4. Rigid Lighting Columns protected with Guardrail
Protecting the point hazard introduces added complexity to the roadside design. In
addition to selecting a type of road restraint system, the designer must also consider
the placement of the safety feature and the size of the area to be protected (see
Figure 4.5). An important issue is the transition from single point hazards to
distributed point hazards (such as a group of trees or lighting poles).
50
Figure 4.5. The culvert is protected with a short guardrail but the gantry pole and the old
rigid lighting column are unprotected (Photo, HUT)
4.2.2 Distributed Hazards

Distributed hazards by their nature encompass larger areas than point hazards (see
Figure 4.6). This can result in higher costs than for the point hazard. A similar
process, Figure 4.1, can be applied to distributed hazards as for the point hazards.
Action 1 Remove the hazard: As for point hazards, the nature of the hazard will
determine what is possible. Man-made distributed hazards should be
designed so that they are not located within the safety zone. Similarly,
natural hazards (rock faces, groups of trees, etc) should be set back
from the road so that they are not located in the safety zone.
Figure 4.6. Examples of the distributed hazards. Left: unprotected rock cut very close to the
carriageway. Right: untested guardrail with concrete posts and too low
positioning (Photo, HUT)
Action 2 Modify the hazard: Almost all distributed hazards of concern are related
to the roadside geometry. RISER statistical accident data indicate that
roadside geometry; including slopes, embankments and ditches (or no
specific impacted object), contribute to almost half of all run-off-road
accidents involving injury or fatality. These roadside features are
believed to be the leading cause of rollover in single-vehicle, run-off-road
accidents. The layout of the side slopes and ditches adjacent to the road
51
are the main features that can be modified from a dangerous situation to
a more gentle geometry.
Action 3 Hazard shielding: The application of road restraint systems (safety

barriers) to protect distributed hazards is the best alternative when a
hazard cannot be relocated outside the safety zone. The safety barrier
must be selected to provide suitable protection for the exposed traffic
and the dimensions of the hazard.
4.3 Analysis and Recommendations of Specific Hazard Protection

4.3.1 Point hazards Case Examples
Trees
RISER statistical database shows that trees are the most commonly impacted object
in fatal single vehicle accidents among RISER countries (28 % of fatalities).
RISER detailed database includes 30 accidents (of 211) where tree was the main or
only impacted object. The set-back of trees from the edge of the carriageway ranged
from 2.0 m to 10.8 m. In all fatal accidents with reconstruction data the vehicle struck
the tree at a speed greater than 70 km/h.
In all RISER countries the trees are considered as the main concern when they are
too big and too close to the carriageway.
Recommendation:
1. Remove trees which are inside the safety zone

Usually the easiest and best method to decrease the risk of collision to trees
is removal of all trees inside the safety zone (see Figure 4.7).
Figure 4.7. Example of a Clear Safety Zone when trees have been removed (Photo, HUT)
2. Shielding with safety barrier

In some cases the trees (row of trees) are considered as an important
aesthetic part of the roadside area. They can also have an important role in
functionality of a self-explaining road scene.
52
Boulders
All boulders and smaller hazardous stones inside the safety zone should be
removed. If that is not possible for some reason the boulders must be shielded with a
safety barrier.
Bridge Abutments and Bridge Piers

Earlier studies indicate that the risk of fatalities is outstandingly high in crashes to
bridge abutments and bridge piers (Figure 4.8) compared to most other roadside
obstacles [1, 2, 3]. The easiest, most cost-effective and usually also the only way to
protect the bridge abutments and bridge piers is shielding it with a safety barrier.
Figure 4.8. Example of hazardous, unshielded bridge pier and bridge abutment
Recommendation:
1. Shielding with guardrail

All bridge abutments and bridge piers must be shielded with a safety barrier.
Special attention must be paid to the sufficient working width and sufficient
length of the safety barrier.
Underpasses
The underpassing minor road or small water course – especially their opposite
embankment or retaining wall - is extremely dangerous for the encroaching vehicles
(see Figure 4.9).
Figure 4.9. Hazardous underpass on motorway, notice too short safety barriers
53
1. Shielding with guardrail

Travelling on the roadside or on the median and encroaching to the
underpass must be prevented by using safety barriers which are long
enough for this purpose. The length of need and the flaring of the safety
barrier will probably ensure that an errant vehicle cannot collide into the
underpass (see Figure 4.10).
Figure 4.10. Underpass opening on motorway median protected with safety barriers.
(Photo, HUT)
Culvert Ends
Culvert ends are very hazardous impact objects, e.g. on minor road junctions (Figure
4.11). If the vehicle runs off the road into the ditch, there is a risk of an impact to the
culvert end or to the minor road side slope.The solutions are to remove the obstacle,
design or redesign of the culvert location or use of safer construction in the ends of
the culvert.
Figure 4.11. Examples of culverts in minor road junctions: traditionally the ends are vertical
and the side slopes of the minor roads steep which makes them very hazardous
objects in collisions (Photo, HUT)
Recommendation:
1. Reduction of the number of minor road junctions (and culverts) by the

means of traffic planning. The safest solution is always the removal of minor
road junctions and culverts, if that is possible with reorganisation of the
entrances to the major road.
2. Culverts parallel to the main road: new design or redesign of the position.
Culverts must be positioned far enough from the main road so that an errant
54
vehicle which is travelling along the ditch cannot collide with the termination
of the culvert (Figure 4.12).
Figure 4.12. Errant vehicle is prevented to impact the end of the culvert by adequate set back
of the culvert. Safe set back of the culvert depends on the width of the ditch and
width of the safety zone (figure: Finnish Road Administration; TYLT 6800-6870).
3. Culverts parallel to the main road: design or redesign of the terminal. The
bevelled ends of the culverts and gentle side slopes of the minor roads are
safer in collision than traditional vertical culvert ends and 1:1 – 1:2 side
slopes (Figure 4.13)
Figure 4.13. Bevelled culvert ends in minor road junctions (Photo, HUT)
4. On existing roads it may be more cost effective to protect the culvert

terminals than reshape them and the side slope or relocate the culverts.
Some experimental applications exist, but as far as is known there are no
generally accepted applications yet in any European design guidelines.
Culverts transverse to the main road
As stated in RISER deliverable 5 [4] culverts which are transverse to the
road should have the terminal bevelled with gradient 1:1 to 1:4 (Figure 4.14).
55
Figure 4.14. Bevelled end of a concrete culvert: safer design than traditional vertical end
shape (Photo, HUT)
Lighting Columns
Rigid lighting columns and utility poles are very common and severe hazards in the
roadside. On new roads all poles should be located beyond the safety zone. If the
location inside safety zone can not be avoided only passively safe columns should be
used. All columns should be tested according to the current EN standards (EN
12767).
Case Finland: today more than 90 % of new lighting columns installed for Finnish
Road Administration are break-away ones. Most of them are wooden poles or
yielding steel columns, not many slip base ones. Break-away columns are only
slightly more expensive than columns with no passive safety [5].
Recommendation:
1. New installation or replacement to passively safe columns

If installation into safety zone is necessary only passively safe columns
should be used (see Figure 4.15). Energy absorbing columns should be
preferred if there is pedestrian path or hazards (e.g. trees) not far behind a
column. Old rigid steel, concrete and wooden lighting columns can be
replaced with energy absorbing or break-away steel, aluminium, wooden or
composite columns.
2. Modification of rigid columns

Wooden and steel columns can be modified to break-away ones (see Figure
4.16). There are several techniques for that purpose.
56
Figure 4.15. Left: Break-away wooden column; glued and laminated. Right: Yielding, energy
absorbing steel column (Photo, HUT)
Figure 4.16. Left: old wooden lighting column. Right: old steel lighting column, both are
modified to break away at impact.(Photo, HUT)
Utility Poles
Utility poles are commonly located along roadsides. Nowadays there is no traffic-
related reason for the location, but due to the huge number of utility poles it is not
possible to relocate all the poles in the near future. In TRB Guide for Reducing
Collisions Involving Utility Poles, the objectives to reduce the number and severity of
impacts include:
reduce the hazard of specific utility poles in high-crash and high-risk

locations
prevent placing new utility poles in high-risk locations
treat several utility poles along a road to minimize the likelihood of crashing
into a utility pole when vehicles run-off-road
57
Recommendation:
1. Remove poles inside the safety zone

2. Locate/relocate poles beyond the safety zone
3. Place the utilities underground
4. Decrease the number of poles
– sometimes it is possible to decrease the number of utility poles by
widening the spacing or relocating the utilities to be placed only on one
side of the road
5. Use break-away poles
– recommended alternative for the new installations if the poles for some
reason must be located inside the safety zone
– old poles can also be modified to break-away ones
6. Shield a pole with a safety barrier
– option if an utility pole cannot be removed, relocated or modified with
reasonable costs
Gantry Poles
Gantry poles are normally rigid, very strong steel or concrete supports without any
energy absorbing properties. In most cases they are located close to the roadway
(see Figure 4.17). It is recommended that all kind of impacts to rigid gantry poles are
prevented.
Figure 4.17. Unprotected gantry poles in the safety zone are fatal hazards (Photo, VALT)
Recommendation:
1. Shielding with safety barrier (see Figure 4.18) note that the use of unflared,
turned down barriers is not recommended from the results of RISER
accident studies.
2. Use break-away gantry poles. Nowadays there are also break-away gantry
structures on the market.
58
Figure 4.18. Gantry pole shielded with safety barrier, note that the use of unflared, turned down
barriers is not recommended (Photo, HUT)
Small Sign Supports

Traffic sign supports are considered as hazards if they are not proved to be safe in
impact or if their dimensions exceed the limits presented in Chapter 1.
Recommendation:
1. Relocate the sign support beyond the safety zone.

Locating the traffic sign or other sign support outside the safety zone is
preferable if it is possible to carry out without degrading the visibility of the
sign.
2. Use passively safe supports
If sign support must be located inside safety zone, the impact severity can
be significantly reduced by using a passively safe device (see Figure 4.19).
Figure 4.19. Small break-away sign support
4.3.2 Distributed Hazards Case Examples
Fill Slopes
Fill slopes and fore slopes of the ditches are considered as recoverable if they are
1:4 or flatter. Slopes from 1:3 to 1:4 are considered as non-recoverable, but
traversable. Slopes steeper than 1:3 are critical; possibility to rollover of the vehicle
increases substantially [6].
59
Recommendation:
1. Low traffic main roads, design speed 70 km/h or less

– embankment height < 3.0 m: side slope 1:3 or flatter
– embankment height 3.0 m: also fill slopes steeper than 1:3 can be used
with an appropriate safety barrier (see Figure 4.20)
2. Busy main roads
– embankment height < 3.0 m: side slope 1:4 or flatter, toe of the slope(or
bottom of the ditch) rounded but note that heights over 1 m were
hazardous in the accident analyses (Chapter 1)
– embankment height > 3.0 m: fill slopes steeper than 1:3 can be used with
an appropriate safety barrier
Figure 4.20. Fill slope with a safety barrier, height exceeds 3.0 m.
Ditches and Cut Slopes

Ditches are needed for the drainage of the road structure. Ditches cause two kinds of
severe accidents: rollovers and crashes to the back slope. To decrease these kinds
of accidents the geometry of the cross-section must be gentle enough. When a
traversable shape of the ditch is considered, the terrain behind the ditch (trees,
pedestrian path etc.) must also be taken into account.
The behaviour of errant vehicle in 1:3 fore slope and 1:2 back slope ditches is
investigated with full-scale tests [7]. In V-shaped ditch it was discovered that
passenger car tends to crash to the back slope and then overturn, if the speed
exceeds 80 km/h and exit angle is 20°. With smaller angles the vehicle travelled in a
1:2 back slope beyond the height of 1.5 m. Next tests proved that rounding of the
bottom prevents a car to overturn. As a conclusion it was decided to prefer rounded
bottom ditch shape with fore slope 1:4, bottom width 1.0 m and back slope 1:2.
Particularly in cases when a deep ditch is needed the use of a covered drainage
system should be considered.
Recommendation:
1. Ditches and cut slopes along busy main roads

If the back area allows the traversable design then it is very preferable: fore
slope 1:4 or flatter, width of the bottom 1 m or wider, back slope 1:2 or
60
flatter. Examples of the preferable design are shown in Figure 4.21 and
Figure 4.22.
Figure 4.21. Recoverable ditch shapes: 1:4 fore slope, rounded bottom and 1:2 back slope.
(Photo, HUT)
Figure 4.22. Recoverable fill slope; bottom of the ditch is rounded, no hazardous obstacles
in fill slope. (Photo, HUT)
2. Ditches and fill slopes along low volume main roads, design speed 70 km/h
or less. On low volume and low speed road sections also the V-shape ditch
with fore slope 1:3 and back slope 1:2 can be considered.
Forest/Line of Trees
Recommendation:
1. Removal
2. Shielding with safety barrier
In some cases the trees (row of trees) are considered as an important
aesthetic part of the roadside area. They can also have an important role in
functionality of self-explaining road scene (see Figure 4.23).
61
Figure 4.23. There is evident need for better roadside safety because of trees and steep
slope. In this road section the only solution for better roadside safety is safety
barrier – by choice the guardrail with aesthetical value
Rock Face Cuttings

Recommendation:
1. Removal from safety zone (new roads) (see Figure 4.24)

2. Shielding with safety barrier (see Figure 4.25)
3. Protective earth slope (see Figure 4.25)
Figure 4.24. Rock face cutting removed from the safety zone (Photo, HUT)
Figure 4.25. Rock face cutting shielded with safety barrier (Photo, HUT)
62
Retaining walls (parallel to the road)

Recommendation:
1. Safety barriers are needed when retaining walls are within the safety zone.
4.3.3 Roadside and Median Road Restraint Systems Considered as Hazards

Existing guardrails may have many critical weaknesses in their positioning,
dimensioning and crash safety properties.
Recommendation:
1. Installation of guardrail at the places where there has been no guardrail

before
2. Increasing the length of pre-existing guardrail i.e. extension of guardrail
(Figure 4.26)
3. Modernisation of old guardrails
Figure 4.26. Left: too short guardrail, Right: lengthened guardrail (new section with weaker
posts). (Photo, HUT)
Transition
Figure 4.27. Left: good solution of transition structure, the guardrail is overlapped with the
concrete barrier and fastened with steel bolt. (Photo, HUT) Right: poor solution of
the transition from steel guardrail to the concrete pillar of the bridge parapet.
(Photo, VALT)
63
Terminations of Safety Barriers

Safety barrier ends which do not fulfil the requirements of EN 1317 (or NCHRP 350)
can be point hazards themselves. Blunt ends of safety barriers are well-known
hazards, but also ramped ends of guardrails parallel to the road can easily cause a
vehicle vault or rollover and hence lead to more severe consequences.
In RISER detailed accident database there are 41 accidents where the barrier was
the only obstacle involved. In 14 cases the termination of the barrier was impacted; in
four of those cases the vehicle travelled along the top of the barrier until it came to a
stop or impacted another object, in 10 cases the vehicle was launched into the air.
The injury risk of impact to the safety barrier end can be reduced by flare the end into
the roadside or installing tested end treatments like energy absorbing terminals
Recommendation:
1. Replacement with tested energy absorbing end treatment

Energy absorbing end treatments are safer than ramped ends if an errant
vehicle collides with the end of the guardrail (see Figure 4.28).
2. Bending the end to the slope
The ramped ends can be turned to the slope so that the straight impact onto
the ramped end is prevented. With this treatment the vehicle is also disabled
to travel behind the barrier and possibly collide with the shielded object (see
Figure 4.28).
Figure 4.28. Energy absorbing guardrail terminal and flared guardrail terminal (Photo, HUT)
3. Replacing the blunt end with ramped end

The blunt ends should not be used on the European main road network. As
a minimum requirement it is recommended that those terminations are
replaced with ramped ends. Recommendations 1 and 2 should be the
choice for any new design or redesigned road sections.
64
References
1 Kelkka, M. Performance of guardrails and the requirements for guardrail development.

Helsinki University of Technology. Espoo 2002. ISBN 951-22-5942-7.
2 Zegeer, C. V. & Reinfurt, D. W. & Hunter, W. W. & Hummer, J. & Stewart, R. & Herf, L.
Accident Effects of Sideslope and Other Roadside Features on Two-Lane Roads.
Transportation Research Record 1195. TRB, National Research Council. Washington
1988
3 . G. Viner. "The Roadside Safety Problem," Transportation Research Circular 435,
Transportation Research Board/National Research Council, Washington, D.C., January
1995
GRD2/2001/50088, 2003
5 Finnra Engineering News No 9E. Break-away Lighting Columns in Finland, year 2005.
Finnish Road Administration. Helsinki 2005. http://alk.tiehallinto.fi/thohje/fen.pdf NCHRP
Report 500. Volume 8: A Guide for Reducing Collisions Involving Utility Poles.
Transportation Research Board. Washington D.C. 2004.
6 Roadside Design Guide 2002. AASHTO. Washington D.C. 2002. ISBN 1-56051-132-X.
7 Thomson R., Valtonen, J., Vehicle Impacts in V Ditches, Transportation Research
Record 1797, Pages: pp 82-88, 2002
65
CHAPTER 5: PASSIVE SAFETY ROAD EQUIPMENT
5.1 Introduction
Passive safety road equipment includes Road Restraint Systems (RRS), breakaway
equipment supports, energy absorbing supports, and arrester beds. The equipment
included in RRS can be divided into two groups, vehicle and pedestrian restraint
systems. Vehicle restraint systems comprise safety barriers and bridge barriers
(parapets), terminals for barriers, crash cushions, transitions among different RRS,
and Motorcyclist Protection Devices (MPD). Pedestrian restraint systems include
pedestrian parapets and guardrails for separation of pedestrians and traffic. Passive
safety structures constitute, together with road markings and signing, a basic element
for road safety.
Passive safety road equipment (or road equipment) is implemented at locations

where there are serious consequences when a vehicle makes an uncontrolled exit
from the roadway. It must be recognized that passive safety equipment represents
objects that can be struck by a vehicle. However, they are constructed and tested to
ensure that any collision with a passive safety structure is less severe than with a
hazard located on the side of the road.
The necessity of installing road equipment is made when the roadside area cannot
be made safer using the safety zone concept (Chapter 2). As described in Chapter 4,
some roadside hazards can be removed or made less dangerous through different
countermeasures. However many situations arise where the hazard cannot be
removed or when it is more economical to install road equipment. The purpose of this
chapter is to identify the criteria and design implications for road restraint systems.
5.2 Classification and Function of Passive Safety Road Equipment

The main categories of passive safety road equipment are presented to provide
information on the system structure and typical installation locations.
5.2.1 Safety Barriers

Safety barriers are installed lengthwise on the roadsides or the central reservation of
a roadway. They are designed to restrain an errant vehicle from leaving the roadway
and redirect the vehicle towards the traffic lane. Safety barriers are developed and
designed for oblique impacts only.
Safety barriers can be:
Longitudinal barriers
– barriers designed for vehicle impacts to one side only and are installed
on the edge of the carriageway.
Median barriers
– barriers designed to be mounted between opposing travel lanes and can
be impacted on both sides.
Bridge parapets
– specialised longitudinal barriers designed to be installed on bridge decks.
66
Safety barriers can be subdivided into temporary or permanent barriers, depending

on the intended lifetime for their installation. The main difference between these
systems is the type of anchorage to the roadbed and the designed impact protection
level. Temporary barriers are almost exclusively installed in work zones where lane
widths and traffic speeds limit the severity of a vehicle impact.
5.2.2 Crash Cushions

Crash cushions (sometimes called energy absorbers or impact attenuators) are
designed for protecting a point hazard (see Figure 5.1). In contrast with safety
barriers, crash cushions are designed for impacts at the end points of their structures,
as well as impacts along their sides. Their performance for oblique impacts is used to
divide crash cushions into the following categories:
Redirect
– designed for oblique impacts and can behave like a safety barrier for
short sections.
Non redirect
– no capacity for oblique impacts.
Crash cushions should only be installed at a location when the protection of vehicles
from a hazard cannot be solved by the regular process outlined in Chapter 4. When
protection of a hazard is required, safety barriers should be the first option if space
permits. Crash cushions are notably more expensive than safety barriers.
Figure 5.1. Example of Crash Cushions
5.2.3 Terminals/End Treatments

The terminals or end treatments for safety barriers are special constructions applied
to the end of the barriers (see Figure 5.2). All safety barriers have ends and these
sections should not represent a hazard for the traffic. The terminal sections are
designed to be impacted by a vehicle in both "end on" and oblique trajectories.
67
Figure 5.2. Safety Barrier Terminal
A terminal can work by:
a. reducing the speed of the vehicle.

b. allowing a controlled penetration of the vehicle behind the barrier.
c. retaining and redirecting the vehicle.
d. combining the functions a, b and c.
The terminal type is categorized by means in which it operates and if it is designed to

allow the vehicle behind the barrier. A Non-Redirecting end treatment will not fully
redirect the vehicle during an oblique impact but a Redirecting does not permit the
vehicle to stop behind the barrier.
5.2.4 Transitions
A transition is defined as the structure that connects two safety barriers of different
geometry and/or containment levels and/or lateral deformation. A transition is
designed so that there are no abrupt changes in the safety performance if a vehicle
strikes the area between two different barrier types (see Figure 5.3). Common
locations for transitions are connections between bridge parapets and roadside
safety barriers.
Figure 5.3. Transition
68
5.2.5 Arrester Beds

Arrester beds are areas adjacent to the roadway section in areas with long
downgrades (see Figure 5.4). These facilities are used for vehicles (typically heavy
goods vehicles) that have lost their braking capacity. When braking problems are
identified, the driver intentionally drives onto the arrester bed which has a specific
surface material that allows the vehicle to be brought to a stop.
Figure 5.4. Arrester Bed
5.2.6 Break-Away or Energy Absorbing Structures

Break-away or energy absorbing structures are typically used for lighting columns,
utility posts, sign posts, etc. These structures are designed so they break or deform
in a controlled way when impacted by a vehicle (see Figure 5.5).
Figure 5.5. Break-Away Utilities
5.3 European Standards Applicable to Roadside Infrastructure

The objectives of the European Standards EN 1317 - Road Restraint Systems [1]
and EN12767 - Passive Safety of Support Structures for Road Equipment [2] are to
provide a common testing and reporting system and to provide clear understanding
of the design, performance, production, and construction of road restraint systems.
The current EN1317 document contains 5 parts relevant for roadside design
guidelines. Parts 1-4 describe how the crash protection of different types of RRS is
determined during full scale crash testing. Part 5 describes the durability and
documentation for conforming to the standard. Part 6 applies to pedestrian protection
and only relevant for the separation of pedestrians and traffic. EN12767 lists testing
requirements for passive safety structures like break-away and energy absorbing
69
lighting columns. The following discussion of these standards reflects the printed
version of the standards listed in the reference. These standards are continuously
reviewed and subject to change.
5.3.1 Requirements and Test Methods

It is important to note that the standards are developed to describe the performance
of road equipment using parameters that are appropriate for the protective behaviour
of the specific equipment type. These parameters are relevant for the installation of
the equipment in accordance with the geometric characteristics particular to the road
and traffic conditions. The standard was not developed to provide direct information
on the implementation (installation) of the equipment other than specific details, like
the anchorage of the system, which are specified in the test and conformity reports.
The installation location and selection of a specific product for that location are not
covered by the standards and are the role of guideline documents (like this
document). The role of a standard is to provide a method for comparing different
products that are relevant for a specific installation location.
Structure of EN1317
The European Standard EN 1317 is organized in six parts that are:
1. Terminology and general criteria for test methods.

2. Performance classes, impact test acceptance criteria and test methods for
safety barriers and vehicle parapets.
3. Crash cushions - Performance classes, impact test acceptance criteria and
test methods for crash cushions.
4. Impact tests acceptance criteria and test methods for terminals and
transitions of safety barriers.
5. Durability criteria and evaluation of conformity.
6. Pedestrian road restraint systems.
At the time of publication, parts 1, 2, and 3 are official European standards while part
4 is published as an ENV (Experimental Standard for a 3-5 year period). Part 5 is in
the final stages of becoming a formal standard. Any information presented in the
following sections reflects the current printed status of the standard [1].
The products regulated in part 6 (Pedestrian road restraint systems) are not intended
to protect the occupants of a vehicle during a run-off-road event. The pedestrian
restraint systems are implemented to separate pedestrians from road traffic and are
thus only dimensioned to restrict people from walking onto the roadway. This part is
not applicable for roadside design guidelines.
The following description of the RRS requirements and characterization are based on
parts 1 and 2 and are relevant for safety barriers. Corresponding requirements are
valid for crash cushions, transitions, etc. and more details can be found in the
relevant part of the standard EN1317. It is recommended that manufacturers and
suppliers of road restraint systems are consulted when selecting road equipment and
their components so that no performance conflicts are created.
70
5.3.2 Performance Criteria

The parameters used to evaluate the effectiveness of the RRS, to define the limits of
acceptance, and to identify the technical classes are:
Containment level of RRS

Impact severity
Deformation of the barrier
In addition, there are requirements that the vehicle is redirected smoothly without
rolling over.
The crash tests are conducted using vehicle masses prescribed in the test standards.
Depending on the intended application of the system, vehicle test masses range from
900 kg to 38000 kg. Additional vehicle characteristics such as wheel base and centre
of gravity height are specified for each vehicle type. Test speeds and impact angles
are defined for the different test categories for specific vehicle types.
Containment Level of RRS

The containment level of RRS is the capacity of the system to redirect a test vehicle
with a prescribed mass, impact speed and impact angle, see Figure 5.6. It should be
noted that the containment level indicates the most severe (in terms of lateral kinetic
energy) loading capacity of the system. In addition to the severe loading condition, a
test with a small (900 kg) vehicle is required for the system to ensure that vehicle
occupants are suitably protected. Thus the containment level indicates the maximum
loading of a RRS while providing safe protection for smaller vehicles.
α = impact angle
Safety Barrier
Figure 5.6 Impact Configuration
Suitable behaviour of the RRS requires that the vehicle never penetrates or overrides
the system. No deformation of the system shall cause parts of the RRS to penetrate
the vehicle passenger compartment. The system shall contain or redirect the test
vehicle in a controlled and predictable manner.
The European standard establishes that a RRS tested and meeting the desired
containment level is also qualified for application for less severe containment levels.
Containment levels for a road restraint system can be determined from the test
conditions described in the following 2 tables. Table 5.1 describes the test
configurations with the specific vehicle, speed and impact angle and Table 5.2
provides the test combinations that define the different containment levels for
longitudinal barriers. Similar information is available in EN1317 for crash cushions
and barrier end terminals.
71
Table 5.2 also provides information about the lateral kinetic energy for the test and
the vertical position (height) of the vehicle'
s centre of gravity above the ground. The
lateral kinetic energy is a useful reference for comparing the impact severity of a
crash test and is calculated by the equation:
M veh
T= (V sin(α )) 2 Equation 5-1
2
where T= lateral kinetic energy, Mveh = vehicle test mass [kg], V = test speed [m/s],
and α = impact angle
Table 5.1. Vehicle Impact Test Criteria
Lateral Centre of
Impact Impact Total Test
Type of Kinetic Gravity
Test speed angle mass
Vehicle Energy Height
(km/h) α [deg.] Mveh[kg]
T [kJ] [m]
TB 11 100 20 900 Car 41 0.49
TB 21 80 8 1 300 Car 6
TB 22 80 15 1 300 Car 22
0.53
TB 31 80 20 1 500 Car 43
TB 32 110 20 1 500 Car 82
TB 41 70 8 10 000 Rigid HGV 37 1.50
TB 42 70 15 10 000 Rigid HGV 127 1.50
TB 51 70 20 13 000 Bus 287 1.40
TB 61 80 20 16 000 Rigid HGV 462 1.60
TB 71 65 20 30 000 Rigid HGV 572 1.9
TB 81 65 20 38 000 Articulated HGV 725 1.9
Table 5.2 Containment Levels for Longitudinal Barriers

Containment levels Acceptance test
Low angle containment
T1 TB 21
T2 TB 22
T3 TB 41 and TB 21
Normal containment
N1 TB 31
N2 TB 32 and TB 11
Higher containment
H1 TB 42 and TB 11
H2 TB 51 and TB 11
H3 TB 61 and TB 11
Very high containment
H4a TB 71 and TB 11
H4b TB 81 and TB 11
NOTE 1: Low angle containment levels are intended to be used only for temporary safety barriers.
Temporary safety barriers can also be tested for higher levels of containment.
NOTE 2: A successfully tested installation at a given containment level shall be considered as
having met the test condition of a lower level, except that N1 and N2 do not include T3.
NOTE 3: Because testing and development for very high containment safety barriers in different
countries has taken place using significantly different types of heavy vehicles, both tests TB 71 and
TB 81 are included in the standard at present. The two containment levels H4a and H4b should not
be regarded as equivalent and no hierarchy is given between them.
By using the energy values presented in Table 5.1 and Table 5.2, the lateral kinetic
energies associated with the different containment levels can be presented
72
graphically as in Figure 5.7. This figure shows the exponential increase in lateral
kinetic energy as the containment level increases.
800
700
600
Lateral Kinetic Energy [kJ]
500
400
300
200
100
0
T1 T2 T3 N1 N2 H1 H2 H3 H4a H4b
Low Angle Containment Normal Containment Higher Containment Very High
Containment
Figure 5.7 Comparison of Lateral Kinetic Energy for the Different Containment Levels
Impact Severity
The severity of the impact is defined as the injury risk for the vehicle occupants.
There are 3 parameters used to evaluate injury risk, mostly based on the
accelerations measured at the vehicle' s centre of gravity. In addition to these
parameters, one other criterion on vehicle deformation is provided as information so
that the end user can learn more about the system performance.
Normative Information:
ASI (Acceleration Severity Index)

The objective of the ASI is to compare the maximum acceleration levels that
an occupant is exposed to during the impact. ASI is a function of time,
computed using an equation based on limit acceleration values and the
acceleration of a selected point P of the vehicle, averaged over a moving
time interval of 50 ms.
THIV (Theoretical Head Impact Velocity)
THIV is the calculated impact speed of a free moving mass with an interior
surface of the vehicle. The objective of the THIV is to indicate the risk of
injury for an unbelted occupant inside the vehicle. It attempts to recreate the
motions of an occupant head inside the vehicle.
73
PHD (Post-impact Head Deceleration)

PHD is a parameter that is reported in conjunction with the THIV. At the time
the THIV is calculated, the vehicle decelerations are subsequently monitored
and the maximum resultant of the lateral and longitudinal vehicle
accelerations is reported. The objective of the PHD is to indicate the severity
of occupant loading after they have contacted the vehicle interior.
Informative information:
Deformation of the cockpit

The deformation of the cockpit of the vehicle is measured through the index
VCDI (Vehicle Cockpit Deformation Index). It describes how much the
interior surfaces of the passenger compartment are deformed inwards
toward the occupants. This reduction of the passenger compartment space
is strongly linked to occupant injuries in all motor vehicle accidents.
How to report the impact severity, shown in Table 5.3, is under discussion at the time
of publication (February 2006). The official version of standard currently classifies the
impact severity into two classes. This classification is currently under review.
Table 5.3. Impact Severity Levels [1]
Impact severity level Index Values
A ASI 1.0 THIV 33 km/h
and
B ASI 1.4 PHD 20 g
NOTE 1: Impact severity level A affords a greater level of safety for the occupants of an errant
vehicle than level B and is preferred when other considerations are the same.
NOTE 2: At specific hazardous locations where the containment of an errant vehicle (such as a
heavy goods vehicle) is the prime consideration, a vehicle restraint system with no specific impact
severity level may need to be adopted and installed. The index values recorded in the test of the
restraint system shall however be quoted in the test report.
Deformation of the Safety Barrier

The deformations of the safety barrier are described by the working width and the
dynamic deflection and are recorded during the crash test. Figure 5.8 illustrates the
parameters of interest during a crash test.
The working width (W) is the distance between the side facing the traffic before the
impact of the road restraint system and the maximum dynamic lateral position of any
major part of the system. If the vehicle body deforms around the vehicle restraint
system so that the latter cannot be used for the purpose of measuring the working
width, the maximum lateral position of any part of the vehicle shall be taken as an
alternative (lower images in Figure 5.8).
During impact tests using buses and Heavy Gods Vehicles (HGV), the extreme
lateral position of the system and the extreme lateral position of the test vehicle shall
be recorded separately in the test report.
The dynamic deflection (D) is the maximum dynamic lateral displacement of the side
facing the traffic of the restraint system. For narrow restraint systems, the dynamic
deflection can be difficult to measure and if such is the case, the dynamic deflection
may be taken as the working width.
74
The purpose of these measurements is to identify the amount of lateral space that is
required for the system to operate properly. If the system is installed beside a hazard,
but the hazard is within the working width of the system, then the system cannot
properly protect the hazard since it may come into contact with the hazard during an
impact.
Figure 5.8. Barrier and Vehicle Deflection Parameters
Test Vehicle Behaviour

Since the purpose of the RRS is to contain and redirect a vehicle, the motions of the
vehicle during the test should not result in hazardous conditions for the vehicle
occupants. For these reasons the vehicle behaviour during a test must fulfil the
following criteria:
The centre of gravity of the vehicle shall not cross the centreline of the
deformed system.
The vehicle shall remain upright during and after impact, although moderate
rolling, pitching and yawing are acceptable.
The vehicle shall leave the safety barrier after impact so that the wheel tracks of the
vehicle are within a prescribed "exit box" (Figure 5.9). The exit box is determined
from the contact locations on the barrier and the size of the vehicle prescribed for the
test.
75
Figure 5.9 Vehicle Redirection Requirements
5.4 Influence of Passive Safety Road Equipment on Road Safety

The presence of road equipment should not have, in general, any influence on the
frequency of vehicles leaving the travel lanes and entering the roadside area in an
uncontrolled manner. They should, however, affect the consequences of the vehicle
encroachments into the roadside. The role of road equipment is to improve the
survivability of vehicle occupants during a single vehicle accident.
The driver simulator study (RISER Deliverable 4 [3]) demonstrated that the presence
of a safety barrier can have minor influences on the traffic positioning and speed.
Improper placement of the barriers can conceivably increase this effect and care
should be taken when placing the barrier along the carriageway that it does not
adversely affect the traffic flow.
5.4.1 Cost Benefit

It is difficult to accurately determine the cost/benefit ratios for road infrastructure in
Europe. The complete life-cycle cost analysis of a specific safety feature must include
the installation cost, maintenance costs and the changes to the injury (societal) costs.
As demonstrated in RISER Deliverable 1 [4], there is a lack of maintenance
information for developing European wide cost/benefit figures. The information
provided for one region of Spain showed that 5 times the number of encroachments
are identified by maintenance workers than reported to the police. Until this type of
data becomes easily accessible, the information from regional studies must be used
for reference purposes.
Some general information can be obtained from Table 5.4 and Table 5.5 which show
the benefit cost ratios for different road infrastructure in the USA and Spain,
respectively.
Table 5.4. Benefit of some RRS Measures of Road Safety Programme in the USA (FHWA,
1993)
Action Type Benefit/Cost %
Improvement of median barriers 1370
Installation of median barriers 850
Improvement of safety barriers 790
Crash Cushions 760
Improvement of bridge barriers 710
76
Table 5.5. Benefit/Cost of some Infrastructural Safety Remedial Measures in Spain

Action Type Benefit/Cost %
Signalling & Traffic Guidance Equipment 460
Road Restraint System 390
Pavement 180
Geometric Characteristics 30
Grade separated interchanges and intersections 20
Effect on accidents
Elvik and Vaa [5] have evaluated the effects of roadside and median safety barriers
on the accident severities. These meta-analyses prove that installing or improving the
safety barriers reduce significantly severe injuries caused by running off the road
(see Table 5.6)
Table 5.6. Effects on accidents of guardrails along the roadside. Percentage change in the
number of accidents [5]
95 % confidence
Accident severity Types of accidents affected Best estimate
interval
New guardrail along embankment
Fatal injury Running-off-the-road -44 -54…-32
Any injury Running-off-the-road -47 -52…-41
Accident rate Running-off-the-road -7 -35…+33
Changing to softer guardrails
Fatal injury Running-off-the-road -41 -66…+2
Any injury Running-off-the-road -32 -42…-20
5.5 Installation Recommendations for Passive Safety Road

Equipment
The following sections describe the procedures for selecting and installing passive
safety road equipment. The main process of treating roadside safety issues is
covered in Chapter 4. The most involved implementation issues are involved with
road restraint systems. Since these must be located on the road to protect traffic from
an obstacle, their positioning is crucial to their performance. Equipment such as
energy absorbing or breakaway poles replaces the hazard directly. Therefore
equipment positioning is usually defined by their purpose (i.e. gantry support or
lighting column).
The criteria for installing road equipment in European countries were reviewed in
RISER Deliverable 5 [6]. The results of this review are summarized in Table 5.7.
Table 5.7 RISER Review of Installation Criteria for Road Equipment
Classification FI FR DE GB NL ES SE
Road class Yes Yes
Traffic flow Yes Yes Yes Yes Yes Yes
Type of obstacle Yes Yes Yes Yes
Distance from edge Yes Yes Yes Yes Yes Yes
Position in curves or straight lines Yes Yes Yes
Design speed Yes Yes Yes
Risk assessment Yes Yes Yes
Estimation of the safety effect Yes
Severity of potential accident Yes
77
The steps for installing road equipment recommended by RISER results are
developed by combining the observed best practices in Europe with the information
gained in the RISER project. The following procedures only apply when the obstacle
or hazard being treated can not be removed from the roadside safety zone or
replaced by a non-hazardous (i.e. passively safe) structure.
The procedures in the following section have a natural logic if they are followed in the
order they are presented. The first step is to identify the hazards that must be
addressed. This will determine which type of road restraint system is necessary,
protection of a point hazard or a distributed hazard. The second step determines the
containment level, or strength of the system. The lateral location of the hazards is
necessary to determine the working width of the system, identified in step 3. Finally in
step 4, the length of the system is determined based on the size of the hazard.
5.5.1 Identification of Hazards

The most important elements of roadside safety design are the identification of the
roadside features that could be struck by vehicles. The focus in the RISER accident
studies has been the protection of the vehicle occupants, but it should also be noted
that third party injuries or damage require special considerations. Railway lines,
sensitive buildings (schools, hospitals, etc), and dangerous goods storage facilities
are often located near a roadway. The results of a vehicle impact with these
structures may lead to casualties beyond those in the vehicle.
The type of object that is being investigated in the roadside design process will
influence the implementation criteria. The main information that must be determined
are:
1. Where is (are) the structure(s) located relative to the roadway?

2. What are the consequences of the impact to the passengers of a vehicle?
3. What are the dimensions of the object?
4. What are the consequences of an impact to the object itself (third party
damages)?
These points must be resolved to fulfil the following procedure for selecting passive
safety road equipment.
Set-back (Lateral) Position of the Hazard

The set-back of obstacles from the roadway is the first issue that must be addressed
for ensuring safe roadside environments. The safety zone for the road section should
have been identified (Chapter 2) as this allows the surveying of roadside obstacles to
be focused to a corridor adjacent to the roadway. All obstacles and their set-backs
should be recorded. Characteristics of the obstacle that influence their risks of
causalities (Chapter 1) in an impact should be recorded as this will influence the
subsequent selection of the containment level. It is important that no further design
solutions are determined until the remaining information for the obstacles is collected.
A common mistake in the roadside safety process is solving a local problem without
investigating potential problems in the vicinity. A small incremental cost to a road
equipment installation may create larger savings in casualty and property damage
costs. One example discussed in Chapter 4 (Figure 5.10) is presented again to
highlight the importance of creating an inventory of all hazards in the vicinity.
78
Figure 5.10. The culvert is protected with a short guardrail but the gantry pole and the old
rigid lighting column are unprotected (Photo, HUT)
5.5.2 Selection of Containment Level

The containment level is an important characteristic of a road restraint system. The
containment level identifies the strength of the system, essentially specifying the
maximum capacity for redirecting a vehicle. As presented in Section 5.3.2, the
containment level is specified by the crash test conditions (vehicle mass, impact
speed, impact angle). Higher containment level produces stronger restraint system.
Background
The main consideration when specifying a containment level is to assess the risk for
a severe impact with the system and a subsequent penetration of the vehicle through
the protective equipment. Thus the containment level must be assigned with
considerations to:
1. Type of vehicles operating on the road (usually the percentage or number of

Heavy Goods Vehicles –HGV observed in annual average daily traffic).
2. Type of roadway (motorway versus rural road).
3. Roadway speeds.
4. Local information affecting risk of accidents (weather, road geometry, etc.).
5. Additional factors of risk (identified above in terms of third party damages).
Once this information has been collected, a containment level selection for the
particular location can be determined. Most national road administrations will have
some general policy on the containment level for their road network [7]. However a
more detailed selection procedure should be available for local installations. A good
example of such a selection process is presented in RPS 2003 [8] for Germany
(Figure 5.11). This flow chart contains all the elements identified above. The process
starts at the left side with the obstacle or hazard of concern. Then the roadway type
is defined by the speed and AADT. Specific conditions leading to higher risk of
accidents are included as well as the influence of heavier vehicles. It is important to
note that there are solutions that do not require the installation of a road restraint
system.
79
Figure 5.11. Selection Criteria for Safety Barrier Containment Level in Germany (RPS 2003)
Another approach to selection of containment levels is applied in Spain. When a

hazard is identified, it first is necessary to determine minimum distance from the
hazard and the roadway edge that justifies the installation of a safety barrier. In the
case of bridge parapets, their installation is always justified. The installation criteria
presented below are fully described in the Spanish national document O.C. 321/95 T.
and P. [9 ].
For Spain, the containment level depends on: carriageway features, roadway
environment features, roadway design speed and AADT[9]. As well as for the layout
of the RRS, the Spanish guidelines take into account a critical distance between the
hazard and carriageway as seen in Table 5.8.
The Spanish guidelines [9] establish the containment levels as a function of the
accident type. Three accident types are defined according to the risk level - normal,
serious and very serious. This guideline was developed before European Standard
EN1317-2 approved 1998. Thus the Spanish criteria were based on the containment
levels existing at that time (L1=N1, L2=N2, M=H2, P=H4b) and the draft version of
European Standard in 1995. More containment levels are found in the current
standard than the draft in 1995, thus Spanish guidelines [9] could be further refined
to specify more containment levels.
80
Table 5.8 Critical Distances between Carriageway and Hazards [9]

Minimum Distances for Given
Slope Gradient of Accident Severity
Road Alignment Shoulder Embankment [m]
[horizontal:vertical] Serious or Very
Normal
Serious
Single Carriageway
Straight, Inside Curves, > 8:1 7.5 4.5
Outside Curves 8:1 to 5:1 9 6
Curve Radius > 1500 m <5:1 12 8
> 8:1 12 10
Outside Curves
8:1 to 5:1 14 12
Curve Radius < 1500 m
<5:1 16 14
Dual Carriageway
Straight, Inside Curves, > 8:1 10 6
Outside Curves 8:1 to 5:1 12 8
Curve Radius > 1500 m <5:1 14 10
The containment levels can be assigned to the accident severity type defined in O.C.
321/95[9], according to the following severities.
• Normal Severity Accidents = N1 - N2

• Serious Accidents = H1 - H2 - H3
• Very Serious Accidents = H4a - H4b
Some (but not all) examples of these accident categories are:
Design speeds> 80 km/h, collisions with trees

and post over 15 cm in diameter, masonry walls,
Normal Severity Accidents
foreslopes steeper than 5:1, backslopes steeper
than 3:1, etc.
Collisions that result on significant debris on the

roadway, collisions causing structural damage to
Serious Accidents buildings or bridge structures, possible
encroachments on neighbouring roads or
opposing traffic lanes, etc.
Proximity to railways, dangerous goods,

Very Serious Accidents proximity to below grade motorways,
intersections at bridges, etc.
The development of a specific, European containment level selection procedure was

not pursued in the RISER project since the information is very sensitive to regional
issues. However, the 5 pieces of information identified above are readily available for
the regional administrations and the information in Chapter 1 provides relevant
information for determining the consequences of striking different obstacles.
Chapter 2 also provides important information for determining the likely impact
configurations (speeds and angles) that determine the strength requirements for
81
different road conditions. It is apparent from the information in Chapter 2 that the
actual impact speeds and angles experienced by road restraint systems are lower
than those used in the performance standards.
5.5.3 Lateral Placement of Road Equipment

The positioning of any road equipment near the roadway must take into consideration
the influence of the structure on surrounding traffic and the maintenance implications
of the structures. Many safety barriers are located very close to the carriageway edge
line (typically around 1 m). Driver simulator studies in RISER [3] included the
influence of the lateral position of guardrails. When an emergency lane is placed
between the safety barrier and the traffic lanes, drivers tended to driver closer to the
edge lines as opposed to conditions without an emergency lane. Thus the lateral
position of the safety barrier can affect the lateral positioning of traffic. Chapter 3
(Recovery Zone) discussed the advantages of a paved shoulder beside the travel
lanes that allowed for vehicle manoeuvres. The recommended width of a paved
shoulder was 1-1.5 m and it is advisable to provide this amount of space between the
safety barrier and the carriageway edge line.
The distance behind a safety barrier (or other passive safety system that deflects) is
important for the proper operation of the system. As defined in Section 5.3.2, the
working width and dynamic deflection are determined from a crash test. These
distances must be used when selecting a safety barrier to ensure that there is
enough free space behind the system. As shown in Figure 5.12, there must be
sufficient space for the safety barrier and vehicle to deflect during the crash without
contacting the hazards placed behind the system.
Figure 5.12 Lateral Distance Behind a Road Restraint System
Information about the deflection characteristics ( W, D) for any road restraint system
is available from the manufacturer.
82
5.5.4 Installation Length of Need

An important requirement for any road restraint system, particularly safety barriers, is
the length of installation. This parameter can be determined from characteristics of
the roadside environment, the expected impact conditions, and the road restraint
system:
1. What is the length of system tested and reported for compliance with
EN1317?
2. What is the length of the obstacle?
3. What are the expected impact angle and impact speed?
4. What is the potential for the vehicle driving behind the barrier?
The length of need should be understood as the length of a road restraint barrier
without the accompanying anchorages or end terminals. The length of need identifies
the barrier section that is expected to be struck by a vehicle leaving the roadway. End
conditions and anchorages can also be struck, but they should not be considered as
part of the primary protection purpose of the barrier.
Minimum Test Length

The length of the safety barrier that is used in the test report should always be the
smallest length of safety barrier to be installed. The evaluation criteria are usually
sensitive to the length of system installation and it is important that the field
performance can be predicted from the test results. The relevance of this length is
also discussed in a following section.
Shielding of Hazardous Zones

The roadside features that must be shielded from traffic will influence the length of
the system installation. Figure 5.13 shows a roadside hazard protected by a safety
barrier. Length "a" is the projected length of the hazard onto the road and is never the
length of a system installation. Length "b" is required to protect a vehicle that may run
off the near side of the road prior to the obstacle. The length of "b" will depend on the
expected exit angle and exit speed. Similarly, "c" represents the length of safety
barrier to protect a vehicle crossing over the oncoming traffic lanes and striking the
hazard. Thus the system length of need is defined by lengths a,b, and c. Lengths
denoted "d" represent the end terminals of the system required for structural strength
as well as protecting the vehicles from impacts with the end of an unprotected safety
barrier.
83
d c a b d
Figure 5.13 Hazard Length and Dimensions for Road Restraint Systems
The principle for dimension b (and c) is illustrated in Figure 5.14. The angle α is
determined from expected exit angles for the road section and is not the angle used
for the crash tests in EN1317. The speed that the vehicle will strike the hazard after
leaving the road is also important to take into consideration since there can be
considerable braking effects in off-road terrain, reducing the length required for b.
These exit angles and possible impact speeds should be determined for local
conditions using Chapter 2 as a reference. The 5 degree value (representing the
median value of vehicle exit angles) is a good reference value for α.
Figure 5.14 Approach Length to Shield Hazards

Referring again to Figure 5.10, one should make sure to consider all obstacles in the
vicinity of an identified hazard to determine if the installation length should be
reconsidered. Figure 5.15 depicts 4 point hazards in the roadside area. If one
considers a radius of influence around each hazard, then one can identify if there are
cases where separate point hazards can be combined. Items c and d should
obviously be treated as one distributed hazard. Cost analyses can be conducted to
see if the two additional crash cushions needed to protect a and b are more
expensive than a safety barrier installation extending from a to d.
84
b
a c d
Figure 5.15 Review of Multiple Hazards
Supplementary Concerns
As listed at the start of Section 5.5.4, the possibility of vehicles moving behind the
system should be considered. Experience from the RISER data collection indicated
that vehicles drove off the road prior to an installation of safety barrier and struck a
hazard behind the barrier. This is depicted in Figure 5.16 where the barrier length x
extending before a hazard is too short. When the terrain is flat behind the safety
barrier, the vehicle may move behind the barrier, striking the hazard. This is
particularly problematic for overpass supports or culverts in motorway medians.
RISER accident analyses [10] found cases where safety barrier installations of 50-
60 m from the hazard were insufficient to protect against this type of collision.
Figure 5.16 Possible Vehicle Motions Behind Safety Barrier
The two alternative solutions for the problem depicted in Figure 5.16 are shown in
Figure 5.17. In one case the barrier is angled away (flared) from the road – indicated
with the red dotted line. This will result in a vehicle impact on the barrier terminal
which is less severe than the impact with the hazard. The other alternative is to
extend the barrier in front of the hazard (dashed black line). This may still allow the
vehicle to move behind the barrier, but the potential braking distance available for the
vehicle allows the resulting impact to be of lower severity. The use of a flared barrier
installation to provide extra shielding of an obstacle may also reduce the required
length of barrier to be installed. The flare rate (expressed in lateral shift of the barrier
for the installation length) should be made in consultation with the test barrier
85
manufacturer. The use of a flared barrier will increase the impact angle of the vehicle
against the barrier so modest flare rates are recommended.
Figure 5.17 Design Solutions
The length g of a barrier needed to protect a vehicle can use the hazard information
from Chapter 1 in combination with off-road braking distances calculated from the
procedures in Chapter 2. An example is presented in Figure 5.18.
220
200
180
160
140
Distance [m]
Desirable distance (m)

120
100
Minimum distance (m)
80
60
40
20
0
70 80 90 100 110 120 130
Exit speed (km/h)
Figure 5.18 Length of Need Prior to a Hazard
The minimum distance curve represents the distance needed when an impact of
50 km/h with a hazard is considered acceptable. The desirable distance represents
the case where the vehicle stops before reaching the hazard. These curves both
assume a braking acceleration of 3 m/s2.
The concept that barrier length must be adjusted for hazards that would be near the
end of a barrier installation has been recognized by some countries already. In
Germany, the standard [8] specifies a minimum length of need based on the tested
length of barrier. However the risk of hitting a hazard by driving up on to the barrier or
driving behind the barrier is known if the hazard is less than 1.5 m from the traffic
face of the barrier. A deviation for the standard barrier length is required according to
this risk. The new lengths of barrier segments surrounding the hazard are given in
Figure 5.19and Table 5.9. This requirement provides sufficient length of barrier to
86
prevent the vehicle coming in contact with the obstacle in close proximity to the
barrier installation. Without this deviation from regular requirements, the terminals of
the safety barrier may be too close to the obstacle and allow the vehicle to impact the
hazard. Examples of these collision types have been identified in the RISER accident
analysis[10] where barrier lengths were too short. Referring to Figure 5.18, the
distances quoted in the German standard are consistent with the suggested barrier
lengths when exit speeds are 100-110 km/h.
L≥ L1
d a d
L2≥ 1/3L1 L2 ≥ 1/3L1
Figure 5.19 German Dimensions for Road Restraint Systems
Table 5.9 Safety Barrier Lengths (L2) Before and After Hazards
Criteria for deviation Lengths for Safety Barriers:
from test length Type of Road Angled from
(Figure 5.19) Parallel to Road
Road
Riding on the barrier Single Carriageway 100 m -
and Hazard 1,5 m
behind front section Dual Carriageway 140 m -
of the barrier
Driving behind the Single Carriageway 80 m 60 m
barrier (if possible) Dual Carriageway 100 m 60 m
The Spanish guidelines also use a look up table for the installed lengths of barriers
preceding the obstacles. As shown in Figure 5.20, the Spanish guidelines include
elements of the two previous approaches. The location and size of the obstacle are
used to determine the length La required before the obstacle. Both parallel and
angled approach lengths for the barrier are permitted.
87
a) Barriers Installed Parallel to the Road
b) Barriers Angled from the Road
Figure 5.20 Spanish Dimensions for Road Restraint Systems
Table 5.10 Barrier Approach Lengths for Spain (Figure 5.20a)

Setback from Carriageway Minimum Distance for La (Figure 5.20)
to Hazard Single Carriageway Dual Carriageway
a <2m 100 140
2 to 4 m 64 84
b 4 to 6 m 72 92
>6m 80 100
Table 5.11 Angled Barrier Approach Lengths for Spain (Figure 5.20´b)
Maximum Setback (b) to Hazard Distance La of the Angled Barrier Section [m]
[m] Single Carriageway Dual Carriageway
<4 36 40
4 to 6 44 52
>6 52 60
Barrier Components
The installation of a safety barrier length of need would have end points as shown in
Figure 5.13 just prior to the sections marked "d". These would be blunt objects and
these points should be treated as hazards (Chapter 1). There are two different
88
methods to protect the end of the length of need section of the barrier. The first
method is to employ a flared barrier section, thus eliminating the risk of an end on
impact to the barrier. This is shown schematically in Figure 5.13. This requires that
the end of the flared section does not present a new hazard to the vehicle. One
approach is to move the end of the flared section outside of the safety zone. When a
flared end terminal is not possible, an energy absorbing barrier end terminal can be
installed (Section 5.2.3). This is an effective solution when the barrier end must be
placed close to traffic. The barrier end terminals have containment and deformation
classes, similar to safety barriers, which should be used to select the end terminal
suitable for the road section.
Often the installation of a safety barrier is adjacent to another type of safety barrier or
bridge parapet that has different characteristics (containment level, structural
configuration, etc.). The transition between different safety barriers systems (Section
5.2.4) should be tested in accordance to EN1317 to ensure there is no degradation in
the crash performance in the transition section.
The selection of barrier components should be done in conjunction with the

equipment manufacturer to ensure compatible structures are selected. The
maintenance personnel should also be involved to identify that any training,
inspection, and part logistics issues are resolved.
Structural Issues for the Length of Need.

As described previously, the length of need is the total length of a longitudinal barrier
needed to shield a hazardous location. However, the installed length must also have
the structural capacity required for the crash conditions. The length of the safety
barrier to be installed must be longer than the length tested to demonstrate
compliance to EN1317. This is because when a vehicle restraint system is tested, its
technical description comprises not only the design of the parts (beams, posts, bolts,
etc.), but also other important details such as the anchoring conditions, end sections,
and the total installed length.
A typical installation length in crash tests is 60 m plus terminals. The impact point is
at approximately one third of the length, i.e. 20 m from the initial terminal. In this
situation, the entire system carries the loads caused by the vehicle impact during the
test. Impact loads are distributed along the barrier elements and transferred to the
ground. However, real world crashes on safety barrier installations do not duplicate
the EN1317 crash test conditions and thus result in different load distributions than
observed in the crash tests. It is commonly expected that for example, when vehicle
mass, speed and impact angle increase from their standard testing values, the
system is overloaded and may not ensure a safe performance. But variations in other
impact parameters such as impact point and end anchorage resistance can have a
strong influence in its safety performance and should be taken into account
Crash analyses were performed for RISER using a metal safety barrier. Simulations
were carried out for a standard EN1317 crash test and simulation where the point of
impact was shifted closer to the terminal. The results indicated that higher tensile
forces were predicted for the second simulation case. When this happens, it is critical
that the end anchorages of the barrier can withstand these conditions and is
dependent on the system design, material resistance, and installation process. If the
end sections fail to constrain the system, then there is a high probability that the
89
barrier deflects more or becomes detached from the ground and is unable to contain
the vehicle. This highlights the need for a barrier installation length that is greater
than the tested length in EN1317. It is preferable to provide an installation length that
reduces the probability of vehicles impacting close to the end sections of the barrier.
Otherwise, a barrier design performing successfully in an EN 1317 Part 2 test for
structural adequacy may be exposed to failure of its anchorages due to impacts near
the terminals. This failure would not be due to the barrier design, but due to an
improper installation. In all cases, it should be assured that the barrier that is selected
to be placed on a roadside includes all the design and installation characteristics that
were featured in the EN 1317 tests.
The issues of short barrier sections and end anchorage failure have been identified
as a possible cause for some cases in which barrier end anchorages were detached
in the analysed accidents (see Figure 5.21). They were cases in which the vehicle
impacted the system near the end anchorage and the barrier system was not able to
contain the vehicle.
Figure 5.21 Failure in end sections
Barrier performance is even more critical when safety barriers that are designed to
include ramped or tied-down terminals are installed with unrestrained end sections.
When this happens, there is an important risk that the structural resistance of the
system is dramatically decreased: barrier posts can experience large deformations
and unable to support the loads transferred by the longitudinal beams. This results in
larger deflections than expected, and eventually, breaking of the barrier. Simulation
results (see Figure 5.22) show how in a barrier with terminals not anchored to the
ground, the loads caused by a vehicle’s impact cause the system’s failure.
90
Figure 5.22 Comparison between the behaviour of safety barriers with (green) and without
(red) end terminal fixation
Soil conditions
Road restraint systems work by transferring impact loads to the terrain on which they
are installed. It can be concrete, asphalt or soil of varying composition and
compaction. Due to its resistance during deformation, a soil has capacity to absorb
some of the impact energy. However, the working mechanism of road safety barriers
comprises a combined, balanced deformation of the barrier elements such as rails,
posts, and the ground.
The performance of road restraint systems to be installed on roads is described by

their behaviour in tests according to EN 1317. It has to be noted that in the crash
tests, the ground conditions are controlled and reported. Consequently, in order to
ensure that safety barriers offer their best performance in the event of a vehicle
impact, the road and roadside soil properties should be defined carefully so that they
can match the characteristics that are required by the barrier for a good passive
safety protection.
5.6 Summary
An overview of passive safety road equipment was presented in this section with a
focus on road restraint systems, particularly safety barriers. The procedures for
selecting other passive road equipment generally follow the same procedure. The
first step is to identify the hazards that must be addressed. This will determine which
type of passive safety road equipment is necessary. The hazard may be a lighting
pole that can be replaced with an energy absorbing column or a rock cutting that
needs to be shielded by a safety barrier.
91
The second step in selecting passive safety road equipment is to determine the
containment level, or strength of the system. Safety barriers and crash cushions are
classified by the size of the largest test vehicle used in the crash testing program.
Energy absorbing poles are classified by the amount of impact energy they absorb in
the crash test. Both of these ratings identify the structural capacity for the system.
The third step for selecting equipment is to identify the amount of space available for
the systems dynamic performance. This is established by the proximity of the
hazards being shielded. The location of the hazards is necessary to determine the
working width of the system (for safety barriers), and the deflection classes for crash
cushions.
The final step for determining the installation requirements of passive safety road
equipment is to identify the length of the system. This is most relevant for safety
barriers and is determined by the size and position of the hazard(s) and the expected
accident configuration for the specific location.
Experience has shown that typical problems associated with road equipment are:
• Insufficient length of systems to shield hazards.

• Installations shielding hazards neglect neighbouring hazards.
• Insufficient free distances behind the system.
• Inappropriate end terminals for barriers.
Any passive safety structure used for protecting roadside hazards must be tested to
European test requirements specified in EN 1317 (road restraint systems) and
EN12767 (passive safety supports). Accident data and causation information
analyzed in the RISER project (Deliverables 3 [10] and 4 [3]) can be used to develop
local policies for the selection and installation of roadside infrastructure.
The selection of road restraint systems should include maintenance and operation
program of the road function (RISER Deliverable 8 [11]). Safety performance or
roadside infrastructure can only occur if the equipment is maintained in good working
order. This requires regular inspections and repairs when necessary. A reliable
source of replacement parts and qualified service personnel is thus needed to keep
all safety equipment within the manufacturers’ specifications.
92
References
1 EN1317 – Road Restrain Systems (Parts 1-6), CEN - European Committee for
Standardization, Central Secretariat: rue de Stassart 36, B-1050 BRUSSELS, ©1998
2 EN12767 Passive safety of support structures for road equipment -Requirements and
test methods, CEN - European Committee for Standardization, Central Secretariat: rue
de Stassart 36, B-1050 BRUSSELS, ©2000
3 D04: Identify envelope of vehicle and driver response prior to collisions, European
GRD2/2001/50088, 2005
4 D01: Accident Databases For Collisions With Roadside Infrastructure, European
GRD2/2001/50088, 2004
5 Elvik.R & Vaa.T. The Handbook of Road Safety Measures. Elsevier 2004. ISBN 0-08-
044091-6
GRD2/2001/50088, 2003
7 Safety Standards for Road Design and Redesign – Appendix 6, SAFESTAR, European
project, 1998
8 Richtlinen für passiven Schutz an StraBen durch Fahrzeug-Rückhaltesysteme, RPS
2003
9 Recommendation on Road Restraint Systems. Orden Circular 321/95, 1995
10 D03: Critical Vehicle and Infrastructure Interactions, European Community R&TD
Project, 5th Framework Programme “Growth”, Project "RISER" GRD2/2001/50088, 2005
11 D08: European Best Practice for Roadside Design: Guidelines for Maintenance and
Operations of Roadside Infrastructure, European Community R&TD Project, 5th
Framework Programme “Growth”, Project "RISER" GRD2/2001/50088, 2003
93
CHAPTER 6: ACCIDENT DATA COLLECTION
6.1 Introduction
A road accident data collection system is required to monitor the performance of
newly designed as well as existing roads. Through data collection, it is possible to
learn more about the road transport system and help identify the need for safety
countermeasures. For re-designed roads it is important to compare the safety levels
of the road before and after the changes to find out if the changes were successful
and quantify the benefit obtained with these modifications. The information in this
chapter gives an overview of the purpose of accident data collection, the information
to collect, and organisations that can collect it.
6.2 Overview of Accident Data Collection

Road safety data analyses are usually separated into two categories. The first (and
most common) monitoring approach involves collecting large numbers of events that
are processed to identify the number of accidents, types of vehicles involved and
other easily tabulated variables. This approach generally intends to cover the whole
population of accidents occurred in one country. These databases are usually carried
out by the Governments and they are referred to as national, intensive, statistic or
police accident databases. This kind of data is essential to understand the magnitude
and the relevance of the different traffic safety problems, although the level of detail
of the information registered within this databases for each accident is not enough to
understand completely all the events contributing to the accident.
The second approach addresses several specific issues that are not present in the
national accident databases as they are less easily compiled. Detailed data collection
programs provide specific technical details and more understanding of the accident
event, i.e. the cause of the accident, information about the occupants of the vehicles
involved in the accident, the vehicles themselves and the environment are registered
within these so called in-depth or extensive accident databases. These information
systems are usually created to address specific problems, thus, depending on the
specific purpose of the database, more information of certain topics will be compiled.
The combination of these two approaches provides answers to "who", "when" and
"where" using statistical databases while the "what" and "why" questions are
addressed from in-depth accident reports.
There are three main processes which should be considered when creating or
analyzing road safety information systems. These processes can lead to
improvements in traffic safety when they are implemented and optimized: accident
prevention eliminates accidents before they happen, severity reduction minimizes the
risk for serious injuries when an accident does occur and post-crash treatment
provides the best possible medical attention to accident victims. Through the
collection of accident data, information to improve our knowledge of these three
processes is gained.
94
6.2.1 Accident Prevention

To prevent accidents there is a need to understand the factors contributing to
accidents. Through this knowledge, it is possible to develop (active) safety systems
and implement countermeasures for the road, vehicle and driver. In an accident
analysis, the understanding of the human factors interaction with the vehicle and
road environment is essential. Identifying the factors that generate accident scenarios
and establishing the methods to address these factors will prevent accidents from
happening.
6.2.2 Severity Reduction

When a collision is unavoidable within the road network, it is necessary that the
passive safety systems for the road, as well as in vehicles, are reducing the risk of
injuries. Many years of in-depth data collection has focused on the vehicles and has
led to the development of passive safety systems in vehicles. However, not as many
studies have been made for the development of safer roadsides. There is still a need
for in-depth studies focusing on the passive safety aspects of the road environment.
Test methods have been developed and are mandatory for road restraint systems. In
tests, vehicle impact conditions such and speed and angle are controlled, and they
are meant to ensure that roadside equipment provides protection in as many accident
conditions as possible. But in reality, actual accident conditions are not generally
known, because few accident scene data include this kind of information (e.g. exit
angle, exit speed, guardrail deformation etc.). Other information about the roadside is
also often lacking e.g. shoulder width, slope gradient, ditch depth, material properties
etc. To reduce the injury risk when an accident occurs, it is important to have real-
world accident data that help to refine, validate and develop test methods. It is also
important to collect accident data to monitor the safety of the road and different road
and roadside designs.
6.2.3 Post-Crash Treatment

When an accident has occurred it is crucial that the rescue services arrive at the
accident scene as quickly as possible to decrease the injury outcome. The most
direct contribution from road infrastructure to this objective is to incorporate automatic
or direct warning systems that notify the event to the emergency systems. Data
collection activities can contribute to improve the post-crash treatment systems by
providing medical personnel with accurate information about injuries, that can help
identify and locate the injuries more precisely. Other important issues to take into
account are, for example if the road design contributed to delays for the rescue
services and ambulance to get to the scene, if the vehicle design contributed to any
delays for rescuing the occupants and to what extent the traffic flow was affected by
the accident. Some of these post-crash factors should be taken into account for the
design of roadside infrastructure.
6.3 RISER results

The different levels of accident data collection was used in RISER to get a broad
base of information about single vehicle accidents and to obtain detail information
about the accident event.
95
6.3.1 RISER Databases

Two databases were developed in the RISER project , one containing a high number
of base level data from the police (referred to as the statistical database) and one
containing a smaller number of in-depth cases (referred to as the detailed database).
From seven countries (Austria, Finland, France, Spain, Sweden, The Netherlands,
United Kingdom) almost 265 000 single vehicle accidents were collected and stored
in the statistical database. From the same countries 211 in-depth cases were
reported in the detailed database.
RISER Statistical database

A comparison of the variables collected by the police from the seven countries was
undertaken [1]. The criteria used when selecting the variables for the statistical
database was that they should be of interest for road infrastructure design and be
available from at least two countries. All of the important variables for road
infrastructure were chosen. Fourteen variables were selected for general information
about the crash and six variables were selected for information about the occupants.
Crash Data Casualty Data

Date Person Class
Time Age
Road Type Gender
Carriageway Type Alcohol Involvement
Road Conditions Injury Severity
Weather Seatbelt Usage
Speed Limit
Light Conditions
No of Vehicle Occupants
Accident Type
Road Alignment
Hit Object
Vehicle Type
Deformation Location
Each country has different ways to code these variables within its own national
accident database. A recoding procedure was undertaken by the countries before the
data was merged into the database, in order to have the same variables with the
same values within the RISER database. It was found that the different organisations
of the police in the different countries make it very difficult to obtain all data from all
countries. Some of the abovementioned variables could only be collected by a few
countries, and in some cases, it was not possible to merge one specific variable from
one country with the same variable from the others, as the information contained in
that specific field was rather different. This makes it a difficult task to perform
complete analyses on the results.
RISER Detailed Database

Before the RISER project, no in-depth databases has been developed containing
thorough information about the road infrastructure involved in accidents. RISER has
96
developed a unique variable list [1] for detailed analyses of accidents involving road
infrastructure especially for run-off-road (ROR) accidents. In the detailed database,
detailed information about the accident, vehicle, occupant, roadway, struck
infrastructure and causation was reported. The database was fed with detailed
accident information that the involved organisations collected from different sources.
As a result, having seven different countries and organisations reporting to the same
database some issues arose. In the RISER detailed database some organisations
had very thorough information about the vehicle (external and internal deformations
etc.) and the occupant injuries but less information about the road environment,
whereas other organisations had much information about the road environment and
the occupant injuries but not so much information about vehicle deformation.
6.4 Conclusions
The current situation regarding accident data collection in European roads shows
that, in order to find countermeasures to improve safety, more thorough information
about the road infrastructure and the causation of the accident is required.
Statistic or base-level accident data are the most commonly available sources of
information. These accident data, usually collected by the police, are often very
general and the type of information differs considerably among European countries.
All variables proposed to describe accidents in RISER statistical database are not
fully available in national databases across Europe. Harmonised data elements
describing European roadside infrastructure allow European-wide statistical studies.
The information gathered in RISER provides a base to further develop this resource.
The data collected by the police is not sufficient for improving roadside infrastructure
safety. Other information sources (i.e. road maintenance and detailed investigations,
etc.) are needed to have a complete overview of the traffic environment. A valuable
resource would be to have a system which collects data of all maintenance
performed on road equipment and store it in a computerised database. When this
data is compared to the police data, unreported accidents can be identified found and
the frequency of accidents can be identified, independently of the injury severity. This
makes it possible to monitor the actual accident rate, the performance of the road
and roadside, and calculate the real costs raised by accidents on the road.
In-depth level accident data provide specific information that helps improve the
knowledge of the accident event, and makes it possible to design or improve safety
measures for identified problems. Detailed databases do not cover populations as
large as the statistical ones, but the information they provide allows describing and
analysing specific safety problems.
To reach a complete description of accident scenes and outcomes, it is important to

not only have collection procedures covering a broad scope of information, but also a
co-ordinated approach is desirable among the different collecting actors. The
following procedures are a proposal to improve the type and quality of data collected
for roadside safety issues.
6.4.1 Police
Road accidents are reported differently in the European countries. Even if the police
are often called to an accident scene, in some countries the police only report injury
97
accidents, where in other countries they also report property damage accidents. It is
preferable to register all of the accidents to get reliable statistics. By considering all
types of accidents, it is possible to improve the usefulness of statistical indicators for
accident risk rates and accident severity. These are used for the safety effectiveness
evaluation of roads and road infrastructure.
When nationwide collection of accidents is not possible, an alternative is to collect

data from all the accidents within a limited region. The geographical area selected
should be representative of the whole country. Using statistical methods, the data
can be scaled from the region to the whole country.
6.4.2 Maintenance and Operation Organisation

In many countries the road maintainers report all repairs that are due to a road
accident. Unfortunately not many countries have a system to store this data. Use of
this data can contribute to finding:
the safety performance of the road

the actual cost for a road
unreported (property damage) accidents
6.4.3 Hospital and Rescue Services

Many injuries, especially to vulnerable road users, are never reported to the police. If
a cyclist or a pedestrian falls on the road/pavement for example (without involvement
of a motor vehicle) the injured person will probable be taken to hospital without being
reported injured to the police. Only some of the countries (e.g. Sweden) have a
system where both the police and the hospital report road accidents to the same
database. Combining police and hospital data makes it possible to:
improve the statistics on vulnerable road users

compare injury outcome (police reported injury severity versus real injury
severity)
find unreported (injury) accidents
In order to improve the relevance of injury databases, it is important that unified

procedures and criteria are set. Victims classified as “injured” should be followed up
over fixed periods of time to report the final actual outcome of injuries, including
eventually death. Currently, different methods are used throughout the European
countries to report injuries.
6.4.4 In-depth Investigations

When special studies are required, in-depth accident investigations must be
performed. To find countermeasures for the road network it is important to include
variables related to the road environment. Most of the existing variable lists are
focusing on vehicles and the injuries to the road user e.g. the STAIRS protocol [2].
Within the RISER project an in-depth study on single vehicle accidents was
performed. From the in-depth study a minimum set of road infrastructure variables
was identified that is essential when investigating single vehicle accidents [1].
98
6.5 Variables
The variable matrix presented below (see Table 6.1) is a proposal and should not be
seen as a complete list. Fundamental variables are assumed to be collected by the
police. Additional variables can be collected by road operators and the hospital and
rescue services. In-depth accident investigations should collect all or part of the
variables listed in Table 6.1, depending on the aim of the data collection.
All forms of accident data collections should include:
date
time
type
place (preferable GPS coordinates)
If the police for example could include more environmental variables to gain better
statistics on road and roadside infrastructure it would enhance national databases
and benefit the whole traffic safety area.
The colour coding for Table 6.1 is as follows:
Police
Road operators
Hospital and rescue services
99
Table 6.1. Variable Matrix for Accident Data Collection

Background factors Pre-crash Crash Post-crash
Driver under the influence of
Age medicine, alcohol, drugs Injury severity Evacuation
Time of accident within the
drivers normal circadian Type of injuries (AIS code if
Gender rhythm possible) Rescue
Human
Driver familiarity to the road

Type of road user system (surroundings) Injury cause Care
Vehicle speed compared to
Placement in vehicle posted speed Impact points in the interior Died at scene
Usage of passive safety Function of passive safety Post-crash times (alarm,
Type of trip equipment systems to/from scene, to hospital)
Driving licence (type, Death because of illness or
possession time) suspicion of suicide Ejection, ejection path
Make Technical failures Collision type Trapping
Tyres (make, dimension, Rescue delayed due to the
Model track depth) Collision speed vehicle
Vehicle movement during
Model year Tachograph data collision Possibility to open doors
Colour Load (weight) Rollover movement Fire
Kerb/gross weight Load contributed in accident Hit object Water submersion
Vehicle
Vehicle characteristics
(driven axles, type of Vehicle movement on road Principal direction of force
gearbox, effect) before collision (PDOF)
Type and position of Rest position (wheels, side,
passive safety systems roof)
Vehicle deformation
(Collision Deformation
Type of active safety Classification CDC if
systems possible)
Load influence on injury
extent
Light condition (daylight, Problem for rescue services
Road location (urban/rural) darkness) Points of collision to get to the accident scene
Traffic effected by the
Type of road (classification) Weather condition Roadside type/outline accident
Distance to medical rescue
Traffic flow Road condition Hit object service
Traffic flow at time of Distance to object in
Posted speed accident roadside Temperature
Road characteristics
(geometry, surface type,
width, camber, Type of road- signs, signal Object protected by road Cost of the damage or
Environment
deformations) system, markings present restraint system repair

Road restraint system (type,
containment level,
deformation, function, effect,
Roadside type Brake/skid marks failure) Repair time
Roadside characteristics
(shoulder width and
material, fore/back slope Road equipment influence
gradients, ditch depth) Exit angle on injury extent
Type of road equipment/ Preventive maintenance
restraint system actions
Road construction or
maintenance zone
Vehicle heading (north,
south, west, east)
The background factors are a description of the condition of the including components. Some of the background factors are the
same for all accident types e.g. age, gender and type of road user etc. Other background factors are more specific depending
on the accident type e.g. vehicle type, road geometry and roadside area etc. The background factors can also be placed under
the different accident phases respectively if they are considered to be contributing factors in the accident. Example; if an
accident occurred in a curve it might not be a contributing factor to the accident. However, if the curve camber was incorrect the
curve can be considered as a pre-crash factor.
100
Reference
1 Accident Databases for Collisions with Roadside Infrastructure, Deliverable 1, European

Community, R&TD-Project, 5th Framework Programme “Growth”, RISER Project
GDR2/2001/50088/RISER/S07.15369, 2004
2 Standardisation of Accident and Injury Registration Systems, Final Report, Transport
RTD Programme of the 4th Framework Programme, STAIRS Project RO-96-SC.204,
1999
101
GLOSSARY
The definitions on road and roadside terms as presented below are developed by the
members in the RISER project. There can be differences to National and European
definitions. The phrasing in the European Best Practice for Road Design is based on
this glossary to be sure that the definitions are used the same way.
Arrester bed
Arrester beds are designed to stop
vehicles that have lost braking ability on
long downgrades. They are parallel
ramps filled with smooth, coarse, free-
draining gravel. They stop the vehicle by
increasing the rolling resistance. Arrester
beds are commonly built on an up grade
to add the benefits of gravity to the
rolling resistance.
Back slope (see ditch)

Slope associated with a ditch, opposite the road edge beyond the ditch bottom
Boulder
A large rounded mass of rock lying on
the surface of the ground or embedded
in the soil in the roadside, normally
detached from its place of origin.
Break-away structure
A design feature which allows a device
such as a sign, luminary, or traffic signal
support to yield or separate upon impact.
The release mechanism may be a slip
plate, plastic hinges, fracture elements,
or a combination of these.
102
Bridge abutment
A bridge abutment is the end of a wall
(bridge, tunnel etc.), and is normally
located in the roadside.
Bridge pier/pillar
A bridge pier is an upright structure,
often a series of columns, which
supports a bridge. It can be located in
the central reserve or in the roadside.
Bridge parapet
A longitudinal barrier whose primary
function is to prevent an errant vehicle
from going over the side of the bridge
structure. It can be either steel or
concrete.
CCTV Masts
A mast on which a Closed Circuit Television camera is mounted for traffic surveill-
ance
Carriageway
The part of the roadway constructed for use by vehicular traffic. Includes the travel
lanes in between the edge line markings.
Central reserve
The portion of a divided roadway separating the travel lanes for traffic in opposite
directions.
103
Clearance
Unobstructed horizontal dimension
between the front side of safety barrier
(closest edge to road) and the traffic
face of the protected object (figure
shows the plan view of road with
definitions of set back and clearance
measurements).
Clear/Safety zone
The total roadside border area, starting at the edge line of the carriageway, available
for safe use by errant vehicles. This area may consist of a shoulder, a recoverable
slope, a non-recoverable slope, and/or a clear run-out area. The desired width is
dependent upon the traffic volumes, speeds and on the roadside geometry.
Contained vehicle
For example, a vehicle which comes in contact with a road restraint system and does
not pass beyond the limits of the safety system.
Containment level
Description of the standard of protection offered to vehicles by a road restraint
system. In other words, the Containment Performance Class Requirement that the
object has been manufactured to (EN 1317).
Crash cushion
Energy absorbing system that prevents
an errant vehicle from impacting fixed
object hazards by gradually decelerating
the vehicle to a safe stop or by
redirecting the vehicle away from the
hazard.
Culvert
A structure to channel a water course. Can be made of concrete, steel or plastic.
104
Culvert end
The end of the channel or conduit,
normally a concrete, steel or plastic
structure (figure illustrates a concrete
culvert end which is hazardous).
Cut slope
Earth embankment created when a road
is excavated through a hill, slopes up
from the roadway.
Design Speed
A speed determined for the design of the physical features of a road that influence
vehicle operation. It is the maximum safe speed that can be maintained over a
specified section.
Distributed Hazards
Also known as 'continuous obstacles', distributed hazards are hazards which extend
along a length of the roadside, such as embankments, slopes, ditches, rock face
cuttings, retaining walls, safety barriers not meeting current standard, forest and
closely spaced trees.
Ditch
Ditches are drainage features that run
parallel to the road. Excavated ditches
are distinguished by a fore slope
(between the road and the ditch bottom)
and a back slope (beyond the ditch
bottom and extending above the ditch
bottom).
105
Divided roadway
Roadway where the traffic is physically divided with a median and/or road restraint
system. Number of travel lanes in each direction is not taken into account. See also
dual carriageway.
Drainage gully
A structure to collect water running off
the roadway.
Drop-off
The vertical thickness of the asphalt
edge.
Dual carriageway
A divided roadway with two or more travel lanes in each direction and traffic is
physically divided with a median and/or road restraint system. See also divided
roadway.
Edge line
Road marking indicating where the carriageway ends and the roadside or median
begins. If a shoulder or emergency lane is present, these are located in the roadside
beyond the edge line.
Embankment
A general term for all sloping roadsides, including cut (upward) slopes and fill
(downward) slopes (see cut and fill slope).
Encroachment
When the vehicle leaves the carriageway and enters the roadside.
106
Energy absorbing structures

Any type of structure which, when
impacted by a vehicle, absorbs energy
to reduce the speed of the vehicle and
the severity of the impact.
Fill slope
Earth embankment created when extra
material is packed to create the road
bed, slopes down from the roadway.
Frangible
A structure readily or easily broken upon impact (see also break-away structure).
Fore slope (see ditch)

Associated with a ditch, the slope beside the roadbed before the ditch bottom
Guardrail
Another name for a post and beam safety barrier.
Hard/Paved shoulder
A hard shoulder is defined as being an asphalt or concrete surface immediately
beyond the carriageway edge line. Shoulder pavement surface and condition as well
as friction properties are intended to be as good as the road surface.
Hard strip
A typically narrow paved strip, located in
the roadside adjacent to the carriageway
edge line. A hard strip provides a
surfaced strip that is abutting the
carriageway.
Hard strip
Soft strip
107
Highway
Roadway is used instead because highway is often thought of as a motorway.
Horizontal alignment
The geometric description of the roadway within the horizontal plane.
Impact Angle
For a longitudinal barrier, it is the angle between a tangent to the face of the barrier
and a tangent to the vehicle’s longitudinal axis at impact. For a crash cushion, it is
the angle between the axis of symmetry of the crash cushion and a tangent to the
vehicle’s longitudinal axis at impact.
Impact attenuators
A roadside (passive safety) device which helps to reduce the severity of a vehicle
impact with a fixed object into a less severe collision. Impact attenuators decelerate a
vehicle both by absorbing energy and by transferring energy to another medium.
Impact attenuators include crash cushions and arrester beds.
Kerb (Curb)
A border or row of joined stones forming
part of a gutter along the edge of a
street. A unit intended to separate
surfaces to provide physical delineation
or containment. Often used as edges of
road islands in intersections.
Length of Need
Total length of a longitudinal barrier needed to shield an area of concern.
Median
See Central reserve
Nearside
A term used when discussing right and left hand traffic infrastructure. The side of the
roadway closest to the vehicle's travelled way (not median).
Non Paved surface

A surface type that is not asphalt or concrete (e.g. grass, gravel, soil...).
Non Paved roadside

A roadside which contains very little or no paved surface immediately beyond the
edge line.
108
Offside
A term used when discussing right and left hand traffic infrastructure. The side of the
roadway closest to opposing traffic or a median.
Overpass
A structure including its approaches
which allows one road to pass above
another road (see also underpass).
Overpass
Retaining wall/abutment
Underpass
Paved shoulder
See hard shoulder.
Pedestrian restraint system

Classified as a group under road restraint system. A system installed to provide
guidance for pedestrians.
Point Hazard
Narrow item in the roadside that could
be struck in a collision for example trees,
bridge piers, lighting poles, utility poles,
sign posts (figure illustrated a rigid utility
pole).
Recovery zone
Zone beside the travel lanes that allows avoidance and recovery manoeuvres for
errant vehicles.
Rebounded vehicle
A vehicle that has struck a road restraint system and then returns to the main
carriageway.
Retaining wall
A wall that is built to resist lateral pressure (especially a wall built to support or
prevent the advance of a mass of earth) (see figure for bridge abutment and
overpass).
109
Road restraint system (RRS)

General name for all vehicle and pedestrian restraint systems used on the road.
Road Equipment
General name for structures related to the operation of the road and located in the
roadside.
Road Furniture
See Road Equipment
Roadside
The area beyond the edge line of the carriageway. The area between a divided
roadway (median or central reserve) may also be considered roadside.
Roadside Barrier
See safety barrier
Roadside hazards
Continuous or punctual, natural or artificial, fixed objects or structures endangering
an errant vehicle leaving its normal path. The risks associated with these hazards
include high decelerations to the vehicle occupants or vehicle rollovers.
Roadway
The paved area of the road including shoulders, for vehicular use.
Rock face cuttings

Rock face cuttings are created for roads
constructed through hard rock
outcroppings or hills.
Rumble strip (Shoulder rumble strips)

A thermoplastic or grooved transverse
marking with slight vertical profile which
is designed to provides audible and
tactile warning by the use of the ribs. It is
normally located between hard
shoulders and nearside travel lanes of
carriageway.
Rumble Strip
Paved Shoulder
110
Safety barrier
The purpose of safety barriers is to provide a shield between hazardous roadside
areas and vehicle traffic. Types of safety barriers include for example steel, concrete
and cable safety barriers.
Safety zone
See Clear zone
Set-back
Lateral distance between the carriage edge line and an object in the roadside (see
diagram for clearance).
Shoulder
The portion of the roadway contiguous with the travel lane, primarily for
accommodation of stopped vehicles, emergency use, lateral support of the
carriageway (see figure for rumble strip). On non motorways the shoulder can be
used by pedestrians and bicycles.
Single carriageway
See undivided roadway
Slope
A general term used for embankments. It can also be used as a measure of the
relative steepness of the terrain expressed as a ratio or percentage. Slopes may be
categorized as negative (fore slopes) or positive (back slopes) and as parallel or
cross slopes in relation to the direction of traffic.
Soft strip
A narrow strip of gravel surface found in the roadside beyond the roadway (normally
beyond a hard strip/shoulder) (see figure for hard strip).
Termination (barrier)
The end treatment for a safety barrier,
also known as a terminal. It can be
energy absorbing structure or designed
to protect the vehicle from going behind
the barrier.
Travel/Traffic lane
The part of the roadway that is travelled on by motor vehicles. A carriageway can
include one or more travel lanes.
111
Underpass
Opposite to overpass, the roadway passing under another roadway (see figure for
overpass).
Undivided roadway
Roadway with no physical separation. Also known as single carriageway.
Vehicle restraint system

Classified as a group under road restraint system. A device used to prevent a vehicle
from striking objects outside of its travelled lane. This includes for example safety
barriers, crash cushions, etc.
Verge (grass)
Grass border in the roadside.
Vertical alignment
The geometric description of the roadway within the vertical plane.
112
APPENDIX A: HUMAN FACTORS CONSIDERATIONS
Introduction____________________________________________________________________ 114
The Driving Task _______________________________________________________________ 114
Driver characteristics ____________________________________________________________ 115
Human factors issues related to roadside infrastructure _______________________________ 118
Literature review interaction roadside elements and driver behaviour ______________________________ 118
RISER Detailed database_________________________________________________________________ 119
Human Factors Case Examples ____________________________________________________________ 119
Detailed Road Scene Analyses ____________________________________________________________ 125
The driving simulator study_______________________________________________________________ 126
Human Factors recommendations relevant for road(side) design and redesign_____________ 127
References _____________________________________________________________________ 129
113
Introduction
It is obvious from the accident statistics that single-vehicle accidents are a significant
component of annual road casualties, according to Collin (2000) 33.8% of all fatalities in
the European Union in 1998 were the result of single-vehicle collisions. Run-off-the-road
(ROR) accidents apparently occur because drivers fail to keep their vehicle on the road.
The objective of good road design is to provide a roadway that enables drivers to
recover safely in case a vehicle unintentionally crosses the carriageway edge-line. The
reasons for unintentional roadside encroachments are various and may include drivers
trying to avoid a vehicle, object or animal in their travel lane, inattentive driving due to
distraction, fatigue, sleep, alcohol or drugs, unexpected effects of weather on pavement
conditions, or travelling too fast through a curve etc. But also roadway design factors,
such as traffic lanes that are too narrow given the road function and design speed, or
substandard curves, inconsistent road design with unexpected transitions that surprise
the driver, may contribute to the occurrence of ROR accidents. To include human
factors it is a good start to look at the road(side) design from the driving task perspective
and to be aware of typical driver characteristics before making the step to human factors
issues related to roadside infrastructure.
The Driving Task

In the literature the task analysis for driving a car is well documented. A frequently used
conceptual model of the driving task consists of three hierarchically ordered levels,
navigation, guidance and control (Allen et al., 1971). Tasks at the navigation level refer
to the activities related to planning and executing a trip from origin to destination. The
need for processing information only occurs occasionally, with intervals ranging from a
few minutes to hours. The guidance level refers to tasks dealing with the interaction with
both environment (roadway, traffic signs, traffic signals) and other road users. Activity is
required rather frequently with intervals of a few seconds to a few minutes. At the control
level the motion of the vehicle is controlled in longitudinal and lateral direction.
Information has to be frequently processed, ranging from intermittent activities every few
seconds to almost continuous control. Alexander and Lunenfeld (1986) visualised the
relationship between the levels by a set of nested triangles (see Figure 1), hierarchically
ordered from a low to a high level with an increasing complexity and from high to low
with an increasing urgency (primacy). For example, a flat tire or being suddenly
confronted with a heavy wind gust will immediately interrupt activities at the navigation
level and put all attention to the control level, since getting lost has less severe
consequences than running off the road.
114
+ -
NAVIGATION
complexity primacy
GUIDANCE
- +
CONTROL
Figure 1 The three hierarchical levels of the driving task according to Alexander and
Lunenfeld (1986).
At each level of the driving task the successive steps of information processing, i.e.
perception, processing and decision-making, and action take place. Moreover, the way a
driver performs these tasks strongly depends on the routine the driver has developed in
task performance. Rasmussen (1985) distinguishes three levels of task performance:
knowledge-based, rule-based, and skill-based. The highest level (knowledge-based)
refers frequently to new situations (for example, finding the best route to a new
destination) or situations that occur frequently in itself, but in which the driver still has
little experience. The choice of behaviour depends on interpretation and deductive
reasoning. When a situation occurs frequently, a rule develops, after some time, how to
deal with that situation and recognising that situation leads to appropriate behaviour
without a ‘need’ to understand exactly what is going on. Skill-based tasks are conducted
automatically, incoming information automatically results in the right behaviour without
any cognitive control. Theeuwes (1993) introduced a nice three-dimensional
representation of the driving task as is given in Figure 2.
Figure 2 The driving task in three dimensions (Theeuwes, 1993).
Driver characteristics
It is obvious, in the road design, to include those driver characteristics that are relevant
for the driving task such as for speed choice or lateral positioning. For example, if it were
established that the 95th percentile of the distributions of standard deviations of lateral
position is 0.25 m at design speed, it is easy to calculate the minimum lane width that
would make an involuntary lane departure unlikely to occur (Janssen & van der Horst,
2000). But a complicating factor may be that drivers may display compensatory
115
behaviour in case the resulting lane width is actually so comfortably to them that they
start driving less carefully.
In the so-called funnel model, various human factors aspects are ordered from
background factors of road users that determine how functions are performed resulting
in certain road user behaviour, which in turn influences the total system performance,
(see Figure 3). Demographic factors include individual differences such as gender, age,
nationality, driving experience, impairment, physical, mental, visual capabilities,
personality, etc. Temporary differences include fatigue, alcohol, drugs, motivation,
attention, emotional state, etc. Road(side) design should take into account the way in
which road users interact with the dynamic, and not only with the static environment
(Van der Horst & Hagenzieker, 2002). Concepts such as expectancy and consistency
should be central. In particular, transitions are inherently difficult (for example, from
straight road to bend, from one road category to another with a different function, etc.).
Three human factors aspects appear to be crucial in road safety:
situation awareness
– including perception of relevant elements in the current situation,
comprehension of the current situation, and projection of the future situation
workload
– too much is bad and resulting in stress; too little is bad because of resulting
in a low arousal level of the driver
attention
– the stage preceding the actual processing by the senses.
As an example of the latter factor, in the USA recently, a large-scale study was
completed, in which 100 car drivers were observed in great detail for a one-year period
in their vehicle during driving to get more insight in the pre-crash phase of collisions, the
so-called 100-car naturalistic study (Neale, Dingus, Klauer, Sudweeks & Goodman,
2005). In this study, 82 crashes were registered and 761 near-crashes (defined as a
conflict situation requiring a rapid, severe evasive manoeuvre to avoid a crash).
116
Background factors
demographic factors
permanent factors
temporary factors
motives
Functions
attention
perception
cognition
motor
User performance
driving behaviour
System performance
safety
throughput
comfort
Figure 3 Ordering of (categories of) characteristics in ‘funnel’ model (Janssen & van der
Horst, 2000)
Figure 4 gives an example of a categorisation of events in which the driver was attentive
or inattentive. Driver inattention included secondary task engagement, fatigue, driving-
related inattention to the forward roadway and non-specific eye-glance away from the
forward roadway. These categories were extracted manually by data reductionists from
video. It appeared that a majority of events (78% of the crashes and 65% of the near
crashes) have driver inattention as a contributing factor. The sources of inattention that
generally contributed to the highest percentages of events were wireless devices,
internal distractions and passenger-related secondary tasks.
90
80
70
Percent of Events
60
50 Inattentive
40 Attentive
30
20
10
0
Crash Near Crash
Severity
Figure 4 Percentage of crashes and near-crashes with inattention/distraction as a contributing

factor (Neale, et al., 2005).
It is obvious that also other aspects of the driver such as his intentions, attitudes,
emotions and subjective norms play a role in driver behaviour. One possible
representation of this is given in Figure 5.
117
Figure 5 A driver behaviour model according to van der Horst (1998).
Also in the road traffic, road users, inevitably, will make errors and human errors are
causing most of the accidents. Reason (1990) distinguishes slips and lapses (the plan is
adequate but there is minor failure in execution), errors (misapplication of a rule or
failure to apply a correct rule), and violations (deliberate deviations from rules). It is
obvious that all type of errors may apply for roadside accidents, and as indicated in the
introduction the roadside should be made forgiving for road users that do not properly
conduct their task of speed choice and lateral positioning and end up aside of the road.
The first step, of course, should be that the road and traffic situation do not surprise the
driver and meets his expectancies as much as possible and drivers do not run off the
road. But in case they do, try to reduce the consequences both for themselves as for
other road users.
Human factors issues related to roadside infrastructure

This section gives an overview of a literature review on the interaction between roadside
elements and driver behaviour, of the accident causation issues identified in the RISER
detailed database and a selection of some human factors case examples. Moreover, it
summarises the results of a detailed human factors road scene analysis of ten selected
accident sites, and details the results of a driving simulator study conducted at TNO to
investigate the influence of different roadside features including trees, barriers, signs,
and emergency lanes on speed choice and vehicle lateral placement.
Literature review interaction roadside elements and driver behaviour

In the context of RISER a literature review was conducted to identify effects that
roadside elements have on the lane keeping task (the guidance concept of road and
roadside infrastructure directing driver’s path), and to learn why drivers sometimes fail in
performing this task (Janssen, de Ridder & Brouwer, 2004). Moreover, speed, or rather,
speeding, plays a significant role in the occurrence of ROR accidents. In many instances
drivers lose control of their vehicle because the speed chosen is not appropriate in a
given situation. In general it can be said that if a driver can increase speed without
increasing perceived risk difficulty (or perceived risk) he, or she, will do so. Safety
interventions that make the driving task easier (such as straightening out a curve) may
be consumed by increases in speed. It also explains why it is important to inform a
driver about the road environment and its task difficulty for the driver to be able to adapt
118
driving behaviour accordingly (referring to ‘no surprises’ and design roads and road
environments that meet drivers’ expectancies, Self-Explaining Roads). At higher speeds
drivers tend to show a more selective (narrow) visual search pattern, as they do in case
of high workload. A special case is that of so-called behavioural adaptation after
roadside hazards have been treated. It deserves specific mention because this is often
raised as an issue of concern when treatment is being considered. For example, we
may take away the risk of colliding with trees only to observe that drivers now move
closer to the edge of the road and thereby increase their risk of leaving the lane after all.
A Finnish study by Kallberg (1993) indicated that reflector posts on narrow, curvy, and
hilly roads can significantly increase driving speeds and consequently increase the
number of accidents in darkness. Reflector posts may increase visual guidance and help
the driver see the road alignment ahead, but may have an adverse effect on safety due
to increases in driving speeds.
Out of a total of 144 papers reviewed, 61 relevant papers were selected that dealt with
the driver-roadside interaction in one way or another. There are not that many studies
that have actually looked into driver behaviour, or that – at least – present some ideas
about the role of driving behaviour in the chain of events leading to a ROR accident.
Most countermeasures that are presently under discussion in the literature are directly
infrastructure-based. It is surprising that only once or twice the potential of advanced in-
vehicle devices, such as Lane Departure Warning systems, is mentioned as a remedy to
ROR accidents. It should be noted that the relevance of driver behavioural studies is as
high to these systems as it is to the more classical approaches, as these supports can
only be designed adequately if it is known what behaviour they should deal with (i.e.,
prevent or correct).
RISER Detailed database

A review of over 200 detailed cases was conducted. The retrospective nature of this
review meant that all the relevant details for RISER may not have been collected at the
time of original collection. In a general sense it appears that the lateral positioning and
speed of the vehicle were two of the primary factors leading to the reviewed crashes.
This is observed in the vehicle motions prior to the crash where over 50 % of the cases
involved some sort of lateral motion of the vehicle. Following this; speed, fatigue, and
alcohol were risk factors that were attributed to the driver in several cases. Effective
means of controlling the vehicle position in the lane is difficult to determine from the
available accident data. Controlling driver fatigue and speed is also difficult to achieve
using the static roadside elements. For more details, the reader is referred to RISER
Deliverable D04 (de Ridder, Thomson & van der Horst, 2005).
Human Factors Case Examples

This section outlines ten cases examples from the RISER detailed database which show
how the layout of a road can possibly lead to the onset of a road side collision occurring.
These cases were chosen because they were considered to have more relevant human
factors issues related to the highway than other cases in the database.
119
Example 1
The non-motorway road consisted of two lanes in
each direction, separated by a grass central
reservation with some shrubs. At the accident
location, the two lanes in the car' s direction of travel
are being filtered down to one lane. White painted
lines reduce the number of lanes in order to prevent
vehicles from overtaking at intersections. At the time
of the accident, it was daylight, it was raining and the
road was wet. For an unknown reason, the driver of
the car braked too hard and the vehicle ran onto the
nearside verge. The driver steered back onto the carriageway, but went too far and had
to steer towards the nearside again. The car slid off the road to the nearside, hit an
embankment and rolled over. The driver, who was belted, suffered slight injuries.
Human factors issues related to the highway - The driver was reported as being
tired, so concentration levels may have been low. However, the fact that 2 lanes were
merging into one just before the accident location may have been an issue, particularly if
there was inadequate signage to warn of the lane merging. Also, it appears from the
photographs that nearside edge of the road is very uneven, which may have made it
more difficult for the driver to regain control of the vehicle once it left the road.
Example 2
The accident occurred at night on a single carriageway road, which had no street
lighting. It was raining and the road surface was wet. The car was travelling at 90 km/h
through a left hand bend, when the driver thought that they saw a vehicle coming
towards them. The driver steered right and braked to avoid it. The car was actually
parked on the grass verge on the right hand side of the road. When the driver realised
this, they drove off the right hand side of the road and lost control of the car. The vehicle
struck a fence and then a telegraph pole situated just after the grass verge. The driver
was not wearing a seatbelt and was not injured, despite impacting the windscreen. The
passenger was belted and was slightly injured.
Human factors issues related to the highway - It appears from the accident
information that the main reason for the vehicle leaving the road was the driver being
confused about the ' actions'of the parked car. The presence of the vehicle at the point
where the road bended to the left, plus the lack of road lighting and lack of visual cues
120
on the road (e.g. reflective signs or road markings), all would have contributed to the
confusion of the driver and resulted in the steering and braking manoeuvres carried out
by the driver. If the layout of the road had been more self-explaining, the driver might
have realised that avoidance manoeuvres were not necessary and the run-off could
have been avoided.
Example 3
The accident occurred during daylight on a single
carriageway road. The weather and road surface
were dry. The road had been recently resurfaced
and there was loose gravel. There was a temporary
speed limit of 70 km/h. The novice driver panicked
in the traffic situation (thought there was not enough
room to pass an oncoming car), braked and skidded
on the gravel. The vehicle hit a raised section on the
nearside then the driver steered hard to the left. The
car rotated 180° and then rolled over in a ditch
(drainage gully) on offside. The road was straight at site of accident but there were
bends on approach (left bend). The occupants were not injured.
Human factors issues related to the highway - From the photographs, it appears that
the road was being redeveloped at the time of the accident. The newly resurfaced road
was covered in loose gravel, which was probably reducing the visibility of the centre line
markings and therefore, there were very few visual clues of the road ahead to assist the
driver. Therefore, the lack of visual clues led to the driver thinking the road was too
narrow to pass the oncoming car.
Example 4
The accident happened on the exit slip road of a
motorway during daylight hours. The weather and
road surface were dry. The driver of the vehicle
was driving too fast for the road (100 km/h, the
recommended speed was 60 km/h), and the car
left the road to the offside on the slip road. The car
fell into the embankment and rolled over. Both
occupants suffered minor injuries.
Human factors issues related to the highway -

Although the main reason for the accident occurring was the vehicle' s inappropriately
high speed, it is possible that the layout of the road caused the driver to think the bend
of the road was less severe than it actually was. Also, there appears to be no visual
clues to warn drivers of the approaching bend in the road, such as chevron signs, either
on the approach or at the start of the bend.
121
Example 5
The accident occurred on an entry slip road of a non-motorway dual carriageway. The
collision occurred at dawn, it was raining and the road was wet. The car ran off the road
to avoid an impact with a truck that was entering the road from a side road. This
avoidance manoeuvre resulted in the car impacting a lamp post. The car driver was not
wearing a seat belt and was slightly injured.
Human factors issues related to the highway

- The truck did not obey a sign that established
that vehicles from right must give preference to
the other vehicles. It is possible that the truck
driver misinterpreted the layout of the road. From
looking at the photographs, there is only one
small 'give way'sign at the road junction, which
may have been missed by the truck driver, who
would have been at a more elevated position in
the truck cab. Also, the road markings at the
junction location are worn away, so may have also been missed by the truck driver.
Therefore, the poor signage and markings, in addition to the weather conditions, would
have made it less obvious to the driver of the truck that they were approaching a
junction where they would have to give way to oncoming traffic.
Example 6
The accident occurred on a non-motorway dual carriageway road. It was daylight and
the weather and road conditions were dry. The car driver had to steer right suddenly to
avoid collision with a truck entering the road from offside. The car left the road to the
nearside, hit a barrier termination and travelled along the top of the barrier (rotated 180o)
122
for 10.8 m until it came to a stop on the barrier. The driver, who was belted, suffered
slight injuries.
Human factors issues related to the highway - The layout of the road, which allows
vehicles to enter the carriageway from the opposite side of the road through a gap in the
central safety barrier, does not allow much room for larger vehicles, such as trucks, to
negotiate the turn into the central refuge lane before entering the main carriageway, as
shown in the diagram. This may have given the impression to the car driver that the
truck was moving into the same lane they were travelling in. Also, it may not have been
obvious to the driver of the truck that they had to give way to traffic already on the main
carriageway, as there are very few signs at the junction to instruct drivers.
Example 7
The accident occurred on a motorway. It was
dark but the road was lit by street lighting. The
weather and road conditions were dry. The car
driver steered back onto the main motorway from
the exit slip road at the last minute, over-steered,
struck the central reservation barrier and
rebounded back into the main carriageway. The
after impact distance was 15 m. The occupants
of the vehicle were all belted and the front seat
passenger suffered slight injuries (all others were
not injured).
Human factors issues related to the highway - The human factors issues in this
accident are not clear, as it is not known why the driver of the vehicle attempted to re-
join the main carriageway at the last minute. However, it may have been a result of poor
signage or a misleading road layout that led to the driver possibly taking the wrong exit
off the motorway. But there is no information or photographs about the approach to the
accident to know whether the signage or road layout was adequate or not.
Example 8
The accident occurred on a motorway. It was
dark but the road was lit by street lighting.
The weather and road conditions were dry at
the time of the accident. The car impacts a
safety barrier end while trying to enter the exit
slip road. The occupants of the vehicle were
all belted and the rear seat passenger
suffered slight injuries (all others were not
injured).
Human factors issues related to the highway - Similar to Example 7, the human
factors issues in this accident are not clear. From the photographs, the condition of the
road and the road markings appear poor. Therefore, there would be little to help guide
the driver along the correct drive line , especially at night. As with Example 7, there are
123
no photographs of the approach to the site of the accident, so it is not possible to

determine the level of visibility of the road leading to the impact site or to ascertain how
adequate the signage was.
Example 9 (left hand traffic)
Path of van
Path of car
Path of car
The accident occurred on a single carriageway road (100 km/h speed limit). It was
daylight, the weather was dry, but the road surface was icy. The driver of the car
swerves to avoid the van which pulls out of a side road. It leaves the road to the
nearside and collides with a fence followed by a telegraph pole. The driver, who was
belted, suffered slight injuries.
Human factors issues related to the highway - On the approach to the accident site,
the view of the road ahead is restricted by the uphill gradient of the road (not shown in
photographs). Therefore, the driver of the car would not have seen the van pull out from
the side road until it was quite close. Also, the driver of the van would not have seen the
oncoming car until it was quite close, but by then, the van driver would have already
started to pull out. On the approach to the junction, the only signs to warn the car driver
of the approaching road junction were at the top of the incline.
Example 10 (left hand traffic)
The accident occurred on a single carriageway
road. It was daylight, it was raining and there
was oil/diesel on the road. The car approaches
the roundabout junction, skids and leaves the
carriageway ahead, colliding with a multiple
posted chevron sign which is mounted on the
roundabout. A post from the chevron sign
penetrates the windscreen. Both the driver and
the front seat passenger were belted and
suffered slight injuries.
Human factors issues related to the highway - On approach to the roundabout, there
is a bend to the left just before arriving at the roundabout, which cannot be seen easily
by the driver on the approach. The rain may also have obscured the driver' s view of this
even more. There are also no advance warnings to the driver to slow down before the
bend. National speed limit (100 km/h) signs are present right up to the junction itself. So
124
there are very few visual cues to instruct the driver to slow down before the start of the
bend.
Summary:
The ten case examples outlined in this section have a very limited amount of information
about the possible reasons why the vehicle ran off the road, including the roadway
issues that may have contributed to the initial run-off. However, from what is known
about these cases, it is possible that all ten cases have involved at least one of three
types of issues related to the layout of the road:
Road markings (markings missing, poor quality or faded)

Signage (signs missing, location of signs)
Road geometry (bends, inclination)
For the majority of cases in the RISER detailed database, there is not enough
information prior to the initial run-off to know exactly why the vehicles had a run-off in the
first place. Due to a lack of available information with respect to road layout it is difficult
to draw any conclusions whether the run-offs were directly associated to the road layout
or not.
One third of the cases (69 of 211) in the database had no causation information at all. In
19 % of the cases, the driver exceeded the legal alcohol limit, in 15 % the driver was
fatigued, in 9 % the driver was not driving appropriately for the road surface conditions,
in 6 % the driver was not concentrating because of internal distractions, 5 % of the
vehicles had serious defects and 4 % of the drivers were attempting to avoid other
vehicles/objects on the road. Some cases have more than one of these causation
factors.
By comparing levels of alcohol impairment to those in the statistical database, it can be

seen that the proportion of alcohol impairment cases in the detailed database is higher
(19 %) than it is in the statistical database (9 %). However, the sample of cases in the
detailed database is not totally random (e.g. high bias towards fatal cases), which is one
possible reason for the larger proportion in the detailed database.
The data show the importance of retrieving more thorough accident causation data
concerning the road and roadside to be able to use accident data in the road(side)
design process.
Detailed Road Scene Analyses

Detailed Road Scene Analyses (RSA) were conducted by two human factors experts at
a selection of ten accident sites in the Netherlands, to provide more in-depth information
in relation to the interaction of the driver with the road design and roadside
infrastructure. For five out of the ten sites, the RSA did not reveal a specific reason
based on the road layout why the crash with a tree or roadside infrastructure had
125
occurred. Two locations showed a mismatch between the road design and road
category and the speed limit posted (the road layout invites the road user to drive faster
than the speed limit). At one location a multitude of curves might be appealing to the
frequent user (especially motor cyclists) and invite to race. Two accidents occurred in a
work zone area with confusing road markings. At two locations it appeared that, once a
road user had made an error and found him/herself behind a hazard protection device
such as a poorly visible barrier or an interrupted guardrail, the collision with the hazard
was almost inevitable.
The driving simulator study

Given the limited data in the database of accidents collected in the RISER project, a
driving simulator study was developed to investigate two of the main factors leading to
crashes – lateral position and speed of the vehicle. The driving simulator study observed
the influence of different roadside features including trees, barriers, signs, and
emergency lanes on speed choice and vehicle lateral placement. Different constructions
of the man made objects were investigated as well as the lateral positioning of the
different objects.
For the conditions on the motorway, where different types of safety barriers were
introduced, it can be concluded that there was no effect on speed or on lateral position
dependent on the type of safety barrier that was introduced. Overall no guidance of any
type was found. The presence or absence of an emergency lane did have an influence
on driving behaviour. When an emergency lane was absent, drivers tended to choose a
position on the road that was further away from the right side marking. How far away the
driver chose to drive from the right side marking when an emergency lane was absent
depended on the type of safety barrier. When the standard 0.75 m high Dutch concrete
barrier was introduced (this is the lowest safety barrier of all safety barriers introduced)
and when the 0.9 m high concrete barrier was introduced (the second lowest of all
safety barriers) in combination with an emergency lane, drivers drove closer to the edge
markings than in other conditions. However, when the emergency lane was absent, they
moved further away from the right side edge markings in these conditions than in other
conditions, compared to the situation with an emergency lane. It could be the case that
familiarity in the condition with an emergency lane plays a role.
Another conclusion based on these results and the results on the rural road is that when
an obstacle is introduced, drivers tend to temporarily move away from that obstacle and
choose a position in the lane further away from the edge line. This effect was also found
in earlier field studies when introducing a sound barrier or entering a tunnel entrance
(Blaauw, & van der Horst, 1982; Bakker, & van der Horst, 1985; de Vos, Hoekstra,
Pieterse, 1998; Martens & Kaptein, 1997, 1998). The effect was not found in the curve
sections on the motorway.
In the conditions with the curves on the motorway, the results revealed that when drivers
see a particular treatment for the second time, they tend to be less impressed by it and,
as in this case, drive faster through the curve and closer towards the right edge marking
126
than the first time. Also, when a new treatment (like the stripes) is introduced, speed
goes down. However, based on the previous finding, this effect might be temporary.
On the rural road, drivers did change their speed when trees or safety barriers were
introduced, dependent on how close they were positioned to the road. When they were
4.5 m away or more, no effects on speed were found, whereas when they were
positioned 2 m away or less, speeds were reduced.
For the lateral position a different picture occurred. When trees were introduced in
combination with a safety barrier, drivers tended to choose a position away from the
safety barrier and trees. However, when trees were introduced solely, no effects on the
lateral position chosen were found. In fact, the position chosen in these two conditions
was the same as in the condition with the trees at 30 m away from the roadside. One
has to question therefore whether drivers understand the risks of trees alongside the
road and whether that might be an underlying reason of the many incidents and
accidents with trees. Trees along the road do not seem to influence driving behaviour
that much and are not considered to be a serious hazard by road users. This is in line
with the real life studies reported in France, (CETE 2000 & 2002) where the removal of
trees closer than 2 m from the carriageway edge line had no influence on travel speeds.
One important result from the French study was that after the trees have been removed,
the average number of injury accidents decreased by 50 %, and the number of fatal
accidents reduced by 75 %. It is important to realize that the current guidelines
recommend that no obstacles should be in the safety zone which is at least 4.5 m from
the carriageway edge line for 80 km/h roads in most European countries surveyed, and,
if so, should therefore be removed or protected.
Human Factors recommendations relevant for road(side) design and

redesign
Based upon the analyses as presented in the previous sections, we make the following
recommendations relevant for road(side) design and redesign:
Make the road and roadside design such that they meet drivers’ expectancies
by applying the concept of Self-Explaining Roads and road categorisation
based upon the road function.
Inform the driver about transitions in road function and road layout in a timely
manner. Do not surprise the driver.
Drivers make errors, either at the strategic or manoeuvring level (from the
human factors case examples it appears that often the initial event is trying to
avoid other road users in the interaction between road users), or are driving too
fast relative to the situation with consequences at the control level, or there is a
lack of an early action (inattention, distraction, fatigue, etc.). With respect to the
latter, one can consider installing shoulder rumble strips or edge line profile
markings to warn the driver that he/she is starting to exceed the lane. These
measures appear to reduce the number of lane departures (van der Horst, de
127
Vos & Folles, 1997), but also may result in behavioural adaptation effects such
as increased driving speeds during wet-road conditions.
Trees along the road do not seem to influence driving behaviour that much and
are apparently not considered as a serious hazard by vehicle drivers. With
respect to roadside safety, therefore, trees require special attention.
Make the roadside forgivable. Since drivers make errors and inadvertently
exceed their lane, the likelihood of a collision or a rollover should be minimised
and the severity of crashes that occur reduced.
Be aware of behavioural adaptation effects. Make a shoulder such that people
are able to correct their manoeuvre without inviting them to drive on it without
problems (for example by providing discomfort without losing vehicle control), or
provide sufficient visual guidance but not more than that to avoid increases in
speed (Finnish example of reflector posts (Kallberg, 1993)).
128
References
Alexander, G.J. & Lunenfeld, H. (1986). Driver Expectancy in Highway Design
and Traffic Operations. (Report FHWA-TO-86-1). Washington, D.C.: U.S.
Department of Transportation, Federal Highway Administration.
Allen, T.M., Lunenfeld, H. & Alexander, G.J. (1971). Driver Information Needs.
Highway Research Record, no. 366, 102-115.
Bakker, P.J. & Horst, A.R.A. van der (1985). The behaviour of car drivers near
noise barriers on motorways. (Report IZF 1985 C-20). Soesterberg, The
Netherlands: TNO Institute for Perception.
Blaauw, G.J. & Horst, A.R.A. van der, (1982). Lateral positioning behaviour of car
drivers near tunnel walls – Final report. (Report IZF 1982 C-30).
Soesterberg : TNO Institute for Perception.
CETE (2000). Évaluation sur les vitesses de la suppression des plantations
d’alignement – cas de la RD1314 en Seine-Maritime. CETE Centre d’Études
techniques de l’équipement Normandie Centre, Division exploitation sécurité
gestion des infrastructures.
CETE (2002). Evaluation sur les vitesses de la suppression des plantations
d’alignement – Cas de la RD27 dans le Calvados. CETE Centre d’Études
techniques de l’équipement Normandie Centre, Division exploitation sécurité
gestion des infrastructures.
Collin, C. (2000). Statistics in Focus- transport. (Catalogue number CA-NZ-00-
003-EN-I). Eurostat.
Hogema, J.H., Hoekstra, W. & Stel, I. (2004). Online vehicle model coupling
Driving Simulator - ADVANCE. (TNO Memo TNO-DV3 2004-M 059).
Soesterberg: TNO Human Factors.
Horst, A.R.A. van der (1998). Factors influencing drivers’ speed behaviour and
adaptation. (TNO Report TM-98-D006). Soesterberg: TNO Human Factors.
Horst, A.R.A. van der & Hagenzieker, M.P. (eds.) (2002). Report of the European
Workshop on International Human Factors Guidelines for Road Systems.
(TNO-report TM-02-D009). Soesterberg: TNO Human Factors.
Horst, A.R.A. van der, Vos, A.P. de & Folles, E. (1997). Lane-keeping behaviour
at profiled road markings on motorways: A before-and-after study. In:
Proceedings of the ICTCT 97 Conference, 5 Safety Statistics and Evaluation
Studies, November 5-7, 1997, Lund. Lund: Lund University of Technology.
Janssen, W.H., Ridders, S.N. de & Brouwer, R.F.T. (2004). Summary of driver
behaviour and driver interactions with roadside infrastructure. (RISER
Deliverable D02). Göteborg, Chalmers University of Technology.
Janssen, W.H. & Horst, A.R.A. van der (2000). Contributions to an international
research agenda on ‘Human Factors in Highway Design’. (TNO-Report TM-
00-C042). Soesterberg: TNO Human Factors.
Kallberg, V. P. (1993). Reflector Posts – Signs of Danger? Transportation
Research Record 1403. Washington, D.C.: National Academy Press, 57-66.
129
Martens, M.H. & Kaptein N.A. (1997). Effects of tunnel design characteristics on
driving behaviour and traffic safety: a literature review. (Report TM -97-B05).
Soesterberg, The Netherlands: TNO Human Factors.
Martens, M.H. & Kaptein N.A., (1998). Effects of an emergency lane and exits
and entries in tunnels on driving behaviour: driving simulator studies. (Report
TM- 98-C058). Soesterberg, The Netherlands: TNO Human Factors.
Neale, V.L., Dingus, T.A., Klauer, S.G., Sudweeks, J. & Goodman, M. (2005). An
overview of the 100-car naturalistic study and findings. In: Proceedings 19th
Enhanced Safety Vehicles Conference, Washington, D.C., June 2005. Paper
number 05-0400.
Rasmussen, J. (1985). Trends in human reliability analysis. Ergonomics 28(8),
1185-1195.
Reason, J. (1990). Human Error. Cambridge University Press.
Ridder, S.N. de & Martens M.H. (2004). Roadside Infrastructure for Safer
European Roads : Road-scene analyses of ten accident sites. (TNO Report
TM-04-D009). Soesterberg: TNO Human Factors.
RISER (2004). Accident Databases for Collisions with Roadside Infrastructure.
(RISER Deliverable D01). Göteborg, Chalmers University of Technology.
Theeuwes, J. (1993). Visual Attention and Driving Behaviour. In: Proceedings of
the International Seminar Human Factors in Road Traffic. 5-6 April 1993,
Braga, Portugal: Universidade do Minho, 103-123.
Vos, A.P. de, Hoekstra. W., Pieterse, M.T.J. (1998). Effects of lane width and
physical separation on driver comfort. (Report TM-98-D003).Soesterberg,
The Netherlands: TNO Human Factors.
130
APPENDIX B: FORTHCOMING UK STANDARDS1
Passive Safety
The current Advice Note (TA89/04) states that any sign posts larger than 90 mm
diameter require protection or the use of passively safe posts. Passive signs can be
used in the verge, at roundabouts, at repeat collision sites, and at nosings or splitters
to avoid the need for ramped ends of safety fence.
Passive posts should not be used in central reserves on gantry signs. The new
Advice Note, TA 89/05 (release late November 2005) will allow lighting columns and
traffic signal poles to be placed on passively safe posts, and provide some relaxation
and further guidance on the use and positioning of passively safe posts.
IRRRS and Risk Assessment

The Interim Requirement for Road Restraint Systems (IRRRS) was introduced in
2002 and updated in 2004. The standard requires the use of protection where the
object is located within 4.5 m of the carriageway, objects include trees with a
diameter of more than 300 mm. Other issues of importance include a review of
embankment height and slope, the closure of gaps in safety fence less than 50 m
long, protection of lighting columns, provision of safety fence termination systems
including crash cushions, and passively safe posts (see above).
From 2007 the Highways Agency is seeking to replace its current proscriptive siting
criteria for impact protection, with a new risk assessment based criteria. This aims to:
make risks and risk assessment more transparent in the Standard,

enable designers to carry out site specific risk based design,
minimize the need for approval of Departures from Standard by the
Overseeing Organisation.
It is intended that the risk will be presented in terms of accidents per year/100km and
determined by the probability of the vehicle leaving the road, the probability of the
errant vehicle reaching the hazard, and the probability of injury if the hazard is
reached.
If the hazard is reached, what is the likely consequence (without barrier) on the
vehicle occupants and on third parties? If the risk is considered too high, can the
object be moved, become passively safe, or risk reduced by using a vehicle restraint
system? Designers will be provided with a database system to help them calculate
the risks in each scenario.
1
Trunk roads and motorways with a speed limit of 50 mph or more
131
APPENDIX C: THE ROLLOVER ACCIDENT FOR
PASSENGER CARS
The rollover accident

A rollover, which is a very complex and individual event, is divided into 4 phases
(Gugler et al, 2003):
1 Pre-roll phase
2 Point of no return
3 First phase of roll
4 Rolling phase
Figure 1 The Four Phases in a Rollover
Pre-roll phase: The pre-roll phase is when the vehicle is coming into a
destabilized driving mode till the “point of no return” where a rollover cannot
be avoided. In this phase active safety systems can be used to stabilize the
vehicle and avoid exceeding the “point of no return”. The closer the vehicle
comes to the “point of no return” the vehicle passive safety systems can also
be activated in this phase. If possible, a severity estimation of the impending
rollover should be done.
Point of no return: Is a short time interval when the rollover cannot be
avoided and passive safety systems have to be activated to reduce the risk
of injuries to the occupants.
First phase of roll: The first phase of roll starts from the “point of no return”
and covers approximately the first 90 degrees of the roll angle. It ends with
the first impact of the vehicle structure to the ground. The vehicle can always
be in contact with the ground or loose the contact (flying phase).
Rolling phase: The rolling phase is the phase from the end of the first
phase of roll until the vehicle's rest position. The most important parameter
for this phase is the number of rolls.
Classification of Rollovers
The scenarios, presented below, are based on a modification of the NHTSA-
Classification (Asic, 2002) for rollover in Europe.
132
Fall-Over - Rolls downward due to negative

slope
Flip-Over - Rolls downward due to positive slope
Trip-over - when the lateral motion of the vehicle

is suddenly slowed or stopped inducing a
rollover. The opposing force may be produced by
a curb, pot-holes, or pavement dug into vehicle
wheels.
Flip-over - when vehicle is rotated along its

longitudinal axis by a ramp-like object such as a
turned down guardrail or the back slope of a
ditch. The vehicle may be in yaw when it comes
in contact with a ramp-like object.
Bounce-over - When a vehicle rebounds off a

fixed object and overturns as a consequence.
The rollover must occur in close proximity to the
object from which it is deflected.
Turn-over - when centrifugal forces from a sharp

turn or vehicle rotation are resisted by normal
surface friction (most common for vehicle with
higher COG.). The surface includes pavement
surface and gravel, grass and dirt. There is no
furrowing or gouging at the point of impact.
Note that if rotation and/or surface friction causes
a trip, then the rollover is classified as a turn-
over.
Fall-over - When the surface on which the

vehicle is traversing slopes downward in the
direction of movement of the vehicle cog. such
that the cog. becomes outboard of its wheels.
133
Climb-over – when the vehicle climbs up and

over a fixed object (e.g. guardrail, barrier) that is
high enough to lift the vehicle completely off the
ground. The vehicle must roll in the opposite side
from which it approached the object.
Collision with another vehicle - When an

impact with another vehicle causes the rollover.
The rollover must be the immediate result of the
impact between the vehicles. For example, this
could occur at an intersection where a vehicle is
struck in the side and the momentum of the
struck vehicle results in a rollover.
End-over-end - When a vehicle rolls primarily

around its lateral axis.
The Categories of Rollovers on the Road and in the Roadside Area (Case Study
Based Analysis) (Gugler et al, 2003)
Kinematic parameters like angle and roll rate were investigated in addition to tire
forces and velocity for the accidents collected in the Rollover project. The following
categories has shown to be relevant in Europe see Figure 2 for distribution
1 Impact induced rollovers

1.1 v > 30 km/h
1.2 v < 30 km/h
2 Ramp-like object induced rollovers
3 Skidding & Yawing
3.1 Trip induced rollover
3.2 Turning and rollover
4 Others
The different scenarios focus on causes for the inducement for the roll. It has to be
stated that also the roll trajectory is influenced by the road infrastructure. The car can
hit any objects off the road which intrudes and/or deforms the passenger
compartment or lead to high acceleration influencing the occupant movement.
Different analysis showed that the contact of the occupant to a vehicle part during its
intrusion is critical for severe injuries (e.g. head is in contact with roof when the roof
is intruded by an impact of the roof with the ground).
Because the rollover event is a high complex and individual event the focus is set to
the causes of rollover to avoid the roll (before the “point of no return”) rather than the
rolling phase.
134
Figure 2 Distribution of Rollover Scenarios (Gugler et. al, 2003)
Category 1.1 “Impact induced rollovers with v > 30 km/h”

The change in vehicle velocity due to an impact is caused by:
an obstacle which is part of the road like curbs, drain cover etc.
an obstacle which is part of the landscape like rocks, trees etc.
another vehicle.
Regarding injuries to occupants the inducing impact itself is assumed to be more

severe than the rollover caused by the impact.
Category 1.2 “Impact induced rollovers with v < 30 km/h”

The causes for this scenario are the same as for category 1.1. After analysing the
accidents it can be assumed that the severe event is the rolling phase itself and not
the inducing impact ( v < 30 km/h). Therefore each roadside infrastructure
influencing the intrusion or deceleration (“hard contacts”) is critical as well as the
condition of the vehicle itself (roof strength).
Category 2 “Ramp-like object induced rollovers”

The following mechanisms can cause an accident in this category:
The vehicle is launched of an object acting like a ramp (e.g. safety barrier
termination with ramped end, concrete element, cut slopes) or be induced by
the overrun of another vehicle (hood of the vehicle is acting like a ramp).
These are so called active ramps because both the geometry and the
velocity of the vehicle are inducing the rollover.
The vehicle is driving on an embankment and the rollover is induced by the
high gradient (often downward). This is a so called passive ramp because
the rollover is induced by the geometry of the vehicle (wheel base, centre of
gravity) and the gravity.
135
Category 3.1 “Trip induced rollover”

In this category, which is the most common type, the rollover is induced by tripping of
the vehicle. The tripping is caused by increasing lateral forces due to increasing of
friction which can be caused by the tyres grabbing into soft soil (e.g. embankments)
or contact of the rim with the road surface (defect tyre).
Category 3.2 “Vehicle dynamic induced rollover”

When a vehicle comes into unstable driving mode caused by the driving manoeuvre
the rollover is induced by the vehicle dynamics. The causes for this type or rollover
are:
Driving manoeuvre like the elk test leads to oscillations of the vehicle. The
driving manoeuvre can be caused by the course of the road in combination
with speeding (e.g. driving to fast into a round about) or an evasive driving
manoeuvre (e.g. evasive a obstacle on the road).
The suspension of the vehicle does not damp the oscillations.
Category 4 “Others”
This category summarises all events which cannot be related to the abovementioned
categories. These are events like falling down by crashing through a guardrail or a
jumping down an embankment etc. A big contributor to this category is the
geometrical properties of the road.
References
Gugler, J. et al: Rollover scenarios for Europe, Task 2.4 report, Graz University of
Technology, European Community – R&TD-Project – 5th Framework-Programme
“Growth” – Project “Rollover” G3RD-CT-2002-00802; 2003
Asic, S.: Field Conditions Definitions - NASS-CDS Classification, Presentation,
Delphi, 2002, European Community – R&TD-Project – 5th Framework-
Programme “Growth” – Project “Rollover” G3RD-CT-2002-00802; 2002
136
APPENDIX D: VEHICLE AND ROAD RESTRAINT SYSTEM
CRASH TEST INFORMATION
FRONT
European New Since 1997 frontal impact
Car against deformable barrier
Assessment (40 % overlap) with 64
Program kph
(EuroNCAP)
SIDE
European New
Car since 1999 side impact,
Assessment right angle with 50 kph
Program
(EuroNCAP)
POLE
European New
Sideways at 29 kph into a
Car
rigid pole (diameter
Assessment
254 mm).
Program
(EuroNCAP)
7 vehicle classes, impact

EN 1317
velocity between 65 and
Road restraint
110 kph; impact angle
systems
between 8 and 20°
EN 12767
Passive safety impact velocity 35, 50 ,70
of support and 100 kph; Definition of
structures for 3 energy absorbing level
road equipment
137
Frontal impact test

The frontal impact test simulates a collision with another structure that overlaps 40%
of the car’s bonnet on the driver’s side. The structure is made out of a deformable
mesh with a stiffness representative of a car bonnet, and impacts the vehicle at
40mph. The purpose of this test is to simulate a partially offset collision between the
tested vehicle and an oncoming vehicle and this type of collision accounts for a large
percentage of all car crashes on the road. It is therefore important that a car’s ability
to withstand this type of impact is tested.
Side impact test

The side impact test represents another important simulation of an accident that
commonly occurs in the real world. A deformable crash barrier, approximately the
width of the bonnet of a car, is impacted into the side of the tested vehicle.
For side and frontal test the possible impact velocity at the end of the clear zone is
acceptable under the presumption that the hit object dissipates energy, otherwise the
loads for the occupants can not be compared to the crash tests.
Pole test
A pole test is a representation of what would happen if the vehicle impacted with a
thin pole (e.g. a road sign or tree). The main difference between the pole test and the
other tests is that due to the pole being much thinner, the energy of the collision
concentrated in a much smaller area and the crush depth measured on the vehicle is
greater than in other tests. Without appropriate airbags and car designs, the type of
accident that this represents can result in a severe injury.
138

Riser 2006 D6

Uploaded by

Copyright:

Available Formats

Riser 2006 D6

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Riser 2006 D6

Uploaded by

Copyright:

Available Formats

D06: European Best Practice for Roadside Design:

Guidelines for Roadside Infrastructure on New and

Project ACRONYM: RISER

TITLE: Roadside Infrastructure for Safer European Roads

PROJECT START DATE: 1/01/2003 DURATION: 36 months

Original Date of issue of this report: 28/02/2006

Submitted by: Chalmers University of Technology

Project funded by the

Organisation Acronym Country

3.4.6 Discussion ......................................................................................................................................45

Figure 5.18 Length of Need Prior to a Hazard.........................................................86

Principles for Safe Road Networks

Figure I.1. Design and Operation Knowledge Base

The Self Explaining Road and Forgiving Roadside Philosophies

A second design philosophy that must be highlighted is the "Forgiving Roadside".

Figure I.2. Self Explaining Road with Forgiving Roadside

Framework for Roadside Infrastructure Design

Safety Zone Recovery Zone

Chapter 4 Hazard Yes

Hazard Yes Final

Figure I.3. Procedure for Roadside Infrastructure Design

Chapter 4 provides practical information important for the technical (engineering)

Chapter 5 lists information outlining the performance features of different types of

CHAPTER 1: DEFINITION AND IDENTIFICATION OF

Due to the poor energy-absorbing qualities of many roadside objects, an impact

1.2 Definitions of Roadside and Median Objects (Point and

Distributed objects, also known as ' continuous obstacles'

1.2.1 Point Objects

Trees and Tree Stumps

arguably be distributed hazards, especially in the cases where there are

Poles and Posts

Rocks and Boulders

Bridge and Overpass Structures

Culverts and Culvert Ends

Other Point Hazards

Hazardous Safety Barrier Ends

1.2.2 Distributed Objects

Slopes and Embankments

Rock and Concrete Hazards

Old Design/Poorly Installed Safety Barriers

Other Distributed Hazards

1.3 Overview of Research

The Highway Perspective of Side Impacts [1]

Strategic Highway Safety Plan [5]

1.4 RISER Analysis

Table 1.1. Distribution of Struck Infrastructure Types in RISER Statistical Database

1.4.2 RISER Detailed Database

Table 1.2. Distribution of Struck Infrastructure Types in RISER Detailed Database

Tree Impacts Only

Post/Pole Impacts Only

Impacts with Rock/Concrete Objects Only

Impacts with Sloping Ground Only (Embankments, Ditches etc.)

All Impacts with Culverts

All Impacts with Non-Safety Fences

Impacts with Safety Barriers Only

For example, in a number of cases:

However, it is inevitable that barriers will often be impacted in accidents because of

1.4.3 Current European Guidelines