Innovative Infrastructures - Toward Human Urbanism 1

Bridge barriers in relation to the crash testing standards

Wasim QADIR Wasim Qadir, born 1978, received his

Sr. Structural Engineer post graduate degree in structural
Design Management (Roads) engineering from the V.T. University,
ASHGHAL, P.W.A., Qatar Karnataka, India. He has experience
wqadir@ashghal.gov,qa of over ten years in project planning
and structural design of various
prestigious projects like Qatar
Expressway, Mecca Metro, Dubai
Metro.

Summary
This document aims to review and compare the information on the function, level of service, design and
specific types of bridge parapets used under the two most widely recognized and accepted crash-testing
standards viz. American NCHRP Report 350 and European Union EN 1317.This document will help to
promote discussion and raise issues on the use of traffic parapets associated with bridges or similar
structure in developing countries around the world where such standards and testing facilities are not
presently in place . Reference to research documents and codes is made within the text.

1. Introduction
Parapet is a safety barrier installed on the edge of a bridge or on a retaining wall or similar structure
where there is a vertical drop and which may include additional protection and restraint for pedestrians
and other road users.
Basically two types of barriers exist on bridges, these being:
• traffic barriers; and
• pedestrian barriers.
Pedestrian barriers are typically the post/baluster type. Traffic barriers however come in a variety of
configurations to suit particular circumstances, including flexible, semi-rigid, rigid beam and post, and
rigid concrete barriers. Concrete parapets are preferred in some places as they present a cleaner fascia
and have improved noise attenuation characteristics. Post/baluster type traffic barriers improve the
visibility options where such measures are endorsed.

2. Pedestrian Barriers

Pedestrian barriers on road bridges and footbridges are required to safeguard pedestrians and/or cyclists
and are not a vehicle restraint system; i.e. they are not designed to resist the penetration of an errant
vehicle. These are designed for applications where there is a likely hood of a fall from height.
 2 18TH CONGRESS OF IABSE, SEOUL, 2012

The AASHTO LRFD section 13 specifies geometric

requirements in clause 8.1 and design loads in Clause
8.2. The minimum height for pedestrian railing is
1070 mm from top of the walkway. Where cyclists
may use the walkway, minimum railing height shall
be 1070 mm in accordance with the AASHTO Guide
for the development of bicycle facilities. However,
locations where high speed high angle impact with
the railing are more likely to occur or in locations
with site specific safety concerns, railing above
minimum should be considered.
The minimum height of the pedestrian parapet in Figure 1 : Typical Pedestrian Barrier
accordance with Table 1 of BS 7818 is specified
below. On cycleway bridges or accommodation bridges frequently used by equestrians, the height above
the adjoining paved surface must be increased to 1800mm.
Minimum heights of pedestrian restraint systems as per Table 1 of BS 7818
Parapets
Use Guard Rail
Not Over Railway Over Railway
Pedestrian 1000 1150 1500
Cyclist 1000 1400 1500
Equestrian 1000 1800 1800

3. Traffic Barriers

Traffic barriers are protective devices that are designed to reduce the severity of an accident when an
errant vehicle leaves the roadway. At the same time, these parapets are also an obstacle which means
that a vehicle hitting a parapet can result in injury to the occupants as well as vehicle damage.

Traffic barriers are normally installed on

bridges to provide level of containment that
limit penetration of errant vehicles, and to
protect others in the vicinity by redirecting
vehicles on to a path as close as possible to the
line of the parapet to protect pedestrians or
other motorists in adjacent areas.

Containment capacity refers to the ability of the

parapet to contain the impact of an errant
vehicle and to deflect it away in a controlled Figure 2 : Typical Concrete Traffic Barrier
manner. A containment level is normally
expressed in terms of the angle of incidence, the weight and speed of the vehicle which represents the
magnitude of the impact that the parapet is designed to sustain. A strong parapet designed to a higher
containment level may stop a heavy vehicle in the desired manner, but may cause considerable damage
to a smaller vehicle. Occupants of a small vehicle may also be subject to severe acceleration force and
injury. Conversely, a parapet with a lower containment level designed for light vehicles would not
 Innovative Infrastructures - Toward Human Urbanism 3

perform as equally well for larger vehicles which, in case of a severe impact, may even penetrate the
parapet. It is therefore important to seek a balance between risk and the level of containment.

Previous bridge design codes required bridge barriers to be designed for a given static load, at key
locations on the railings and satisfy certain geometric requirements. Latest revisions of Bridge Design
Code still provides static design loads, however these loads are only provided to allow the design of
prototype barriers for crash testing, for the design of deck cantilevers supporting the barriers and
designing the bolted anchorage connections while emphasis is given to the use of barriers that have been
crash tested to a particular standard.

3.1 Function

The aim of traffic barriers is to improve site safety. This aim equates to minimal damage and injury to
impacting vehicles, vehicle occupants, others on the bridge and traffic, property, roadways, railroads or
waterways below the bridge, consistent with the assessed level of risk at the bridge site. It may be noted
that a parapet designed to a high containment level may stop a heavy vehicle in the desired manner, but
may cause considerable damage to a small vehicle and subject the occupants to severe acceleration force
and injury. Conversely, a parapet designed to a lower containment level for light vehicles would not
perform equally well for larger vehicles that may penetrate the parapet.

Assessment of the level of risk is the critical component in determining the type of traffic barriers that
are suitable for a particular bridge site. Some factors which contribute to the assessment of risk include:
• total traffic volume;
• types and proportions of vehicles in the traffic population;
• identifying the type of vehicle to be contained;
• road alignment;
• bridge width;
• general site conditions; and
• consequences of not containing the identified vehicle within the roadway.

A good example of the selection procedures used to determine the performance level required of a
railing at a particular bridge site is the AASHTO Guide Specification for Bridge Railings 1989.

3.2 Vehicle Impact Test Criteria

The evaluation of safety barriers with a higher containment level require carrying out two different tests:
a test using a heavy vehicle for estimating the maximum level of containment, and a test using a car in
order to verify that the satisfactory attainment of the maximum level is also compatible with the safety
of the light vehicle.

3.2.1 General requirements of Vehicle Impact Test

Test vehicle and the safety barrier needs to satisfy the following for any particular containment level:
• The center of the vehicle shall not cross the centerline of the deformed system.
• The vehicle shall remain upright during and after impact although the moderate rolling,
pitching and yawing are acceptable depending on the specific standards.
 4 18TH CONGRESS OF IABSE, SEOUL, 2012

• The vehicle shall leave the safety barrier after impact within the permissible exit box.
• The barrier shall contain and redirect the vehicle without complete breakage of the principal
longitudinal elements of the system.
• No major part of the barrier shall become totally detached or present an undue hazard to the
other traffic or pedestrians.
• Elements of the safety barriers shall not penetrate the passenger compartment of the vehicle.
• Deformations or intrusions into the passenger compartment causing injuries are not
permitted.
• Ground anchorages and fixings shall perform according to the design of the safety barrier.

3.2.2 Vehicle Occupant Impact Severity Assessment

The vehicle occupant impact severity assessment measure the same impact phenomenon from different
aspects:
• Acceleration severity index(ASI) and is defined by the following formula:
2 2 2
⎛ ax ⎞ ⎛ ay ⎞ ⎛ az ⎞
AIS = ⎜ ⎟ + ⎜⎜ ⎟⎟ + ⎜ ⎟  where
⎝ aˆx ⎠ ⎝ aˆy ⎠ ⎝ aˆz ⎠ ,
a   =  component of the acceleration along x, y and z. Values averaged over a moving time
interval of 50ms
â   =  Limit values for the components of the acceleration. The limit accelerations are
interpreted as the values below which passenger risk is very small (light injuries, if any).

• Theoretical head impact velocity (THIV) is a magnitude of the velocity of the component’s
head striking against a surface within the interior of vehicle.

• Post-impact head deceleration (PHD) is a maximum value of the post-impact head resultant
acceleration.
These parameters are values of a point in the vehicle centre of gravity computed from the light vehicle
test. The coordinate axes of the vehicle body used in measured data during the impact are x
(longitudinal), y (transversal) and z (vertical).
Any one of them can be determinant depending on the test structure. The calculation of the PHD starts
after the calculation of THIV ends; when the head of the occupant hits the notional impact surface inside
the vehicle. ASI is calculated throughout the collision and is averaged over the moving time interval like
PHD, but the size of the time interval is longer. Thus it does not react as sensitivity to changes in the
components of acceleration along the x- and y-axes as PHD, but it includes also the component along z-
axis.

3.2.3 Impact Severity Index (IS)

The “Impact Severity” index is used in both NCHRP 350 and in the European Union code EN1317
(although it is described as “Containment Level”) for comparisons between individual test within a
testing regime to account for tolerances in the various variables i.e. the mass of the test vehicles, its
speed and angle of impact. The Impact Severity formula measures the available kinetic energy just prior
to vehicle impact.
 Innovative Infrastructures - Toward Human Urbanism 5

The Impact Severity is defined by the following formula:

1
M (V sin θ ) ( joules )
2
IS =
2
where: M = Mass of vehicle (kg)
V = Velocity (m/s)
θ = Angle of Impact

The IS is a valuable tool as it can function as an equivalence measure and be used to evaluate tested
barriers against other crash test standards. This approach is valid where the nature of the test vehicles
(for the relevant categories) and other acceptance criteria (i.e. occupant risk and post impact vehicular
behaviour) are similar. With all other parameters being equal, the resulting kinetic energy or IS value of
the test vehicle can be used to rank or derive the equivalency of a tested barrier under an alternate crash
standard. The IS cannot be used in-lieu of a crash test and therefore cannot be used to increase the rating
of a tested barrier. The IS does not examine the performance of a bridge rail directly, rather the
performance of a bridge rail is inferred from the highest test level that was passed. Hence, while a
particular barrier may be able to satisfy the requirements of a higher test level, the IS cannot be used to
demonstrate this potential nor can it be used to predict the post impact behaviour of the vehicle. This
point is particularly relevant to the containment level of a tested barrier and its ability to prevent a
colliding vehicle from tripping over the barrier.

3.3 Summary of International Test Levels

Different international standards have slightly different ‘test levels’ defined. Acceptance criteria for the
tests are also different, but are normally expressed in terms of structural adequacy, vehicle occupant risk,
vehicle damage and exit path requirements
The American NCHRP Report 350 and the European Union EN 1317 are currently the two most widely
recognized and accepted crash-testing standards. The American Federal Highway Administration
(FHWA) issues letters of approval for bridge rails that comply with NCHRP 350 and maintains a
website of all approved bridge rails. At this stage, a similar central repository of Comité European de
Normalisation (CEN) approved bridge rails does not appear to exist. It is understood that CEN stipulates
that compliant systems have to be crash tested to the requirements of EN 1317 but does not issue
certificates of compliance. Without such a system, it would be difficult for a road authority to verify
CEN compliance of any proposed barrier system.
Owing to limited research activities in Hong Kong, and the lack of testing facilities, the highway agency
relied heavily on international standards, in particular the British Standard BS6779 (Highway Parapets
for Bridges and Other Structures), based on which the SDM was developed. After the European
Standard – EN1317 (Road Restraint Systems) superseded the BS6779, all the new barriers need to
comply equivalent with a database of compliant EN 1317 bridge barriers.
Australia similar to the Hong Kong does not have their indigenous testing board and have developed
traffic barrier specifications which are based on crash testing requirements developed by the United
States Transportation Research Board (National Cooperative Highway Research Programme (NCHRP)
Report 350). Table “A” provided below document a comparison between Australian Standard and TRB
Report 350 and the nearest equivalent to a database of compliant EN 1317 bridge barriers as a “Study of
Overseas tested Bridge Barriers’.
 Table A – Comparison of Overseas tested Bridge Barriers in terms of Impact Severity Values
NCHRP 350 EN1317 AS5100
6

Mass Speed Angle Mass Speed Angle Mass Speed Angle

TL Type IS (kJ) Level Type IS (kJ) PL Type IS (kJ)
(kg) (km/h) (Degrees) (kg) (km/h) (Degrees) (kg) (km/h) (Degrees)
820 C 50 20 9.3
1
2000 P 50 25 34.5
N1 1500 C 80 20 43.3
820 C 70 20 18.1 820 C 70 20 18.1
2 Low
2000 P 70 25 67.5 2000 P 70 25 67.5
900 C 100 20 40.6
N2
1500 C 110 20 81.9
900 C 100 20 40.6
H1
10000 R 70 15 126.6
820 C 100 20 37.0
3
2000 P 100 25 137.8
820 C 100 20 37.0 820 C 100 20 37.0
4 Regular
8000 S 80 15 132.3 8000 S 80 15 132.3
900 C 100 20 40.6
H2
13000 B 70 20 287.5
900 C 100 20 40.6
H3
16000 R 80 20 462.1
900 C 100 20 40.6
H4a
30000 A 65 20 572.0
820 C 100 20 37.0 820 C 100 20 37.0
5 Medium
36000 V 80 15 595.4 36000 V 80 15 595.4
820 C 100 20 37.0 Special 820 C 100 20 37.0
6
36000 T 80 15 595.4 Cat.1 36000 T 80 15 595.4
18TH CONGRESS OF IABSE, SEOUL, 2012

900 C 100 20 40.6 Special 820 C To be specified by

H4b
38000 A 65 20 724.6 Cat.2 44000 V/T road authority
Special 820 C To be specified by
Cat.3 To be specified road authority
Ledged of “Vehicle Type”
A=Articulated Heavy Good Vehicle R=Rigid Heavy Good Vehicle
B=Bus S=Single Unit Van Truck
C=Small car T=Tanker Type semi-Trailer
P=Four Wheel Drive, Utility V=Van Type Semi Trailer
The data is arranged in the order of increasing Test Levels within a standard and increasing “Impact Severity” across the standards
 Innovative Infrastructures - Toward Human Urbanism 7

4. Performance Levels
Each bridge need to meet a particular performance level in terms of traffic barrier and a traffic
barrier suitable to that performance level should be chosen and specified. In evaluating the type of
barrier to be used, it is critical to note that the Road Authority must carefully investigate the degree
of risk, clearly identify the types and mix of vehicles that are to be contained and determine the
Performance Level required at the bridge site conditions with proper road restrain risk assessment.
The selection should be based on road type, grade, curvature, deck height and under-structure.

NCHRP 350 is a performance based standard. This basically means that the standard is focused on
the performance of a product. The major difference between the American and European standard is
the vehicle mass and impact angle. For the most common containment level which is based on the
highest vehicle population ie the car, NCHRP 350 uses low mass vehicle with a greater impact
angle and therefore should be considered where there is high percentage of cars.

As per Australian standards:

NCHRP Report 350 Test Level 2 barriers are to be used in situations of low risk and thus defined
as “Low performance level” traffic barrier which are required to contain a level of impact typical of
a 2.0 tonne utility vehicle (light vehicles) at 70 km/h and an impact angle of 25 degrees.
NCHRP Report 350 Test Level 4 barriers are defined as “Regular performance level” which is
required to contain a level of impact typical of a 8.0 tonne truck at 80 km/h and an impact angle of
15 degrees. This performance level barrier is provided for the effective containment of general
traffic on all roads; which means that the regular barrier will be applicable and appropriate to the
majority of bridge sites.
NCHRP Report 350 Test Level 5 barriers are defined as “Medium performance level” which is
required to contain a level of impact typical of a 36.0 tonne articulated van at 80 km/h and an
impact angle of 15 degrees. This performance level barrier is provided for site-specific, medium to
high risk situations for the effective containment of medium to high mass vehicles and buses on all
roads.” The emphasis is on successful containment of the vehicle on the bridge to mitigate against
injury to third parties at risk off or below the bridge. It is important to distinguish the requirement of
identification of the type of vehicle to be contained as a critical factor. The cost of crash-testing 36
tonne trucks is large and subsequently there are fewer crash tested barrier systems available for
selection. Concrete barriers have proved capable of containing substantial forces. The “F” Type
concrete barriers are based on the AASHTO profiles and are considered to be compliant with
NCHRP 350. The 810 mm tall barrier is rated as TL-4, while the taller 1060mm high barrier has a
TL-5 rating. The New Jersey Concrete Barriers are also NCHRP 350 compliant, with the 810 mm
(32”) tall barrier rated at TL-4 while the taller 1060 mm (42”) barrier is TL-5 compliant.

NCHRP Report 350 Test Level 6 barriers is defined as one of the “Special performance level”
which is required to contain a level of impact typical of a 36.0 tonne tanker as compared to the
articulated truck which has higher centre of gravity and impart higher energy at impact compared to
an articulated vehicle of the same weight. The second Special category identifies a 44 t articulated
van at a test speed of 100 km/h which is almost double the energy or severity of impact of the first
Special category. For the American NCHRP 350 compliant barriers, there is currently no barrier
rated higher than TL-6. Potentially, a compliant EN 1317 H4b barrier may be suitable as a Special
category 2 barrier because its IS value is about 30% higher than the TL-6. The suitability of the H4b
barrier will depend on the actual requirements of the Authority and these will include the height and
location of the centre of gravity of the test vehicle, the collision speed and the angle of impact. The
German Maxi-rail is a proprietary barrier system, which is a compliant H4b barrier. This barrier
8 18TH CONGRESS OF IABSE, SEOUL, 2012

may be suitable as a Special category 2 barrier where the centre of gravity of the T44 vehicle is less
than 1.9 m and the impact speed is limited to 80 km/h at a maximum impact angle of 15 degrees.
The third Special category is to be specified by the Authority.

5. BARRIER ACCEPTANCE 

For bridge traffic barriers to be acceptable it must be crash tested and comply with the acceptance
criteria as set out in National Cooperative Highway search programme (NCHRP) Report 350 or to
other appropriate Standards as determined by the relevant Road Authority. However it is not
practical or economically feasible to test every barrier and components, in which case certified
barriers provide an ease and ensure the required safety even for the small projects. Barriers should
also be acceptable if they can be evaluated as being both geometrically and structurally equal to a
compliant crash-tested system. Alternatively, approval may be given based on the evaluation of
performance of an existing barrier to the Authority’s requirements.

6. ACKNOWLEDMENTS
The author gratefully acknowledges the support from Design Department of “ASHGHAL - Qatar
Public Works Authority” and the colleagues from “Aurecon” in particular Eric De Fleuriot who is
Technical Director and Head of Bridges for Aurecon (Middle East).

7. REFERENCES

[1] AASHTO Guide Specifications for Bridge Railings, 1989.

[2] National Cooperative Highway Research Program (NCHRP), “NCHRP Report 350 -
Recommended Procedures for the Safety Performance Evaluation of Highway Features”, 1993.
[3] NCHRP Report 230 - Recommended Procedures for the Safety Performance Evaluation of
Highway Appurtenances, 1981.
[4] NCHRP Report 289 - Performance of Longitudinal Traffic Barriers, 1987.
[5] American Association of State Highway and Transportation Officials (AASHTO) Highway
Subcommittee on Bridges and Structures, Technical Committee (T-7) for Guardrail and Bridge
Rail, “Bridge Railing Design and Testing” - Discussion Paper”, 14 May 1996.
[6] NCHRP Report 239 - Multiple-Service-Level Highway Bridge Railing Selection Procedures,
1981.
[7] European Committee for Standardization (CEN) - “EN 1317-1, Road Restraint Systems - Parts
1-4, 2010.
[8] BS 7818:1995 Specification for pedestrian restraint systems in metal.
[9] Australian Standards Bridge Design Code 5100.