ISO/TS 20459:2023

Road vehicles - Injury risk functions for advanced pedestrian legform impactor (aPLI)

This document provides definitions, symbols and injury probability functions (IPFs) for the thigh, leg and knee intended to be used with the advanced pedestrian legform impactor (aPLI), a standardized pedestrian legform impactor with an upper mass for pedestrian subsystem testing of road vehicles. They are applicable to impact tests using the aPLI at 11,1 m/s involving: - vehicles of category M1, except vehicles with a maximum mass above 2 500 kg and which are derived from N1 category vehicles and where the driver’s position, the R-point, is either forward of the front axle or longitudinally rearwards of the front axle transverse centreline by a maximum of 1 100 mm; - vehicles of category N1, except where the driver’s position, the R-point, is either forward of the front axle or longitudinally rearwards of the front axle transverse centreline by maximum of 1 100 mm; - impacts to the bumper test area defined by References [1] and [2]; - pedestrian subsystem tests involving use of a legform for the purpose of evaluating compliance with vehicle safety standards.

Véhicules routiers — Critères lésionnels et courbes de risques pour l'impacteur en forme de jambe de piéton (aPLI).

General Information

• Status: Published
• Publication Date: 27-Apr-2023
• ICS: 43.020 - Road vehicles in general
• Technical Committee: ISO/TC 22/SC 36 - Safety and impact testing
• Drafting Committee: ISO/TC 22/SC 36/WG 6 - Performance criteria expressed in biomechanical terms
• Current Stage: 6060 - International Standard published
• Start Date: 28-Apr-2023
• Due Date: 27-Jan-2023
• Completion Date: 28-Apr-2023

Overview - ISO/TS 20459:2023 (aPLI IPFs)

ISO/TS 20459:2023 defines injury probability functions (IPFs) for use with the advanced pedestrian legform impactor (aPLI). The technical specification provides definitions, symbols and statistically derived IPFs for the thigh, leg and knee to predict pedestrian lower-limb injury risk from standardized subsystem tests. The IPFs are intended for harmonized, worldwide use with the aPLI (standard build level B) in legform-to-bumper impacts at 11.1 m/s and relate peak aPLI injury metrics to the probability of real-world human injury.

Key topics and technical requirements

• Scope and applicability
  • Applies to aPLI impacts at 11.1 m/s for many passenger vehicles (categories M1 and N1) with specified exclusions related to vehicle mass and R‑point position, and for impacts to the defined bumper test area.
  • Intended for pedestrian subsystem tests for compliance evaluation with vehicle safety standards.
• Body regions covered
  • Thigh, leg and knee IPFs based on aPLI measurements.
• Methodology
  • Human IPFs derived from biomechanical data using survival analysis and covariates (pedestrian size/weight and age).
  • aPLI IPFs are generated by converting human IPFs using methods such as transfer‑functions (TF) and energy‑equivalent (EE) approaches; paired test method is also referenced.
  • Statistical distributions supported include Weibull, Log‑Normal and Log‑Logistic - the best fit is selected per metric.
• Test conditions specified
  • Representative adult (50th percentile male: 175.1 cm, 76.7 kg), age assumption (60 years), upright posture with knee fully extended, lateral‑to‑medial impact direction, foot sole 25 mm above ground.
• Guidance and annexes
  • Annex A: rationale and methodology
  • Annex B: adjustment of IPFs for real‑world relevance
  • Annex C/D: supplemental and PMHS-related data

Practical applications and users

• Vehicle manufacturers, crash‑safety engineers and test laboratories use ISO/TS 20459:2023 to:
  • Translate aPLI sensor outputs (peak forces, moments, energies) into injury probabilities for thigh, leg and knee.
  • Define injury assessment reference values for subsystem test procedures and regulatory or voluntary pedestrian protection programs.
  • Validate and compare pedestrian lower‑limb countermeasures and front‑end designs (bumpers, bonnets).
• Regulators, research institutes and HBM/FE modelers can use the IPFs to interpret subsystem test results and to harmonize injury risk assessment across jurisdictions.

Related standards

• ISO/TS 18506 (statistical methods for IPFs)
• ISO/TS 18571:2014 (ISO metric/time‑history verification)
• ISO/TR 12350:2013 (paired test methodology)
• References and annexes in ISO/TS 20459 provide further data and adjustment procedures for real‑world applicability.

Keywords: ISO/TS 20459:2023, aPLI, advanced pedestrian legform impactor, injury probability functions, IPFs, pedestrian protection, legform impactor, vehicle safety, thigh leg knee.

Standards Content (Sample)

TECHNICAL ISO/TS
SPECIFICATION 20459
First edition
2023-04
Road vehicles — Injury risk functions
for advanced pedestrian legform
impactor (aPLI)
Véhicules routiers — Critères lésionnels et courbes de risques pour
l'impacteur en forme de jambe de piéton (aPLI).
Reference number
© ISO 2023
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: [phone]
Email: [e-mail]
Website: www.iso.org
Published in Switzerland
ii
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Symbols and abbreviated terms.3
4.1 Symbols . 3
4.2 Abbreviated terms . 4
5 IPFs for the aPLI . 4
5.1 General . 4
5.2 Thigh . . 6
5.3 Leg. 7
5.4 Knee . . 8
Annex A (informative) Rationale regarding background and methodology to develop IPFs
for the aPLI .11
Annex B (informative) Adjustment of IPFs for real-world relevance . 105
Annex C (informative) Supplemental data . 135
Annex D (informative) Influence of PMHS test data (dfbetas > 0,3) against IPFs for human . 136
Bibliography .154
iii
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).
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. ISO shall not be held responsible for identifying any or all such patent rights. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www.iso.org/patents).
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.
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.
iv
Introduction
This document has been prepared on the basis of the existing injury probability functions (IPFs) to be
used with the advanced pedestrian legform impactor (aPLI) standard build level B (SBL-B). The purpose
of this document is to document the IPFs for the aPLI in a form suitable and intended for worldwide
harmonized use.
In 2014, development of the aPLI hardware and associated IPFs started, with the aim of defining
a globally accepted next-generation pedestrian legform impactor with enhanced biofidelity and
injury assessment capability, along with its IPFs, suitable for harmonized use. Participating in the
development were research institutes, dummy and instrumentation manufacturers, governments, and
car manufacturers from around the world.
IPFs for the aPLI specified in this document predict injury probability to specific regions of the lower
limb of a pedestrian that corresponds to maximum values of injury metrics obtained by the aPLI in a
subsystem test, as described in References [1] and [2]. As the IPFs do not provide any threshold values,
users will need to determine target injury probability, based on their specific needs, to define injury
assessment reference values to be used for their test protocol.
It is also important to note that the subsystem test procedure (STP) for pedestrian protection may not
be representative of pedestrian accidents for specific injury metrics, depending on their sensitivity to
pedestrian impact conditions such as lower-limb posture and muscle tone. The IPFs for the aPLI have
been validated against accident data and some ideas to compensate for the discrepancy against accident
data are presented in Annex B.
v
TECHNICAL SPECIFICATION ISO/TS 20459:2023(E)
Road vehicles — Injury risk functions for advanced
pedestrian legform impactor (aPLI)
1 Scope
This document provides definitions, symbols and injury probability functions (IPFs) for the thigh, leg
and knee intended to be used with the advanced pedestrian legform impactor (aPLI), a standardized
pedestrian legform impactor with an upper mass for pedestrian subsystem testing of road vehicles.
They are applicable to impact tests using the aPLI at 11,1 m/s involving:
— vehicles of category M1, except vehicles with a maximum mass above 2 500 kg and which are derived
from N1 category vehicles and where the driver’s position, the R-point, is either forward of the front
axle or longitudinally rearwards of the front axle transverse centreline by a maximum of 1 100 mm;
— vehicles of category N1, except where the driver’s position, the R-point, is either forward of the front
axle or longitudinally rearwards of the front axle transverse centreline by maximum of 1 100 mm;
— impacts to the bumper test area defined by References [1] and [2];
— pedestrian subsystem tests involving use of a legform for the purpose of evaluating compliance
with vehicle safety standards.
2 Normative references
There are no normative references in this document.
3 Terms and definitions
For the purposes of this document, the following terms and definitions 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
adult
person who is sixteen years old or older
3.2
advanced pedestrian legform impactor
aPLI
modified pedestrian legform impactor which incorporates a mass representing the inertial effect of
the upper part of a pedestrian body to enhance biofidelity and injury assessment capability (3.10) of
conventional pedestrian legforms
3.3
biofidelity
aspect of the advanced pedestrian legform impactor (aPLI) (3.2) capability to represent the impact
response of human subjects
3.4
BLE height
bonnet leading edge height
height of the geometric trace of the upper most points of contact between a straight edge and the front-
end of the car
3.5
bumper test area
test area of the legform to bumper impact test
3.6
bumper system
component installed at the hip joint inside the upper mass composed of the bumper, the bumper mount
and the compression surface, designed to apply a force on the upper part of the femur in adduction to
enhance injury assessment capability (3.10) of the advanced pedestrian legform impactor (aPLI) (3.2)
3.7
EE method
energy-equivalent method
method of developing injury probability functions (IPFs) (3.11) for the advanced pedestrian legform
impactor (aPLI) (3.2) by transferring human injury values to those of an aPLI using the absorbed energy
3.8
high-bumper car
car with a lower bumper reference line height (3.14) of 425 mm or more
3.9
hip joint
uniaxial joint that allows abduction and adduction and connects the upper mass with the lower limb
3.10
injury assessment capability
aspect of the advanced pedestrian legform impactor (aPLI) (3.2) capability to produce peak injury values
that correlate with those obtained from human body model impact simulations
3.11
IPF
injury probability function
function which defines the relationship between a peak value of an injury metric and probability of
injury for a specific load case
3.12
ISO metric
objective rating metric used in this document to verify time histories of sensor output against
experimentally or computationally produced target time histories as detailed in ISO/TS 18571:2014
3.13
low-bumper car
car with a lower bumper reference line height (3.14) less than 425 mm
3.14
LBRL height
lower bumper reference line height
height of the geometric trace of the lowermost points of contact between a straight edge and the
bumper, measured from the ground
3.15
low-pass filter
filter which permits only low-frequency (100 Hz or less) oscillations
3.16
paired test method
method of developing injury probability functions (IPFs) (3.11) by correlating human injury occurrence
in a specific impact configuration with the injury value measured by an ATD subjected to the same
impact as detailed in ISO/TR 12350:2013
3.17
subsystem test
test to evaluate safety performance of cars where subsystem impactors representing individual
body regions of a pedestrian are propelled into a front end of a stationary car, in impact conditions
representing specific load cases in car-pedestrian accidents
3.18
transfer function
TF
linear regression function between human injury values predicted by human body models and advanced
pedestrian legform impactor (aPLI) (3.2) injury values
3.19
TF method
transfer-function method
method of developing injury probability functions (IPFs) (3.11) for the advanced pedestrian legform
impactor (aPLI) (3.2) by converting human IPFs to those of the aPLI using corresponding transfer
functions (3.18)
4 Symbols and abbreviated terms
4.1 Symbols
See Table 1.
Table 1 — Symbols and their meanings
Symbol Meaning
C Parameter determined for the Weibull distribution for human IPFs
Scale
C
Parameter determined for the Weibull distribution for human IPFs
Shape
C
Slope of the transfer function
Slope
C
Parameter determined for the Log-Normal distribution for human IPFs
μ
C Parameter determined for the Log-Normal distribution for human IPFs
σ
C Correction factor determined to adjust to the real-world accident data
TA1
C Correction factor determined to adjust to the real-world accident data
TA2
F IPF for human
G Transfer function
I Injury metric for human
human
I Injury metric for the aPLI
aPLI
P Injury probability of human
P
Adjusted injury probability for the MCL
adj
x Value of the injury metric for the aPLI
aPLI
x Value of the injury metric for human
human
4.2 Abbreviated terms
See Table 2.
Table 2 — Abbreviated terms and their meanings
Abbreviation Meaning
ACL Anterior Cruciate Ligament
aPLI advanced Pedestrian Legform Impactor
ATD Anthropometric Test Device
BLE Bonnet Leading Edge
BM Bending Moment
BP Bumper
EE Energy Equivalent
EEVC European Enhanced Vehicle-safety Committee
FE Finite Element
HBM Human Body Model
IPF Injury Probability Function
LBRL Lower Bumper Reference Line
MCL Medial Collateral Ligament
PCL Posterior Cruciate Ligament
PMHS Post Mortem Human Subjects
RCM Real Car Model
SCM Simplified Car Model
SP Spoiler
STP Subsystem Test Procedure
TF Transfer Function
TG Task Group
5 IPFs for the aPLI
5.1 General
The IPFs specified in this document are to be used with the aPLI for the thigh, leg and knee to predict
the probability of injuries to pedestrians when involved in real-world car-pedestrian accidents. The
IPFs provide a statistically derived relationship between the maximum values of injury metrics
obtained from a test conducted using the aPLI by following the subsystem test procedure (STP), and
the probability of injury to a corresponding body region of a pedestrian when subjected to load cases
representative of the majority of real-world accidents.
The specific load case represented by the subsystem legform test is described below:
th
— pedestrian size and weight: 175,1 cm and 76,7 kg representing a 50 percentile adult male (Reference
[3]);
— impact speed: 11,1 m/s;
— impact direction: lateral-to-medial direction to a pedestrian lower limb;
— lower-limb posture: upright (vertical to the ground) with the knee fully extended;
— impact height: sole of the foot positioned 25 mm above the ground to represent a shoe sole height.
First, human IPFs were determined using human biomechanical data available from the literature. Data
obtained by the experiments conducted under the loading conditions equivalent to those specified in
the STP were referred to. The statistical method used to derive human IPFs follows that recommended
by ISO/TS 18506 with the covariates of pedestrian size, weight and age. The pedestrian size and weight
were determined from those specified in STP. The age was set at 60 years old that corresponds to the
average age of the subjects of the biomechanical data as this choice was found to provide the most
reasonable set of assumptions when the IPFs were fitted to the accident data. The recommended
method estimates parameters of any one of the Weibull, Log-Normal or Log-Logistic distribution
(choose the one that best fits to data) with survival analysis method. In this document, one of the three
distributions (Weibull distribution, Log-Normal distribution or Log-Logistic distribution) is used to
define human IPFs for each of the injury metrics. The formulae of the aPLI IPFs for these distributions
are presented below.
The injury probability when the Weibull distribution is applied following Formula (1):
C
 Shape 
Cx×
 
 SlopeaPLI 
P =−1exp − (1)
   
C
  Scale  
 
where
P is the injury probability of human;
C
is the parameter determined for the Weibull distribution for human IPFs;
Scale
C
is the parameter determined for the Weibull distribution for human IPFs;
Shape
C
is the slope of the transfer function (TF);
Slope
x
is the value of the injury metric for the aPLI.
aPLI
The injury probability when the Log-Normal distribution is applied following Formula (2):
 
Cx× −− ln −tC−
()
1 1
SlopeaPLI  μμ 
P = exp dt (2)
 
∫
t
C 2ππ
2C
 
σσ
σσ
 
where
P is the injury probability of human;
C
is the parameter determined for the Log-Normal distribution for human IPFs;
μ
C
is the parameter determined for the Log-Normal distribution for human IPFs;
σ
C
is the slope of the TF;
Slope
x
is the value of the injury metric for the aPLI.
aPLI
The injury probability when the Log-Logistic distribution is applied following Formula (3):
P = (3)
−−1
C ×x
 C
SlopeaPLI
Shape
1++
 
exp C
()
 Scale 
where
P is the injury probability of human;
C
is the parameter determined for the Log-Logistic distribution for human IPFs;
Scale
C
is the parameter determined for the Log-Logistic distribution for human IPFs;
Shape
C
is the slope of the TF;
Slope
x
is the value of the injury metric for the aPLI.
aPLI
For each of the thigh, leg and knee, IPFs for a human body are then transferred to those of the aPLI using
a TF, which is a linear function between the maximum values of a human and aPLI injury metrics. Due
to the lack of biomechanical data, the TFs were determined from the results of computational impact
simulations using FE human body models (HBMs) and aPLI FE models in loading conditions specified
in the STP. Details of the human IPFs from which IPFs for the aPLI are derived can be found in A.2.3. For
the determination of TFs, see A.2.4 for more details.
As the IPFs converted from human IPFs using TFs are for the specific load case defined in the STP, the
number of injuries calculated from each of the injury probabilities predicted by the IPFs were compared
with that of real-world accidents. The IPFs for the knee and the leg were compensated for the real-world
observations for the injury metrics showing a significant inconsistency with accident data. Details of
the compensation to real-world accidents can be found in Annex B.
Supplemental information related to the TFs and IPFs for human is provided in Annex C and Annex D,
respectively.
5.2 Thigh
The IPF for the thigh defines probability of femur shaft fracture to a pedestrian subjected to the load
cases representative of the majority of real-world accidents as a function of maximum value of the
femur bending moment measured by the aPLI.
Figure 1 presents the IPF for the thigh. The injury probability function is shown in a solid line, with the
95 % confidence interval shown in dotted lines. The horizontal axis represents the maximum value of
the femur bending moment measured by the aPLI, and the vertical axis represents the probability of
injury.
The IPF for the thigh is given by Formula (4):
C
Shape
 
Cx×
 
 SlopeaPLI 
P =−1exp − (4)
   
C
 Scale 
 
 
where
P is the injury probability for the femur shaft of human;
C
is the parameter determined for the Weibull distribution for the human IPF for the femur
Scale
shaft as described in A.2.3.4.1;
C
is the parameter determined for the Weibull distribution for the human IPF for the femur
Shape
shaft as described in A.2.3.4.1;
C
is the slope of the TF for the thigh as described in A.2.4.4.1;
Slope
x
is the femur BM measured by the aPLI in Nm.
aPLI
The parameters needed to define the IPF ( C , C and C ) for the function are described in
Scale Shape Slope
Table 3.
Key
X aPLI femur BM [Nm]
Y probability of femur shaft fracture
aPLI IPF for femur shaft
95 % confidence interval
observed data
Figure 1 — IPF for the femur shaft
Table 3 — Parameters of IPF for the femur shaft
C C C
Scale Shape Slope
571 11,0 1,04
5.3 Leg
The IPF for the leg defines probability of tibia shaft fracture to a pedestrian subjected to the specific
load cases representative of the majority of real-world accidents as a function of maximum value of the
tibia bending moment measured by the aPLI.
Figure 2 presents the IPF for the leg. The injury probability function is shown in a solid line, with the
95 % confidence interval shown in dotted lines. The horizontal axis represents the maximum value
of the tibia bending moment measured by the aPLI, and the vertical axis represents the probability of
injury.
The IPF for the leg is given by the Formula (5):
C
Shape
 
Cx×
 
 SlopeaPLI 
P =−1exp − (5)
   
C
 Scale 
 
 
where
P is the injury probability for the tibia shaft of human;
C
is the parameter determined for the Weibull distribution for the human IPF for the tibia
Scale
shaft as described in B.3.3;
C
is the parameter determined for the Weibull distribution for the human IPF for the tibia
Shape
shaft as described in B.3.3;
C
is the slope of the TF for the leg as described in A.2.4.4.2;
Slope
x
is the tibia BM measured by the aPLI in Nm.
aPLI
The parameters needed to define the IPF ( C , C and C ) for the function are described in
Scale Shape Slope
Table 4.
Key
X aPLI tibia BM [Nm]
Y probability of tibia shaft fracture
aPLI IPF for tibia shaft
95 % confidence interval
Figure 2 — IPF for the tibia shaft
Table 4 — Parameters of IPF for the tibia shaft
C C C
Scale Shape Slope
446 3,32 0,881
5.4 Knee
The IPF for the knee defines probability of complete failure of the MCL to a pedestrian subjected to the
specific load cases representative of the majority of real-world accidents as a function of maximum
value of MCL elongation measured by the aPLI.
Figure 3 presents the IPF for the knee. The injury probability function is shown in a solid line, with the
95 % confidence interval shown in dotted lines. The horizontal axis represents the maximum value of
the MCL elongation measured by the aPLI, and the vertical axis represents the probability of injury.
The IPF for the knee is given by Formula (6):
 
Cx××C ××C −− ln −tC−
()
1 1
SlopeaPLITA1 TA2  μμ 
P = exp dt (6)
 
∫
t
C 2ππ
2C
 
σσ
σσ
 
where
P is the injury probability for the MCL of human;
C
is the parameter determined for the Log-Normal distribution for human IPFs for the MCL
μ
as described in A.2.3.4.3;
C
is the parameter determined for the Log-Normal distribution for human IPFs for the MCL
σ
as described in A.2.3.4.3;
C
is the slope of the TF for the knee as described in A.2.4.4.3;
Slope
C
is the correction factor for lower-limb posture and impact angle determined to adjust to
TA1
the real-world accident data as described in B.3.2.2.4;
C
is the correction factor for muscle tone determined to adjust to the real-world accident
TA2
data as described in B.3.2.3;
x
is the MCL elongation measured by the aPLI in mm.
aPLI
The parameters needed to define the IPF ( C ,, C CC, and C ) for the function are described
μσ SlopeTAT12A
in Table 5.
Key
X aPLI MCL elongation [mm]
Y probability of MCL complete rupture
aPLI IPF for MCL complete rupture
95 % confidence interval
observed data
Figure 3 — IPF for the MCL complete rupture
Table 5 — Parameters of IPF for the MCL complete rupture
C C
C C C
μμ Slope
σσ TA1 TA2
3,34 0,291 1,14 0,72 0,90
Annex A
(informative)
Rationale regarding background and methodology to develop IPFs
for the aPLI
A.1 Historical background
A.1.1 General
Although lower extremity injuries are not life-threatening, they are frequent and potentially disabling,
resulting in a substantial cost to the victims and society. The importance of preventing this type of
injuries is illustrated by the United Nation (UN) regulations that aim to mitigate lower extremity
injuries to pedestrians hit by the front-end of cars (See References [1] and [2]).
The UN regulations initially implemented the EEVC pedestrian legform impactor that simply consists
of rigid long bones and a deformable knee joint. In order to improve injury assessment capability, a
new impactor called the Flexible Pedestrian Legform Impactor (FlexPLI) has been developed and
implemented in the phase-2 of Reference [1].
Despite a number of improvements of its capability relative to the EEVC impactor, the FlexPLI still
lacks representation of the influence of the upper part of the body. To address technical issues with
the FlexPLI, including but not limited to the lack of upper body representation, the aPLI with the upper
mass attached to the top of the conventional pedestrian legform to compensate for the lack of the upper
body has been developed. In November 2014, two ISO projects were initiated to develop Technical
Specification (TSs) for "Road vehicles - Modified pedestrian legform impactor for tests of high bumper
vehicles" and "Road vehicles - Injury criteria and risk curves for a modified pedestrian legform
impactor for use with high bumper vehicles", and the aPLI task group (TG) was established by active
participants from research institutes, dummy and instruments manufacturers, governments and car
manufacturers. The aPLI TG has conducted extensive CAE studies to identify optimized specifications
of the aPLI by utilizing HBMs, SCMs and aPLI prototype models. Based on the specifications identified, a
physical version of aPLI SBL-A was fabricated and subjected to international round robin testing.
The aPLI TG also dedicated to their effort to discuss a methodology to develop IPFs. During the
discussion, two different methods were proposed and it was difficult to choose one of the two
proposed methods because both have pros and cons. A new idea taken to facilitate the discussion was
to develop ‘virtual IPFs’ by using parametric human body models (HBMs) with the variability in the
material property of a human body incorporated. The results of this analysis along with some other
consideration resulted in a decision of applying a TF to convert IPFs for human to those for the aPLI.
Due to the lack of sufficient biomechanical data, it was necessary to determine TFs based on the
results of computer simulations using HBMs to take various load cases into consideration. Multiple
HBMs with extensive validation were used to avoid potential bias in case one single HBM is employed.
The specifications of the latest version of the aPLI hardware with the bumper system installed at the
hip joint as defined in ISO/TS 20458 were represented by FE models and used in the analyses. Real
car models (RCMs) were also used to accurately represent geometric and stiffness characteristics
of car front-end structures. The use of multiple different HBMs, RCMs and impact locations allowed
determination of robust IPFs.
In the course of the discussion at the aPLI TG, a question was raised as to real-world relevance of the
IPFs determined in this international effort. The most important decision made by the aPLI TG was that
IPFs shall predict the probability of injury to a pedestrian involved in a real car accident, not necessarily
to a pedestrian hit by a car in a load case specified in the STP. This approach would require adjustment
of IPFs to match field observations, if the load case specified by the STP does not represent real-world
accidents. Among the three body regions (thigh, knee and leg) for which injury values are measured
by the aPLI, a preliminary study done by the aPLI TG revealed that a significant discrepancy in the
prediction of the probability of injury is seen specifically with the MCL elongation, due to the upright
lower limb position and fully extended knee specified in the STP, and the lack of consideration of the
influence of muscle tone. For this reason, the aPLI TG decided to adjust aPLI IPFs to match real-world
observations in such a way that injury metrics obtained from an aPLI car test are converted to the
probability of injury in real-world accidents.
IPFs for human that represent the load case specified in the STP were initially developed because
the biomechanical data available in the literature are from experimental studies that used boundary
conditions representing this load case. By applying the TFs, the IPFs for humans were transformed to the
IPFs for the aPLI that predict the probability of injury in the load case specified in the STP. The number
of injuries estimated from the probability of injury predicted by these IPFs was compared to that of the
real-world accidents for each of the injury metrics for the aPLI. Whenever a significant inconsistency
with the field observation was identified, factors responsible for the discrepancy were investigated
using literature review, HBMs and multiple car models. Based on the results of the investigation, the
IPFs for the knee and leg were adjusted to the real-world accident data as described in Annex B. The
adjusted IPFs were further modified to better match the field observations as needed.
A.1.2 Need for standardized IPFs for the aPLI
Due to the lack of the upper body, the legform test in the pedestrian STP allows assessment of knee and
leg regions only, necessitating a different test procedure for the thigh region with a different impactor
and a test protocol. The lack of biofidelity of the upper legform requires a modified legform with upper
body representation. In addition, a large approach angle of a high-bumper car results in significantly
different kinematics of a legform, leading to a need for a different test procedure specifically defined for
such cars and inconsistency of the bumper test.
For these reasons, new car assessment programs have been interested in introducing a modified
legform test capable of replacing the upper legform tests and enhancing injury assessment. In order to
avoid unreasonable increase of the cost to car manufacturer, and therefore to customers, of developing
different car structures and/or protection systems to comply with different requirements, a globally
accepted and harmonized test tool is crucial for future upgrade of a test procedure for consumer
information and regulatory testing. Accurate prediction of injury probability in real-world accidents
from the test results using such an advanced test tool is also vital to take the best advantage of the tool.
A.1.3 Benefits and economic impact of standardized IPFs for the aPLI
Robustness of IPFs is crucial to determine injury assessment reference values (IARVs) suitable for
globally harmonized use. When investigating correlation between human and aPLI impact responses to
develop IPFs based on biomechanical data, this requires consideration of various shapes and stiffness
distributions of front-ends of cars, along with multiple different HBMs that have been well validated
against biomechanical data. An international effort provided by the experts of the aPLI TG has allowed
incorporation of various impact configurations from a variety of combinations of cars in different
markets and HBMs.
With regard to the benefits, globally harmonized and scientifically valid IPFs for the aPLI are essential
to determine reasonable IARVs for future regulations and/or new car assessment programs. It is
also crucial that the IPFs predict the probability of injury in a load case representative of real-world
accidents, not necessarily the load case specified in a test procedure in case it is not representative.
Determination of reasonable IARVs would ensure effectiveness of pedestrian safety measures of cars
in real-world accidents, and eliminate potentially wasteful and misleading efforts of car manufacturers
needed to develop cars to comply with the requirements imposed by unreasonable IARVs. This also
reduces costs to consumers, along with the reduction of social costs due to injuries to pedestrian lower
limbs.
A.1.4 Survey and general differences from previous IPFs
The IARVs for the FlexPLI have been determined for References [1] and [2] by the informal group on
UN GTR No.9 Phase-2 under the UN working party on passive safety (GRSP). Due to the lack of globally
accepted IPFs for the FlexPLI, the discussion to determine IARVs has been a big challenge, involving
proposals from different rationale and concepts. In such a situation, both technical and political
discussions were mixed up, making it difficult to clarify benefits in a consistent manner. This has led to
a strong need for a globally accepted and scientifically valid IPFs for the aPLI, which is anticipated to be
used in future regulatory and consumer information testing.
In the course of the discussion on the FlexPLI IARVs, IPFs for the FlexPLI have been developed using
[6]
biomechanical data . Although a similar methodology to that used to develop IPFs for the aPLI was
used, there are some limitations due to the statistical method used and limited robustness of the TFs.
A combination of geometric scaling of injury values and a univariate survival model was applied for the
FlexPLI, while a multivariate survival model used for the aPLI defined injury probability as a function
of the height, the weight and the age of a pedestrian to enhance accuracy of prediction models. When
developing TFs for the FlexPLI, impact simulations were conducted using one single HBM and simplified
car models (SCMs) representing geometric and stiffness characteristics of old sedans at one single
impact location. The robustness of TFs for the aPLI has been significantly enhanced by using multiple
HBMs and RCMs representing various types of cars in a global fleet, along with impact simulations at
multiple impact locations. In addition, the relevance of the IPFs for FlexPLI were not validated against
accident data, while the IPFs for the aPLI were adjusted to real-world observations whenever needed.
The enhancement in the statistical method and the robustness of TFs, along with the adjustment of the
IPFs to real-world accident data, would lead to more accurate prediction of the probability of injury to
a pedestrian lower limb when involved in a car-pedestrian accident, and consequently contribute to
further improvement in pedestrian safety performance of cars in a global market.
A.1.5 Summary of development process of IPFs for the aPLI
The following summarizes some of the milestones in the aPLI project, in order to describe the
international development and consensus process.
— June 2015: The aPLI TG was established by ISO TC 22/SC 36/WG 5 and WG 6.
— November 2017: At the 6th aPLI TG meeting, it was recognized by aPLI TG members that the paired
[4]
test method (ISO/TR 12350:2013) is not applicable to develop IPFs for the aPLI because of the lack
of sufficient biomechanical data, and two alternatives, the TF method and the EE method as detailed
in Annex A.2.2.2.2 and Annex A.2.2.2.3, respectively, were proposed by some of aPLI TG members.
— May 2018: The idea of a virtual paired test was proposed by some of aPLI TG members at the 7th
aPLI TG meeting to evaluate the two methods to develop IPFs for the aPLI.
— June 2019: The aPLI TG decided at its 9th meeting in London to use the TF method based on the
results of the virtual paired test and the comparison of the number of biomechanical data available
for each of the two methods.
— June to December 2020: Due to the pandemic situation of COVID-19, aPLI TG activity was put on
hold, and official discussions on the development of IPFs for human and TFs have been suspended.
— February 2021: The aPLI TG activity was restarted and it was endorsed to adjust the IPFs to real-
world accident data. A draft version of ISO/TS 20459 was submitted to the aPLI TG for review.
— March to August 2021: Due to continued pandemic of COVID-19, aPLI TG activity was put on hold
again.
— August 2021: The aPLI TG activity was officially restarted.
— March 2022: A modified draft version of ISO/TS 20459 incorporating the IPFs for the aPLI was
accepted by the aPLI TG and submitted to ISO TC22/SC36/WG6 for their review.
— May 2022: The draft version of ISO/TS 20459 was further modified based on the comments from
WG6 and was approved by WG6. The document was immediately subjected to a ballot of ISO TC22/
SC36.
— July 2022: The proposed ISO/TS 20459 was approved by ISO TC22/SC36 with some requests for
modification.
Organizationally, at every stage, the aPLI TG attempted to include all interested parties involved in
the pedestrian safety field worldwide. In particular, 28 aPLI TG web meetings were held in between
face-to-face meetings to facilitate discussions and consensus development by all participants. All the
development activities were conducted at the expense of participating parties. Collaborative CAE
studies to develop TFs were conducted by contributors from Germany, France, the United States and
Japan.
A.2 Methodology to develop IPFs for the aPLI
A.2.1 General
Any assessment of car safety performance shall provide information relevant to the probability of
injury sustained by the victims involved in real-world car accidents, not in laboratory car crash tests.
In general, testing and assessment protocols are designed to provide such information. In the case of
the assessment of safety performance against pedestrian lower limb injury, the aPLI TG recognized
a large discrepancy between the two load cases, specifically with the MCL elongation. Once IPFs for
the aPLI consistent with the STP are developed, the relevance of the IPFs against accident data shall
be evaluated. The IPFs shall be adjusted to represent injury probability in real-world accidents when
inconsistency is found to be significant.
IPFs for the aPLI were developed by the following steps. First, human biomechanical data were collected
to develop IPFs for the aPLI. The results of the experiments performed under the conditions similar to
the load case specified in the STP were used. The conditions included regions of the lower limb, specific
injuries reproduced, relevant injury metrics to predict such injuries, and specific load cases such as
the loading rate, loading direction and lower-limb posture. Second, a methodology to develop IPFs for
the aPLI in a load case specified in the STP was determined. This step was one of the most challenging
one among all the steps employed. Due to the lack of sufficient biomechanical data available, it was
impossible to use the method defined by ISO for the WorldSID dummy (paired test method described
in ISO/TR 12350), and an alternative method needed to be defined. As a result of intensive discussion
by aPLI TG members, it was eventually decided to use the TF method. Third, IPFs for the aPLI were
developed by using the TF method. In the TF method, IPFs for human are initially developed, and then
converted to those of the aPLI in the load case specified by the STP by applying TFs obtained from
a regression analysis of maximum values of injury metrics between aPLI and human. Finally, the
relevance of the IPFs was assessed against accident data and the IPFs were adjusted to better represent
real-world accident data whenever needed.
More details with regard to the methodology to develop IPFs for the aPLI can be found in the following
sections.
A.2.2 Determination of methodology to develop IPFs for the aPLI in a load case specified
in the STP as functions of aPLI injury metrics
A.2.2.1 General
The paired test method used by ISO to derive IPFs for WorldSID (ISO/TR 12350) relates injury
occurrence to human bodies to injury values from the dummy by performing paired (human and
dummy) tests in the same impact configuration. In case of a pedestrian, however, full-scale impact test
data using human subjects are scarce, and therefore an alternative methodology to develop IPFs for the
aPLI needed to be determined. This resulted in proposals of two alternative methods at the aPLI TG
(TF method and EE method), and an extensive study was conducted to make a final decision. As both
methods involve pros and cons, a new idea of investigating virtual IPFs was proposed that virtually
generates biomechanical data using parametric HBMs that incorporate variability of material property
of human tissues to allow comparison of IPFs from the two proposed methods against those determined
by the paired test method. The results of this investigation along with the difference of the number of
data that can be used in each of the two proposed methods resulted in the choice of the TF method at
the aPLI TG.
A.2.2.2 Virtual IPF
A.2.2.2.1 Parametric HBM
Due to the lack of a sufficient number of full-scale pedestrian impact test data, full-scale impact test data
using human subjects were virtually generated by incorporating variability of the material property
of human tissues in a validated HBM (References [7], [8] and [9]) to create parametric HBMs with
different material property. Since the material property of human tissue is generally characterized by
a constitutive equation that describes the relationship between the stress and the strain, three levels
(mean, upper and lower limits) were set for each, resulting in a total of nine parametric HBMs. Assuming
that the major contributors to the lower limb injury metrics (femur and tibia bending mom
...