eligible for inclusion. The CDC of the most significant event was used to initially select frontal
and frontal-oriented side impact crashes for analysis according to the following criteria:
General Area of
Direction of Force
Elements of the CDC coding are described in SAE J224. The choice of which combinations of codes is determined
by NHTSA. See DOT HS 811 522.
The Frontal Impact Taxonomy (FIT) uses the CDC, crush profile, principal direction of
force (PDOF), and vehicle class-specific geometry indicators
to identify and classify frontal
crash types within the broad set of crashes described above based on the amount of overlap and
the angle (obliquity) of the impact. This approach was developed to more comprehensively
identify small overlap crashes, which had been identified as a potential area for frontal impact
Occupant inclusion requirements for the frontal target
population consisted of belt-restrained occupants, who were not completely ejected, and who
sustained an AIS 2+ injury or were killed. The seat positions and ages considered are
Outboard only (11,13) 13+
See SAE J224, March 1980, Collision Deformation Classification for a guide to the acronyms used here.
These are generic dimensions, by vehicle class, that are used as a guide for determining whether the damage is
small overlap or not. See Bean, J., Kahane, C., Mynatt, M., Rudd, R., Rush, C., & Wiacek, C., National Highway
Traffic Safety Administration, “Fatalities in Frontal Crashes Despite Seat Belts and Air Bags,” DOT HS 811 202,
September 2009 for more detail.
Bean, J., Kahane, C., Mynatt, M., Rudd, R., Rush, C., & Wiacek, C., National Highway Traffic Safety
Administration, “Fatalities in Frontal Crashes Despite Seat Belts and Air Bags,” DOT HS 811 202, September 2009.
The first step in applying the FIT is to identify small overlap crashes based on the CDC
alone for cases with damage described by GAD1 of F and SHL1 of L or R.
That subset of
small overlap crashes is then augmented by the addition of crashes meeting a small overlap
definition based on class-based vehicle geometry and crush. This crush-based assessment looks
at the damage relative to the longitudinal frame rails for cases where the CDC may not indicate a
small overlap impact based on the damage type coded by SHL1 (e.g., when SHL1 is either Y
(left +center) or Z (right+center)). The frontal-oriented side plane impacts with GAD1 of L or R
are examined from a crush perspective relative to vehicle class-specific geometry. In other
words, when certain damage, and impact vector (PDOF) characteristics are met, the crash will be
considered a small overlap frontal crash by the FIT. Frontal crashes not identified as small
overlap at this stage are then classified based on the crush profile relative to the frame rail
locations into left partial overlap, right partial overlap, or narrow center impacts if crush
measures are defined. Remaining frontal crashes are considered full overlap.
After crashes have been classified based on the extent of overlap, they are categorized as
either co-linear or oblique based on the coded PDOF value. All small overlap crashes, even with
0° PDOF angles, are considered oblique to the side of crush based on findings from laboratory
All full overlap and partial overlap crashes with non-zero PDOF angles are
considered oblique. Full overlap crashes with 0° PDOF angle are considered co-linear. Partial
overlap crashes with 0° PDOF angle are divided between oblique and co-linear based on findings
of the study reported by Rudd et al. (2011). In that study, approximately 20 percent of the 0°
Saunders, J. & Parent, D., "Repeatability of a Small Overlap and an Oblique Moving Deformable Barrier Test
Procedure," SAE World Congress, Paper No. 2013-01-0762, 2013.
partial offset cases resulted in oblique occupant kinematics (to the side of crush).
NASS-CDS case weights are apportioned 20 percent to oblique and 80 percent to co-linear for
partial overlap 0° crashes. Note that the narrow center-impact partial overlap crashes are
considered a special category, and will not be further broken into oblique or co-linear groups as
they are not specifically addressed by any of the planned tests. For the purposes of this frontal
target population, the crashes are further restricted to those with PDOF angles between 330° to
0° and 0° to 30°. There are no restrictions on the impacted object or on the model year of the
The data are presented on an occupant basis, so the counts do not correspond to the
number of vehicles meeting a particular crash description. There may be more than one occupant
in a given vehicle. A tree diagram depicting the breakdown of the relevant frontal crash
occupants considered in this analysis is provided in Figure I-1. The weighted 14-year total count
of MAIS 2+ or fatal occupants in each level is shown. Data presented in this analysis have not
been adjusted to account for air bag presence, changes in data collection procedures by case year,
and to match fatality counts from the Fatality Analysis Reporting System (FARS). The counts
presented are therefore only indicative of relative contributions – actual counts may differ.
Table I-1 shows counts of the occupants further broken down by MAIS 2+, MAIS 3+, or
fatal and by seat row. Note that some fatally-injured occupants do not have injury data coded,
and are therefore not represented in the MAIS 2+ or 3+ columns. This leads to small differences
in calculated totals from Table I-1 and Figure I-1. Another difference between the counts shown
Rudd, R., Scarboro, M., & Saunders, J., “Injury Analysis of Real-World Small Overlap and Oblique Frontal
Crashes,” The 22nd International Technical Conference for the Enhanced Safety of Vehicles, Paper No. 11-0384,
NHTSA is currently investigating this topic, and may revise its approach to categorizing frontal crashes as either
co-linear or oblique.
in Figure I-1 and Table I-1 is that variant impacts, in which the PDOF angle is from the opposite
side of the partial overlap, are merged into the “Other” category due to their unique occupant
kinematics characteristics. Partial overlap crashes where the angle of obliquity is on the same
side as the crush are considered coincident.
Figure I-1: Breakdown of MAIS 2+ or fatal frontal crash occupants by overlap and obliquity.
Counts represent weighted totals from 2000-2013 NASS-CDS.
Table I-1. Distribution of total weighted occupants for the fourteen year period by crash type
(overlap) and obliquity for MAIS 2+, 3+, and fatal severity levels.
MAIS 2+ MAIS 3+
MAIS 2+ MAIS 3+
Halloway, D., Pintar, F., Saunders, J., & Barsan-Anelli, A. (2012) “Classifiers to Augment the CDC System to
Distinguish the Role of Structure in a Frontal Impact Taxonomy.” SAE International Journal of Passenger Cars –
Mechanical Systems, 5(2):778-788.
* Includes small and moderate overlap crashes with variant obliquity (e.g. left small overlap with right oblique
PDOF angle). Source: NASS-CDS (2000-2013)
With left and right partial overlap broken out into co-linear and coincident groups, the
next step is to look at co-linear versus oblique crashes. The counts in Table I-1 are combined into
co-linear full overlap, oblique, and co-linear moderate overlap groups and annualized by dividing
by the number of case years (14) included in the analysis. It is important to note that Table I-2
does not distinguish between left and right oblique crashes – they are pooled together at this
Table I-2. Distribution of occupants by crash obliquity for MAIS 2+, 3+, and fatal severity levels
(annualized unadjusted occupants counts)
Other frontal includes variant impacts and crashes that cannot be categorized due to missing data. Source: NASS-
Left oblique and right oblique crashes are similar in that the occupants’ trajectories are
not straight forward relative to the vehicle interior, but the side of obliquity results in the near
side and far-side occupants experiencing different conditions (a driver would be considered a
near-side occupant in a left oblique crash while the right front passenger would be a far-side
occupant). Left oblique crashes represent a greater proportion of the oblique crashes, and Table
I-3 excludes the right oblique crashes (although 80% of the 0° right moderate overlap crashes
have been accounted for in the co-linear full overlap category).
Table I-3. Distribution of occupants in left oblique and co-linear frontal crashes for MAIS 2+, 3+,
and fatal severity levels (annualized unadjusted occupants counts)
Source: NASS-CDS (2000-2013)
Applying the 80/20 rule previously described for the 0° left moderate overlap crashes
leads to the counts shown in Table I-4, which shows the annualized target population for co-
linear and left oblique frontal crashes. A graphical depiction of the distribution of MAIS 2+
counts is shown in Figure I-2. The counts shown are annualized, unadjusted counts, and
represent the number of MAIS 2+, 3+, or fatal occupants in each crash and obliquity group.
Table I-4. Distribution of occupants in left oblique and co-linear frontal crashes for MAIS 2+, 3+,
and fatal severity levels after redefining the dataset using NHTSA’s approach on categorizing
For the co-linear moderate overlap crashes, 20% were assigned to their respective oblique category with the
remaining 80% being assigned to the co-linear category. Source: NASS-CDS (2000-2013)
Documents you may be interested
Documents you may be interested