ISO/TR 12353-4:2024
(Main)Road vehicles — Traffic accident analysis — Part 4: Compilation of methodologies for assessment of vehicle safety system effectiveness
Road vehicles — Traffic accident analysis — Part 4: Compilation of methodologies for assessment of vehicle safety system effectiveness
This document compiles common methods for assessing the effectiveness of vehicle safety systems. This covers active, passive and integrated safety systems including crash avoidance systems. Effectiveness in this context refers to the capability of a safety system or feature to avoid or mitigate injuries, fatalities or crashes. The document covers both prospective and retrospective methodologies. Applicability, advantages, limitations, accuracy and sensitivity are described for each method. Necessary input and output data and format are also presented.
Véhicules routiers — Analyse des accidents de la circulation — Partie 4: Compilation des méthodologies pour l'évaluation de l'efficacité des systèmes de sécurité des véhicules
General Information
Relations
Standards Content (Sample)
Technical
Report
ISO/TR 12353-4
First edition
Road vehicles — Traffic accident
2024-11
analysis —
Part 4:
Compilation of methodologies for
assessment of vehicle safety system
effectiveness
Véhicules routiers — Analyse des accidents de la circulation —
Partie 4: Compilation des méthodologies pour l'évaluation de
l'efficacité des systèmes de sécurité des véhicules
Reference number
© ISO 2024
All rights reserved. Unless otherwise specified, or required in the context of its implementation, no part of this publication may
be reproduced or utilized otherwise in any form or by any means, electronic or mechanical, including photocopying, or posting on
the internet or an intranet, without prior written permission. Permission can be requested from either ISO at the address below
or ISO’s member body in the country of the requester.
ISO copyright office
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii
Contents Page
Foreword .v
Introduction .vi
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols and abbreviated terms. 2
5 Overview of assessment methodologies . 3
5.1 Prospective and retrospective methods .3
5.2 Exposure .3
5.3 Risk .4
5.4 Odds . . .4
5.5 Field of effect .4
5.6 Effectiveness rate .5
5.7 Potential effectiveness .5
5.8 Benefit .7
5.9 Safety performance .7
6 Crash rate estimation using crash case and exposure data (retrospective assessment) . 8
6.1 Introduction .8
6.2 Applicability, advantages and limitations .8
6.3 Methodology .8
6.3.1 Description .8
6.3.2 Input data .8
6.3.3 Output data.9
6.4 Accuracy, sensitivity and validation .9
7 Dose-response model (retrospective assessment) . 9
7.1 Introduction .9
7.2 Applicability, advantages and limitations .11
7.3 Methodology . 12
7.3.1 Description . 12
7.3.2 Input data . 12
7.3.3 Output data. 12
7.4 Accuracy, sensitivity and validation . 12
8 Induced exposure (retrospective assessment) .12
8.1 Introduction . 12
8.2 Applicability, advantages and limitations . 12
8.3 Methodology . 13
8.3.1 Description . 13
8.3.2 Input data .14
8.3.3 Output data.14
8.4 Accuracy, sensitivity and validation .14
9 Paired comparisons (retrospective assessment) . 14
9.1 Introduction .14
9.2 Applicability, advantages and limitations .14
9.3 Methodology .14
9.3.1 Description .14
9.3.2 Input data . 15
9.3.3 Output data. 15
9.4 Accuracy, sensitivity and validation . 15
10 Prospective assessment based on virtual simulations according to ISO/TR 21934-1 .16
10.1 Introduction .16
10.2 Applicability, advantages and limitations .16
iii
10.3 Methodology .17
10.3.1 General .17
10.3.2 Input data .18
10.3.3 Output data.19
10.4 Accuracy, sensitivity and validation .19
11 Crash momentum index, C–V plane (prospective or retrospective assessment) .20
r
11.1 Introduction . 20
11.2 Applicability, advantages and limitations . 20
11.3 Methodology .21
11.3.1 Description .21
11.3.2 Input data . 23
11.3.3 Output data. 23
11.4 Accuracy, sensitivity and validation .24
Annex A (informative) Application of prospective and retrospective methods.25
Bibliography .27
iv
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out through
ISO technical committees. Each member body interested in a subject for which a technical committee
has been established has the right to be represented on that committee. International organizations,
governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely
with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
The procedures used to develop this document and those intended for its further maintenance are described
in the ISO/IEC Directives, Part 1. In particular, the different approval criteria needed for the different types
of ISO documents should be noted. This document was drafted in accordance with the editorial rules of the
ISO/IEC Directives, Part 2 (see www.iso.org/directives).
ISO draws attention to the possibility that the implementation of this document may involve the use of
patents. ISO takes no position concerning the evidence, validity or applicability of any claimed patent rights
in respect thereof. As of the date of publication of this document, ISO had not received notice of patents
which may be required to implement this document. However, implementers are cautioned that this
may not represent the latest information, which may be obtained from the patent database available at
www.iso.org/patents. ISO shall not be held responsible for identifying any or all such patent rights.
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation of the voluntary nature of standards, the meaning of ISO specific terms and expressions
related to conformity assessment, as well as information about ISO's adherence to the World Trade
Organization (WTO) principles in the Technical Barriers to Trade (TBT), see www.iso.org/iso/foreword.html.
This document was prepared by Technical Committee ISO/TC 22, Road vehicles, Subcommittee SC 36, Safety
and impact testing.
A list of all parts in the ISO 12353 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www.iso.org/members.html.
v
Introduction
Many methodologies are used to analyse the effectiveness of various vehicle safety systems. Most methods
are retrospective, have different applicability, advantages and limitations, and are often chosen depending
on the structure and content of the data available. More recently, prospective methods have been presented
and used.
The aim of this document is to compile commonly used methods for assessing the effectiveness of vehicle
safety systems. The document covers assessment methods for active, passive and integrated safety systems
including crash avoidance systems. The effectiveness in this context refers to the capability of a safety
system or feature to avoid or mitigate injuries, fatalities or crashes.
The document provides a general overview of commonly used terms for the assessment methodologies,
including exposure, risk, odds, effectiveness, benefit and safety performance.
Six methodologies, both prospective and retrospective, are described in the document. Each method is
summarized in terms of its applicability, advantages and limitations. The methodology is described together
with necessary input data and the resulting output data. Conclusions are given in terms of accuracy,
sensitivity and validation for each method.
An overview of the applicability of prospective and retrospective assessment methods is also included (see
Annex A).
The methods included in this document were considered to be in use and valid for this compilation by the
time of development. If needed and requested, this document can be expanded with additional methods in a
later revision.
vi
Technical Report ISO/TR 12353-4:2024(en)
Road vehicles — Traffic accident analysis —
Part 4:
Compilation of methodologies for assessment of vehicle
safety system effectiveness
1 Scope
This document compiles common methods for assessing the effectiveness of vehicle safety systems. This
covers active, passive and integrated safety systems including crash avoidance systems.
Effectiveness in this context refers to the capability of a safety system or feature to avoid or mitigate injuries,
fatalities or crashes.
The document covers both prospective and retrospective methodologies. Applicability, advantages,
limitations, accuracy and sensitivity are described for each method. Necessary input and output data and
format are also presented.
2 Normative references
The following documents are referred to in the text in such a way that some or all of their content constitutes
requirements of this document. For dated references, only the edition cited applies. For undated references,
the latest edition of the referenced document (including any amendments) applies.
ISO 12353-1, Road vehicles — Traffic accident analysis — Part 1: Vocabulary
3 Terms and definitions
For the purposes of this document, the terms and definitions given in in ISO 12353-1 and the following apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
3.1
mitigate
reduce the consequences of a hazardous event
Note 1 to entry: In the context of this document, the consequences are injuries, fatalities or crash severity.
3.2
injury risk
IR
probability of occurrence of a personal injury at a specific level
Note 1 to entry: Injury level is often expressed with the abbreviated injury scale (AIS).
3.3
relative risk
risk ratio
ratio of risk of an event in one group versus the risk of the event in the other group
EXAMPLE An exposed group versus a non-exposed group.
3.4
odds
probability that the event occurs divided by the probability that the event does not occur
3.5
odds ratio
probability of an event occurring in one group versus the probability of the event occurring in the other group
3.6
eccentricity
distance between the impact force vector of the centres of gravity of the two vehicles in an eccentric impact
4 Symbols and abbreviated terms
A number of accident situations sensitive to the system
AEB autonomous emergency brake
ADAS advanced driver assistance system
B benefit
C crash-momentum index
E effectiveness
F field of effect
IR injury risk
N number of (all) accident situations
n number of crashes (of a certain type)
P crash rate
PDO property damage only
Q penetration factor
R relative risk (risk ratio)
S safety performance
ΔV change of velocity (delta-v)
X exposure
5 Overview of assessment methodologies
5.1 Prospective and retrospective methods
Assessments are calculations of performance that tell the value of a subject in comparison to an aim or
objective. This document describes commonly used assessment methodologies for traffic safety measures
in vehicles in the complete driver-vehicle-traffic system (i.e. when the measure is deployed in the real-world
traffic with the wide-ranging distribution of environments and traffic participants).
The two main types of assessments are studies performed before (prospectively) or after (retrospectively)
the introduction of a safety measure, see Figure 1. The headings of the clauses in this document indicate
whether the assessment method described is prospective or retrospective.
Methodologies to categorize traffic safety performance assessments use tools (e.g. virtual simulation,
accident database analysis) and input data (e.g. crash data, naturalistic driving data). Tools and input
data are discussed for each method in this document. Some applications of prospective and retrospective
methods are given in Annex A.
Figure 1 — Illustration of prospective and retrospective assessment methods
5.2 Exposure
The traditional definition of exposure is of being in a place or situation where there is no protection from
something harmful or unpleasant. According to ISO 26262-1, exposure is defined under functional safety
as the state of being in an operational situation that can be hazardous, which can occur at any point in a
vehicle's lifetime (a combination of an operational situation and a potential source of harm). A similar
interpretation is given in ISO 21448, which refers to the ISO 26262 series. In these documents, exposure is
a factor for potential risk calculation that describes the (expected) frequency of occurrences of situations of
interest. For prospective accident research, a similar approach is taken in case the frequency of scenarios is
relevant for the calculation of the risk of accident.
For retrospective traffic accident research, the perspective is different, since the focus is on the calculation
of accident rates for different groups. These groups can be defined by other parameters, such as technologies,
vehicle types, road types and driver types. A simple comparison of the number of accidents can be
misleading in the analysed data set. Therefore, a rate is calculated to correct the comparison for the different
representation of the groups. In Reference [3] the problem is illustrated by asking which sex is more likely to
be involved in accidents. Although there are copious data available on the number of accidents in which male
and female drivers are involved, this question remains difficult to solve, since it requires comparison of the
number of accidents per unit of exposure for each group. Typical units for exposure are travelled distance or
time, traffic density and/or crash severity.
Thus, in retrospective accident analysis, exposure is defined according to ISO 12353-1 as a parameter
describing the dose or amount of some physically measurable parameter(s) that are related to an accident or
injury or both.
5.3 Risk
According to ISO 26262-1, risk is the combination of the probability of occurrence of harm and the severity
of that harm.
In this document, risk is the probability of occurrence of an accident with a specific severity (due to the
application of a safety system).
As for the risk of personal injury, injury risk (IR) is the probability of occurrence of a personal injury at a
specific level.
The severity of an accident is estimated by the type of injuries and the extent of affected people, for example:
— property damage only (PDO);
— minor, major or lethal injuries;
— one or several persons, objects.
The probability of occurrence of harm depends on the exposure to the hazard, the occurrence of relevant
situations and the possibility to avoid or limit harm by external factors:
— how often or how much time is spent with the hazardous object;
— relevant statistical, historical or reference information;
— skillset and awareness of user, experience, lead-time to harm.
5.4 Odds
Odds is defined as the probability that the event will occur divided by the probability that the event will
not occur.
The odds ratio is the ratio of the odds of an event in one group versus the odds of the event in the other group.
NOTE See also Reference [4].
5.5 Field of effect
The field of effect, F, defines a specific subset within a superset of considered accident situations. The
superset of traffic situations contains all occurrences recorded in a particular region. The accident situations
in the field of effect are specified by common characteristics. These can be, for instance, accident causes,
accident scenarios, involved participants and other concomitant circumstances.
In practice, the field of effect is the proportion of all accident situations in which a specific safety system can
have a positive effect. The safety system is designed to become active in these situations in order to avoid
or mitigate the accident. Thus, the field of effect describes all accident situations that are addressed by the
safety system.
The field of effect is calculated using Formula (1):
A
F =× 100 (1)
N
where
F is the field of effect, expressed in per cent;
A is the number of accident situations sensitive to the system;
N is the number of all accident situations.
EXAMPLE The field of effect of an autonomous emergency brake (AEB) system comprises all run-up accidents
between cars. About 6 % of all injury accidents in Germany are car-to-car run-up accidents that can be addressed by AEB.
5.6 Effectiveness rate
The effectiveness rate defines the proportion of accident situations in the field of effect that are positively
affected by the regarded safety system. It describes how well the safety system performs within its field of effect.
The effectiveness rate of the safety system depends on how well the system addresses all possible accident
situations within the defined field of effect. It also relies on a reliable and functioning system performance.
Ideally, all system-specific accident situations are avoided or at least mitigated by the safety system.
The effectiveness rate is calculated by Formula (2):
A
Reduction
E =× 100 (2)
Rate
A
where
E is the effectiveness rate, expressed in per cent;
Rate
A is the number of avoided or mitigated accidents that are sensitive to the system;
Reduction
A is the total number of accidents sensitive to the system.
EXAMPLE The effectiveness rate of an AEB system specifies the part of all car-to-car run-up accidents that are
avoided or mitigated by AEB. The AEB effectiveness rate amounts to approximately 90 %.
5.7 Potential effectiveness
The potential effectiveness defines the maximal proportion of all accident situations that are positively
affected by the safety system. It describes the overall benefit of a safety system if all vehicles in the field
were equipped with such a system.
The potential effectiveness is calculated by Formula (3):
A
Reduction
E =× 100 (3)
Pot
N
where
E is the potential effectiveness, expressed in per cent;
Pot
A is the number of avoided or mitigated accidents that are sensitive to the system;
Reduction
N is the total number of all accidents.
If the field of effect and the effectiveness rate are known, the potential effectiveness can be calculated as in
Formula (4):
FE ⋅
Rate
E = (4)
Pot
where
E is the potential effectiveness, expressed in per cent;
Pot
F is the field of effect, expressed in per cent;
E is the effectiveness rate, expressed in per cent.
Rate
EXAMPLE 1 The potential effectiveness of an AEB system specifies the part of all accidents that are avoided or
mitigated by AEB. If the field of effect is 6 % and the effectiveness rate is 90 %, the potential effectiveness is 5,4 %.
Figure 2 shows an illustration of the exposure and effectiveness definitions.
Key
A accidents sensitive to system S
A avoided/mitigated accidents by system S
Reduction
N all accidents, including property damage only
N all accidents with personal injury
PI
ODD operational design domain
Figure 2 — Qualitative visualization of the defined effectiveness terms
The overall effectiveness describes the proportion of accident situations that are positively affected,
assuming a specific number of vehicles are equipped with the related safety system. It defines the concrete
system effectiveness depending on a given market penetration.
The effectiveness is calculated by Formula (5):
EQ ⋅
Pot
E = (5)
where
E is the effectiveness, expressed in per cent;
E is the potential effectiveness, expressed in per cent;
Pot
Q is the penetration factor, expressed in per cent.
NOTE See also ISO 12353-1:2020, 6.4.
The penetration factor defines how the percentage of equipped vehicles directly affects the effectiveness of
the system. In other words, the penetration factor describes the system's dependency on other vehicles or
infrastructure, that can be equipped with appropriate systems.
EXAMPLE 2 The penetration factor of an AEB system solely depends on the number of vehicles equipped. If the
potential effectiveness of AEB is 5,4 %, 50 % of all vehicles have AEB fitted and we assume at least one vehicle per
accident, then the overall effectiveness is at least 2,7 %. This implies that a minimum of 2,7 % of all traffic accidents
would be prevented by AEB.
EXAMPLE 3 Vehicle to vehicle (V2V) applications rely on both participating vehicles to be fitted with the V2V
system. Therefore, the penetration factor is always only half of the current fitting rate.
5.8 Benefit
The benefit of the safety system describes the absolute number of accident situations that are positively
affected. It defines how many accident situations are avoided or mitigated.
The benefit is calculated by Formula (6):
EN ⋅
B = (6)
where
B is the benefit, expressed in number of accidents;
E is the effectiveness, expressed in per cent;
N is the total number of all accidents.
EXAMPLE The benefit of an AEB system specifies the number of all accidents avoided or mitigated by AEB. With
approximately 2,3 million traffic accidents in Germany in 2021 and given an AEB effectiveness of 2,7 % (see 5.7,
EXAMPLE 2) the benefit of AEB would be 62 000 avoided accidents per year.
5.9 Safety performance
Safety performance, expressed in km or miles (mi), is the inverse of the accidents per distance measure,
according to Formula (7):
1 d
S == (7)
A A
Rate
where
S
is the safety performance, expressed in km or mi;
A
is the accident rate, expressed in number of accidents per distance travelled (km or mi);
Rate
d is the distance travelled, expressed in km or mi;
A is the total number of accidents sensitive to the system.
The safety performance represents the distance travelled between two accidents, or more generally the
distance travelled between two events of the same category. The bigger the value of the safety performance
and therefore the distance between two events, the safer the vehicle which is observed.
EXAMPLE In 2019, 755 billion km were driven in Germany. The police recorded 2 685 661 accidents. This means
[8]
one accident happened every 280 000 km. This gives a highway safety performance of 1 420 376 km .
6 Crash rate estimation using crash case and exposure data (retrospective
assessment)
6.1 Introduction
The analysis of crash rates is a basic and straightforward method for evaluating traffic safety
countermeasures. The crash rate can represent a meaningful and comparative measure since it reflects
crash numbers in relation to a measure of exposure. However, some confounding factors can influence the
crash rate of a particular car model, e.g. driver behaviour. If adjustments are made to control for confounding
factors, crash rate can be used as a reference when evaluating effects from both injury- and crash-preventing
arrangements in the traffic environment.
The principle of using crash rates in the traffic safety research community is frequently used or considered
[9]
for many applications. For example, for ranking the performance of countries , comparing safety
[10]
levels of road surface conditions , or, as in this document, evaluating vehicle safety performance (see
References [12] to [16]).
6.2 Applicability, advantages and limitations
For retrospective assessments of traffic safety countermeasures using crash rate analysis, a precondition is
that the technology under study has a sufficient penetration rate (for a certain vehicle subgroup or class) in
the market and that sufficient crash cases with and without the technology are recorded.
Depending on the research question, traffic safety indicators can be evaluated using the case and exposure
data method. However, different definitions of rates are being used related to the datasets available for the
analysis. Exposure data measured in, for example, per capita, per registered vehicles, per km travelled or
per insured vehicle years are frequently used to assess change in crash and/or injury risk. Many studies
have also evaluated the probability of sustaining injuries in the vicinity of a crash.
6.3 Methodology
6.3.1 Description
Typically, the crash rate, P, is defined as in Formula (8):
n
P= (8)
X
where
P is the crash rate;
n is the number of crashes (of a certain type);
X is the exposure.
Crash involvement rates can be compared for relevant and corresponding groups of vehicles, with and
without a traffic safety countermeasure.
Furthermore, it is possible to fit regression models to the data to estimate changes in crash involvement and
control for contributing factors.
6.3.2 Input data
Case data on crashes and exposure data, with and without the traffic safety countermeasure in comparable
situations (e.g. the same car make and models, the same time period, in similar overall traffic environments
and in the countermeasure’s specific target conflict situation or crash configuration) is a prerequisite for the
analysis in a quantity that acknowledges the desired statistical significance and power.
The number of pre-crash factors that can be included in the analysis is limited to what is known for both
case data (crashes) and exposure data.
6.3.3 Output data
In their basic form, output data provide comparisons of crash rates between situations with and without the
traffic safety countermeasure.
6.4 Accuracy, sensitivity and validation
One main issue in retrospective effectiveness estimations is the availability of crash datasets large enough
to provide statistically reliable results. Collecting data from crashes is a time-consuming activity. Further,
traffic safety countermeasures are often not immediately deployed on a large scale, but rather are limited
to geographical areas when considering infrastructural devices and as optional mounted equipment in new
vehicles. In the latter case, it is often hard to know from current datasets whether the safety technology was
turned on or off.
Another challenge is the fact that case and exposure data rarely coexist in the same database. In Reference
[11] possible combinations of Swedish national databases for estimating crash rates for different situations
and crash severities are shown. There were many factors the researchers could not include in the analysis.
7 Dose-response model (retrospective assessment)
7.1 Introduction
A general overview of parameters related to impact severity and injury outcome is shown in Figure 3, which
illustrates a vehicle impact as the relation between impact severity and injury consequences in a chain of
events that can be denoted as dose-response models.
Figure 3 — Parameters related to impact severity and injury outcome, application in
the dose–response model
Three aspects of an accident sample are important for the analysis of safety systems:
— the exposure in terms of frequency of collisions;
— frequency of injured occupants;
— injury risk versus impact severity, in the event of a collision.
This is illustrated by the three curves in Figure 4.
These three curves are assessed and used in analyses in several studies, e.g. in References [17] to [21]. This
way of describing the three aspects of accidents can be denoted as a dose-response model. The link deciding
the response of the dose is the injury risk function.
There are three options for reducing the number of injured occupants in car collisions:
a) by reducing the severity of the impacts, or
b) by reducing the number of collisions, or
c) by reducing the injury risk at a given impact severity.
The three options are illustrated in Figure 4 by the arrows at numbers 1, 2 and 3 (see Reference [21]).
The first option can be achieved by, for example, reducing speed limits, by reducing impact speed or by
redesigning the infrastructure. The second can be achieved by active safety measures aimed at preventing
collisions from occurring. The third can be achieved by the passive safety of the vehicle and the road
infrastructure safety features to prevent injuries from occurring.
The dose-response model can be used to evaluate the effectiveness of various safety technologies. Safety
technologies aimed at mitigating crash severity and at increasing the protection by preparing for a crash
situation can address all three options.
Key
X impact severity
Y number of crashes and number of injured occupants
Y injury risk
1 reducing the severity of the impacts
2 reducing the number of collisions
3 reducing the injury risk at a given impact severity
A number of crashes (exposure)
B number of injured occupants
C injury risk
Figure 4 — Dose-response model including the three methods of reducing the number of injured
occupants
7.2 Applicability, advantages and limitations
Dose-response models can be used to study the effectiveness of new safety technologies, especially those
aimed at mitigating crash severity, preparing for crash protection or even avoiding collisions. At least one of
the functions needs to be known depending on what will be analysed.
If the injury or fatality risk functions as well as the crash distribution are known, it is possible to estimate
the effect on injury outcome depending on the crash severity possible to be reduced.
7.3 Methodology
7.3.1 Description
The three approaches to reducing the number of injured occupants described in Figure 4 can be studied
separately. If the injury risk remains the same, a reduced number of injuries can be calculated based on a
reduction of delta-v for all or part of the crashes. A reduction of injured occupants can also be calculated
based on a reduced number of collisions. Similarly, a reduction of injured occupants can be calculated for
a car with lower injury risk. In this case, the injury risk for a new and old car needs to be known, which is
uncommon. Expected reductions of injured car occupants have been calculated in studies, see for example
References [22] and [23].
Similar calculations can also be made by reducing the mean acceleration instead of change of velocity.
Reducing mean acceleration is relevant to use, especially in the design of road infrastructure, such as
deformable posts, guardrails or mid barriers, see ISO 12353-2.
The dose-response model can also be used to calculate injury risk curves if the distribution of crashes and
injured occupants for a selected crash severity parameter is known.
7.3.2 Input data
Crash distribution for crash severity can be, for example, change of velocity or mean acceleration. The
distribution can, for example, be derived from crash recorder/EDR data (see ISO/TR 12353-3). It is also
necessary to have injury or fatality risk functions for the selected crash severity parameter.
7.3.3 Output data
The direct output is the distribution of injury or fatality. Based on the distributions, reduction of injury or
fatality can be calculated.
7.4 Accuracy, sensitivity and validation
Crash distributions can be estimated based on previous research. It is more difficult and important to find
a suitable risk function. Accuracy depends on the accuracy in the risk function and in the distribution. The
accuracy of the risk function in the low severity segment is especially important since it has a large influence
on the number of injured due to the distribution of crashes.
8 Induced exposure (retrospective assessment)
8.1 Introduction
Real exposure in terms of vehicle mileage or number or registered vehicles is often not available. The
accident sample can also be associated with several confounding variables, influencing bias, for example,
those who choose to purchase vehicles with a certain safety technology are probably more concerned about
their safety in the first place (i.e. selective recruitment). There are some indirect statistical methods that
can be used to overcome this. One is called induced exposure. Instead of comparing crash or injury outcomes
with real exposure, changes in distribution of crashes can be studied to estimate the effectiveness of a safety
technology. See References [25] to [27].
8.2 Applicability, advantages and limitations
A key point is to identify at least one crash type in which the countermeasure under analysis can be
reasonably assumed (or known) not to be effective.
In the case given in 8.3.1, cars with and without low-speed AEB were compared. If the only noteworthy
difference in terms of involvement in rear-end crashes is AEB, the relation between cars with and without
AEB in that non-sensitive situation is considered as the true exposure relation. If the influence of
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.
Loading comments...