[Federal Register Volume 73, Number 134 (Friday, July 11, 2008)]
[Notices]
[Pages 40016-40050]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: E8-15620]



[[Page 40015]]

-----------------------------------------------------------------------

Part II





Department of Transportation





-----------------------------------------------------------------------



National Highway Traffic Safety Administration



-----------------------------------------------------------------------



Consumer Information; New Car Assessment Program; Notice

Federal Register / Vol. 73, No. 134 / Friday, July 11, 2008 / 
Notices

[[Page 40016]]


-----------------------------------------------------------------------

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

[Docket No. NHTSA-2006-26555]


Consumer Information; New Car Assessment Program

AGENCY: National Highway Traffic Safety Administration (NHTSA), 
Department of Transportation (DOT).

ACTION: Final decision notice.

-----------------------------------------------------------------------

SUMMARY: On January 25, 2007, NHTSA published a notice announcing a 
public hearing and requesting comments on an agency report titled, 
``The New Car Assessment Program (NCAP) Suggested Approaches for Future 
Program Enhancements.'' This notice summarizes the comments received 
and provides the agency's decision on how it will improve the NCAP 
ratings program.
    For model year (MY) 2010, the agency will make changes to its 
existing front and side crash rating programs. For the frontal crash 
test program, NHTSA will maintain the 35 mph (56 kmph) full frontal 
barrier test protocol but will update the test dummies and associated 
injury criteria used to assess and assign a vehicle's frontal impact 
star rating. For side impact, NHTSA will maintain the current moving 
deformable barrier test at 38.5 mph (63 kmph) but will update that test 
to include new side impact test dummies and new injury criteria that 
are used to assign a vehicle's side impact star rating. Additionally, 
vehicles will also be assessed using a new pole test and a small female 
crash test dummy.
    For rollover, the agency will continue to rate vehicles for 
rollover propensity, but will wait to update its rollover risk model to 
allow for more real-world crash data of vehicles equipped with 
electronic stability control.
    Also for MY 2010, the agency will implement a new ratings program 
that will rate vehicles on the presence of select advanced technologies 
and establish a new overall Vehicle Safety Score that will combine the 
star ratings from the front, side, and rollover programs.
    Finally, for the agency's vehicle labeling program, we are 
announcing that the side score, rather than being based only on the 
moving deformable barrier test, will be based on the combination of the 
moving deformable barrier test and the pole test. Additionally, the 
agency will initiate rulemaking to include the new overall 
crashworthiness rating on the Monroney label.

DATES: These changes to the New Car Assessment Program are effective 
for the 2010 model year.

FOR FURTHER INFORMATION CONTACT: For technical issues concerning the 
enhancements to NCAP, contact Mr. Nathaniel Beuse or Mr. John Hinch. 
Telephone: (202) 366-9700. Facsimile: (202) 493-2739. For legal issues, 
contact Dorothy Nakama, NHTSA Office of Chief Counsel, Telephone (202) 
366-2992. Facsimile: (202) 366-3820. You may send mail to these 
officials at: The National Highway Traffic Safety Administration, 
Attention: NVS-010, 1200 New Jersey Avenue, SE., Washington, DC 20590.

SUPPLEMENTARY INFORMATION:

I. Introduction
II. Summary of Request for Comments
    A. Frontal NCAP
    B. Side NCAP
    C. Rollover NCAP
    D. Rear Impact
    E. Crash Avoidance Technologies
    F. Presentation and Dissemination of NCAP information
    G. Manufacturer Self-Certification
III. Summary of Comments
    A. Frontal NCAP
    1. Impact Protocol
    2. Test Dummies (in the Front Seating Position)
    3. Injury Criteria
    4. Test Speed
    B. Side NCAP
    1. Oblique Pole Test (Test Dummies and Implementation Time)
    2. Moving Barrier Protocol (Test Speed, Test Dummies, and Injury 
Criteria)
    C. Rollover NCAP
    1. Rollover Risk Model
    2. Dynamic Rollover Structural Test
    D. Rear Impact
    1. Basic Information
    2. Links to the IIHS
    3. Dynamic Test
    E. Crash Avoidance Technologies
    1. Program Implementation
    2. Selected Technologies
    3. Rating System
    F. Presentation of NCAP Information
    Combined Crashworthiness Rating
    G. Manufacturer Self-Certification (of NCAP Results)
    H. Other Suggestions
IV. Discussion and Agency Decision
    A. Frontal NCAP
    B. Side NCAP
    C. Rollover NCAP
    D. Rear Impact
    E. Crash Avoidance Technologies
    F. Presentation and Dissemination of Safety Information
    G. Manufacturer Self-Certification
    H. Other Recommendations
    I. Monroney Label
V. Conclusion
Appendix A
Appendix B
Appendix C
Appendix D

I. Introduction

    The National Highway Traffic Safety Administration (NHTSA) is 
responsible for reducing deaths, injuries, and economic losses 
resulting from motor vehicle crashes. One way in which NHTSA 
accomplishes this mission is by providing consumer information to the 
public. NHTSA established the New Car Assessment Program (NCAP) in 1978 
in response to Title II of the Motor Vehicle Information and Cost 
Savings Act of 1972. Through NCAP, NHTSA currently conducts tests and 
provides frontal and side crash, and rollover ratings and communicates 
the results using a five-star rating system. With this information, 
consumers can make better-informed decisions about their purchases. In 
turn, manufacturers respond to the ratings by voluntarily improving the 
safety of their vehicles beyond the minimum Federal safety standards.
    For MY 1979, when the agency began rating vehicles for frontal 
impact safety, fewer than 30 percent of vehicles tested would have 
received the top ratings of 4 or 5 stars for the driver seating 
position.\1\ By comparison, for MY 2007, 98 percent of vehicles 
received 4 and 5 stars in the frontal NCAP rating for that same seating 
position. Equally impressive is that while it took almost 30 years to 
reach this level for frontal NCAP performance, the more recent NCAP 
programs, like side and rollover NCAP, have started reaching this level 
of safety performance at a pace that can be measured in years rather 
than decades. The agency believes that consumers continue to consider 
safety in their purchasing decisions and are demanding ever-increasing 
levels of safety.
---------------------------------------------------------------------------

    \1\ NHTSA began using stars in model year 1994. See 69 FR 61072, 
Docket No. NHTSA-2004-18765.
---------------------------------------------------------------------------

    Similarly, recent advances in crash avoidance technology offer a 
new opportunity for NCAP to further enhance its ability to inform 
consumers about new systems and encourage them to purchase systems that 
NHTSA has found to be effective in improving safety.
    On January 25, 2007 NHTSA published a notice outlining proposed 
enhancements to the NCAP activities. In this notice, we requested 
comments on any additional actions that the agency could undertake so 
that the program could continue to provide consumers with relevant 
safety information.\2\ These enhancements included new test dummies and 
injury criteria for frontal NCAP, the addition of a new side pole test, 
new test dummies, and new injury

[[Page 40017]]

criteria for side NCAP, an overall summary rating, and a new program to 
promote advanced crash avoidance technologies. Additionally, the notice 
announced a March 7, 2007 public hearing to allow interested parties 
the opportunity to address the suggested approaches for enhancing the 
program.
---------------------------------------------------------------------------

    \2\ 72 FR 3473, Docket No. NHTSA-2006-26555.
---------------------------------------------------------------------------

    Seventy-six (76) individual comments were received in response to 
the notice and the public hearing.\3\ Commenters offered mixed 
responses to the various proposals for enhancing NCAP; however, most 
commenters commended the agency's initiative to reexamine the program 
and supported the proposed approaches. This notice summarizes comments 
to the January 2007 notice, the March 2007 public hearing, and provides 
the agency's decision on how it will proceed with changes to NCAP.
---------------------------------------------------------------------------

    \3\ This count does not include duplicative or multiple comments 
from the same source.
---------------------------------------------------------------------------

I. Summary of Request for Comments

    In its notice, the agency presented proposals to improve not only 
the program's current front, side and rollover activities, but also 
approaches to improve its information with regards to rear impact, and 
certain crash avoidance (or active safety) technologies such as 
Electronic Stability Control (ESC). NHTSA also outlined alternatives to 
enhance the presentation and dissemination of safety information to 
consumers, and solicited feedback for additional considerations that 
would allow NCAP to remain effective and relevant in improving vehicle 
safety.

A. Frontal NCAP

    NHTSA proposed three approaches to enhance the frontal NCAP. The 
first approach was to maintain the current 35 mph (56 kmph) test 
protocol with a 50th percentile male Hybrid III dummy, but to account 
for injuries to the knee/thigh/hip (KTH) complex. This would be 
accomplished by including a new injury criterion into the formula used 
to calculate the frontal NCAP rating for the driver and front passenger 
seating positions. Second, while keeping the test protocol the same, 
the agency considered determining whether injury measures obtained 
below the knee using the Denton or Thor-Lx dummy legs are predictive of 
real-world injuries. Last, the agency considered evaluating vehicles 
based on a lower test speed.

B. Side NCAP

    To enhance its side impact safety ratings, the agency presented two 
approaches for consideration. NHTSA proposed continuing to rate 
vehicles using the moving deformable barrier test protocol but would 
also encourage manufacturers to provide better head and pelvis 
protection by including the side impact pole test and the new test 
dummies recently finalized in Federal Motor Vehicle Safety Standard 
(FMVSS) No. 214 ``Side Impact Protection'' prior to the performance 
requirements being fully phased-in.\4\ Furthermore, the agency proposed 
research that would focus on the assessment of the injury mechanisms in 
a fully equipped side impact air bag fleet. The purpose of the research 
would be to evaluate how serious injuries occur in the new fleet and to 
develop test procedures to reflect these impact conditions. The outcome 
of this research could lead to a new barrier test protocol (which could 
include increased test speed and different barrier characteristics).
---------------------------------------------------------------------------

    \4\ 73 FR 32473, Docket No. NHTSA-2008-0104. On June 9, 2008 the 
agency responded to petitions for reconsideration of the final rule, 
changing the effective date of the pole test. Now, with certain 
exceptions, all vehicles have to meet the upgraded pole test by 
September 1, 2014.
---------------------------------------------------------------------------

C. Rollover NCAP

    To enhance its rollover program, the agency indicated that it would 
continue tracking the rollover rate and the single vehicle crash rate 
of vehicles equipped with ESC to create a new rollover risk model.

D. Rear Impact

    Currently, NHTSA does not provide consumer information on rear 
impacts. However, NHTSA is aware of recent research suggesting that 
consumers are concerned about rear crashes. As such, the agency 
proposed two approaches. First, NHTSA proposed that it could provide 
consumers with basic information on rear crashes such as safe driving 
behavior, proper adjustment of head restraints, real-world safety data 
by vehicle classes, and links to the Insurance Institute of Highway 
Safety (IIHS) rear impact test results. Second, as a longer term 
approach, the agency proposed that a dynamic test, which addresses 
those injuries not covered by the agency's current standards, could be 
investigated and incorporated into the ratings program.

E. Crash Avoidance Technologies

    Technologies such as ESC, forward collision warning (FCW), lane 
departure warning (LDW) and crash mitigation systems have been 
developed and are being offered in the current vehicle fleet. Some of 
these technologies have shown effectiveness in reducing the number of 
relevant crashes in Department of Transportation (DOT)-sponsored field 
operational tests.\5\ Research by the agency and others has shown that 
consumers are generally unaware of these technologies or their 
potential safety benefits. As a result, the agency believed that NCAP 
should be used to better highlight those beneficial technologies to 
consumers and sought to establish a new ratings program that evaluated 
vehicles on the presence of proven crash avoidance technologies. Based 
on technical maturity, fleet availability, and available effectiveness 
data, NHTSA identified three technologies that fit these criteria. 
These technologies are ESC, LDW, and FCW.
---------------------------------------------------------------------------

    \5\ See 72 FR 3475, Docket No. NHTSA-2006-26555.
---------------------------------------------------------------------------

    NHTSA proposed two possible approaches and illustrated a possible 
implementation of the program with an A, B, C letter grade system. 
First, the agency proposed that each of the technologies would have 
equal weight. For example, if a vehicle had only one technology, it 
would receive a C; whereas, another vehicle that had all three 
technologies would receive an A. Approach two would attempt to quantify 
a technology's real-world benefits by taking into account the target 
population and anticipated effectiveness of the technology to decide 
whether a particular type of technology would be given more weighting 
than another and thus prompt a higher score. For example, in this 
scheme, if ESC was found to be more effective than lane departure, a 
vehicle equipped only with ESC could receive a B versus a vehicle 
equipped only with lane departure warning which would receive a C 
rating.
    It was further stated that this second approach could be expanded 
into a more comprehensive performance-based crash avoidance rating. As 
the technologies evolved and as the agency gathered more information 
related to various versions of these technologies and their associated 
safety effectiveness, NHTSA proposed that a safety score (i.e., star 
rating) on individual technologies could then be developed (e.g., 
different version of ESC might yield different performance results and 
thus a different star rating).

F. Presentation and Dissemination of NCAP Information

Combined Crashworthiness Rating
    Several NHTSA-sponsored research reports and consumer surveys, as 
well as a Government Accountability Office and a National Academy of 
Sciences review of NCAP, have all pointed to the public's desire for a 
summary safety rating. Similarly, other consumer information programs 
around the world

[[Page 40018]]

such as the IIHS, Japan NCAP, and EuroNCAP use summary ratings that 
combine their respective crashworthiness tests. The agency proposed two 
summary crashworthiness rating concepts. In both concepts, the existing 
rollover rating was not included in the calculation of the overall 
summary rating, and star rating boundaries would have to be developed 
for both individual crash tests and the overall summary rating.
    The first approach computed the overall crashworthiness rating by 
first averaging the driver and right front passenger dummy injury 
results from the frontal crash mode into a single star rating. The same 
would be done for the seating positions in the side crash mode to 
compute the overall side crash rating. To compute the overall 
crashworthiness rating, the overall frontal and the overall side impact 
performance would be combined by using weighting factors obtained from 
real-world data (i.e. the National Automotive Sampling System (NASS)). 
Each individual total (overall front and overall side) would be 
weighted by that crash mode's contribution to the total injuries 
occurring in the real-world.
    The second approach computed the overall crashworthiness rating by 
normalizing the seating positions for each individual crash mode (front 
and side) using the Injury Assessment Reference Values (IARVs) 
established for that dummy, body region, and crash mode. Using the NASS 
data, these normalized values would then be multiplied by the 
occurrence of that injury in the real-world. Body injury regions that 
are coded by NASS but are not measured by the dummy and/or not selected 
by NHTSA for inclusion in the rating would be equally distributed among 
the remaining body regions.
Presentation of Safety Information
    As the consumer's use of the Internet for vehicle safety 
information has grown, so has the need to consolidate and better 
present NCAP vehicle safety information to consumers on http://www.safercar.gov. The four approaches proposed by the agency were: (1) 
Developing other topical areas under the Equipment and Safety section 
of the Web site; (2) redesigning the Web site to improve organization; 
(3) improving search capabilities on the Web site; and, (4) combining 
agency recall and ratings database information.

G. Manufacturer Self-Certification

    In addition to NHTSA's proposed suggestions in the notice the 
agency also sought comment at the public hearing on whether or not 
manufacturers should be allowed to conduct and publish their own NCAP 
ratings via a self-certification process. We indicated that such an 
approach would be one way to improve not only the timeliness of NCAP 
ratings but also to increase the number of vehicles rated by the 
agency.

III. Summary of Comments

    This section provides a brief summary of the seventy-six (76) 
comments submitted to the docket by vehicle manufacturers, safety 
advocates, public health groups and the general public in response to 
the notice and the public hearing.\6\ It should be noted that comments 
unique to the public hearing are stated as such.
---------------------------------------------------------------------------

    \6\ These submissions are available at http://www.regulations.gov in Docket No. NHTSA-2006-26555.
---------------------------------------------------------------------------

A. Frontal NCAP

    Comments regarding NHTSA's frontal program are grouped into four 
categories: Impact Protocol, Test Dummies (in the Front Seating 
Position), Injury Criteria and Test Speed.
1. Impact Protocol
    The Alliance of Automobile Manufacturers (Alliance), Automotive 
Occupant Restraints Council (AORC), Toyota Motor North America, Inc. 
(Toyota), BMW of North America (BMW), Fuji Heavy Industries USA, Inc. 
(Subaru) and Volkswagen of America, Inc. (VW) supported the retention 
of the current frontal crash test protocol at 35 mph (56 kmph). 
Consumers Union and Public Citizen suggested adding an offset frontal 
crash test rating, which Public Citizen believed would be far more 
useful in assessing the structural integrity of different vehicle 
models. Likewise, Toyota also encouraged NHTSA to investigate ways to 
include information on offset collision conditions in its NCAP program. 
Toyota explained that their investigation of National Automotive 
Sampling System Crashworthiness Data System (NASS-CDS) data showed that 
an overwhelming majority of frontal crashes occur in either the full 
overlap or offset condition. They believed that vehicle performance 
assessed in the offset condition should yield relevant improvements in 
safety technology and provide considerable benefit.
    IIHS and Subaru recommended the addition of a frontal pole test to 
address significant injuries resulting from impacts with narrow 
objects. IIHS asserted that offset tests more closely simulate impacts 
with narrow objects than do full-width tests, and that a narrow-object 
NCAP test could have an important impact on real-world vehicle 
crashworthiness, and would give consumers a wide range of results to 
inform their purchasing decisions. Subaru suggested that NHTSA should 
study and possibly propose a frontal pole test for inclusion into NCAP 
if the frequency of frontal crashes with narrow objects is high. 
However, General Motors North America (GM) asserted that a pole test is 
unlikely to result in significant change or further improvement in 
structural stability and resultant injury reduction. They stated that 
research in this area may yield only limited or incremental gains in 
injury mitigation, and that the public interest is likely to be better 
served by channeling resources into areas that could produce greater 
societal benefit.
2. Test Dummies (in the Front Seating Position)
    With regard to test dummies, the Alliance stated that test dummies 
in frontal NCAP should be the same as those in FMVSS No. 208. 
Additionally, GM, AORC, Consumers Union and the Alliance supported the 
use of the 5th percentile female Hybrid III dummy in the right front 
passenger position. GM provided NASS data which suggested that small 
females were over-represented (with regard to serious injuries) in the 
right front passenger seating position. GM also suggested that in the 
future, the 5th percentile female dummy should be used in both seating 
positions to optimize safety. AORC asserted that the substitution of 
the 5th female for the 50th percentile male would demonstrate a broader 
population range of protection since some data has been shown which 
suggests that the weighted frequency of serious and fatal injuries to 
women is greater than to men in the right front passenger seating 
position.
    Furthermore, Consumers Union asserted that the agency should 
investigate using the 5th percentile female and 95th percentile male 
dummies to evaluate NCAP tests for all sizes of vehicle occupants. 
Subaru supported the continued use of 50th percentile adult male 
dummies in both front seating positions indicating that this was more 
representative of real-world occupants. Subaru also asserted that 
additional tests with other dummies, such as the 5th percentile adult 
female, should be done only if well supported by real-world data.
3. Injury Criteria
    Most vehicle manufacturers agreed that NHTSA should develop and 
incorporate a KTH injury criterion into

[[Page 40019]]

the NCAP frontal rating. They noted that a KTH assessment would drive 
vehicle countermeasures that could mitigate lower leg injuries and also 
yield important information relevant to vehicle design. Likewise, 
adding KTH and/or lower leg injury criteria to the NCAP rating protocol 
could expand the usefulness of the NCAP system by addressing the 
societal cost of Abbreviated Injury Scale (AIS) 2+ injuries. The 
Alliance, Autoliv, Consumers Union and IIHS also supported NHTSA's 
efforts to incorporate a KTH injury criterion into the frontal program. 
However, IIHS urged the agency to concentrate its research tests on 
serious injuries and fatalities in frontal impacts to encourage more 
protective vehicle design. Additionally, Autoliv stated that although a 
reduction in KTH injuries would have a significant impact on societal 
cost, they believed that it would have little effect in reducing 
fatalities.
    Nissan North America (Nissan) stated that the agency should 
consider a KTH assessment only after further study is conducted. 
Instead, Nissan urged NHTSA to harmonize knee and thigh injury values 
with those required in Japanese and European regulations. Likewise, the 
Association of International Automobile Manufacturers (AIAM) did not 
believe that the agency should move expeditiously to include a KTH 
criterion in the current frontal NCAP program since the agency had 
identified crashes of lower test speed as the primary concern regarding 
leg injuries. They recommended that NHTSA present the analysis and 
results of their KTH research for public comment prior to including a 
KTH criterion in the frontal program.
    For lower leg assessments, several commenters suggested that 
additional research was needed to determine whether injury measures 
obtained below the knee were predictive of real-world injury. GM noted 
that adding a femur load injury criterion to frontal NCAP would drive 
many of the same vehicle countermeasures that would mitigate lower leg 
injuries.
    With regards to what anthropomorphic test device (ATD) could be 
used for these new criteria (KTH and lower leg), Honda specifically 
stated that a KTH assessment would be possible using the Denton dummy 
leg. For injuries to the lower leg (below the knee), Honda, Subaru, 
Nissan, and Volvo Cars of North America, LLC (Volvo), suggested that 
the agency adopt the Thor-Lx legs in the future. The Alliance did not 
support the introduction of either the Denton or Thor-Lx legs unless 
they were included in FMVSS No. 208. Furthermore, VW believed that 
these test devices must be validated, and the applicable injury 
criteria and rating must be verified for correlation with real-world 
safety.
    Some commenters suggested that all injury criteria incorporated in 
FMVSS No. 208 (beyond head injury criteria and chest acceleration 
criteria) should also be included in frontal NCAP. Specifically, Honda, 
Ford, GM, the Alliance, and Autoliv supported the inclusion of a chest 
deflection criterion into the frontal NCAP rating based on NASS-CDS 
data indicating a substantial number of injuries to ribs and internal 
organs resulting in AIS 3+ or higher severity injuries. However, Honda 
stated that the current chest deflection calibration procedure may not 
be appropriate to assure that chest deflection measurements are 
accurate enough to provide useful data. GM and the Alliance recommended 
including a chest compression criterion into frontal NCAP. The Alliance 
urged NHTSA to conduct research on neck (tension) injury criteria 
before including it into frontal NCAP. However, GM suggested that the 
agency add neck injury criteria to frontal NCAP since these criteria 
are already measured by the Hybrid III dummies and included in FMVSS 
No. 208.
4. Test Speed
    With regards to adopting a lower test speed, the Alliance, GM and 
Volvo agreed with NHTSA's analysis and supported the agency's proposal 
to conduct more research on lower test speeds. However, VW questioned 
whether lower speed crashes represented a greater risk of occupant 
injury than the current NCAP test procedure. Therefore, VW as well as 
the Alliance believed that an additional test in frontal NCAP would add 
significant expense and strain on available resources without any 
commensurate advantages or benefit.
    Subaru asserted that they did not support adding low speed bumper 
tests to frontal NCAP since those tests would overlap with existing 
IIHS tests.
    Two individual commenters, Mr. Dainius Dalmotas and Dr. Harold 
Mertz stated that a full vehicle crash test designed to promote 
enhanced chest protection in low-to-moderate speed frontal crashes 
would be most promising since the vast majority of serious and fatal 
injuries among belted drivers occur at collision speeds of 25 mph (40 
kmph) or less. They also asserted that incentives to promote improved 
safety in low-to-moderate speed frontal impacts were lacking and could 
be addressed through NCAP.
    At the public hearing, Consumers Federation of America (CFA) and 
the Center for Auto Safety (CAS) suggested that NHTSA increase test 
speeds and challenge manufacturers to post the highest speed at which 
their vehicles are tested, in order to differentiate amongst the 
performance of vehicles. However, the Alliance, Consumers Union, AIAM 
and Subaru opposed a higher speed test for frontal NCAP. The Alliance 
stated that field data did not show the need for higher test speeds. 
AIAM and Consumers Union did not believe that increasing crash test 
speeds would benefit the overall safety of occupants; but rather, it 
could cause vehicles to become stiffer. Subaru asserted that a higher 
speed test is not representative of the vast majority of fatal crashes, 
does not enhance NCAP's consumer information goals, and risks 
increasing vehicle aggressiveness.

B. Side NCAP

    Comments regarding NHTSA's side program are divided into the 
following categories: Oblique Pole Test (Test Dummies and 
Implementation Time), Moving Barrier Protocol (Test Speed, Test 
Dummies, and Injury Criteria), and Side NCAP Research.
1. Oblique Pole Test (Test Dummies and Implementation Time)
    GM, Subaru, Toyota, the Alliance, and Autoliv agreed with the 
agency's proposal to incorporate an oblique pole test into NCAP. 
However, with regards to adopting the oblique pole test prior to the 
completion of the FMVSS No. 214 pole test phase-in, BMW, Ford, Toyota, 
and the Alliance, asserted that such action would be premature, and 
these commenters suggested that NHTSA adopt the test after the oblique 
pole test had been fully phased-in. Furthermore, Subaru suggested that 
3 years be allowed after the agency announced a new test before rating 
vehicles under the new test protocol.
    Toyota explained that they understood NHTSA's intention to use an 
early introduction of the pole test to drive the installation of 
advanced head protection systems (like curtain airbags), but they 
believed that significant benefits in head protection were already 
being realized from the introduction of curtain air bags, which was 
driven by industry's commitment to the industry voluntary compatibility 
requirements.\7\

[[Page 40020]]

Therefore, Toyota recommended additional investigation into whether 
there are merits of an early introduction of an oblique pole test into 
NCAP. Honda recommended adding to the existing side impact test by 
introducing a second side impact test that is similar to the current 
IIHS moving deformable barrier (MDB) test.\8\ Honda suggested that this 
would extend the coverage of NHTSA's side impact testing, be more 
representative of real-world crashes, and help to provide a more 
realistic assessment of a vehicle's crashworthiness in these types of 
two-vehicle collisions.
---------------------------------------------------------------------------

    \7\ IIHS and the Alliance created a voluntary agreement wherein 
automotive manufacturers agreed to improve occupant protection in 
front and side crashes involving cars and light trucks. For front-
to-side impacts, most automakers agreed to design their vehicles to 
meet the head injury performance requirements of NHTSA's FMVSS No. 
201 side-pole test or the IIHS moving deformable barrier test. By 
September 1, 2007, at least half of all new passenger vehicles would 
meet one of the two requirements, and by September 1, 2009 model 
year, all new passenger vehicles would meet the head injury 
requirements of the Institute's moving deformable barrier test.
    \8\ This test would represent an SUV to subject vehicle crash 
(IIHS Side Impact Crash Evaluation test procedure--SICE).
---------------------------------------------------------------------------

    If the agency went forward with an oblique pole test, Subaru 
recommended a side impact assessment based on two tests (the oblique 
pole test and IIHS's MDB test) with head injury criteria and the SID-
IIs dummy, as long as the results could be combined into a single 
rating. BMW and the Alliance suggested that the 5th percentile female 
SID-IIs dummy be used for the driver position in the oblique pole test. 
BMW asserted that the smaller SID-IIs dummy is most appropriate for 
determining the geometric coverage area required for a curtain airbag. 
The Alliance believed that it is appropriate to test only with the 5th 
percentile female dummy in the front seating position because this is a 
very severe test condition, and it would serve to meet the intent of 
NCAP while minimizing additional test burdens on NHTSA and the 
automotive industry.
    Honda, Nissan and VW did not support the inclusion of an oblique 
pole test into side NCAP. Honda believed that introducing an oblique 
pole test would be a temporary measure until the test was fully phased-
in as a requirement for FMVSS No. 214. To comply with the requirements 
of FMVSS No. 214, the head protection benefits of the oblique pole test 
would already have been realized in every vehicle, so there would be 
little practical benefit to consumers as a result of temporarily 
including such a test in NCAP. VW and Nissan, similar to Toyota, stated 
that automobile manufacturers were already committed to front-to-side 
impact protection, and that the addition of a side impact pole test 
would provide no added incentive for the manufacturers to implement 
additional side impact protection. Nissan also believed that 
incorporating the pole test into NCAP is unnecessary to encourage head 
protection in new vehicles.
    IIHS stated that the current NCAP barrier test did not fully 
address the mix of vehicles on the road and that the agency needed to 
improve the existing side impact barrier. IIHS suggested giving greater 
priority to adopting or modifying the IIHS side impact barrier rather 
than incorporating a new oblique pole test. However, GM asserted that 
the pole test is structurally more challenging than the IIHS MDB test, 
and that the IIHS MDB test and the pole test will not necessarily drive 
installation of the same air bag solutions.
2. Moving Barrier Protocol (Test Speed, Test Dummies, and Injury 
Criteria)
    NHTSA proposed a new side NCAP barrier test protocol that would 
include new dummies and additional injury criteria. The Alliance 
supported the maintenance of the current barrier test but they 
suggested a revised, lower test speed of 33.5 mph (54 kmph).
    With regards to the incorporation of new dummies into the side MDB 
test, the Alliance, Subaru, Honda, Nissan, Volvo, and AIAM proposed the 
incorporation of WorldSID into NCAP. Specifically, Volvo and the 
Alliance suggested that the WorldSID dummy should be introduced in 
FMVSS No. 214 and NCAP simultaneously. Honda stated that the WorldSID 
dummy provides excellent biofidelity, and does not present problems 
with rib guide shape that the ES-2re dummy appears to have based on 
their evaluation. AORC believed that the current test dummy does not 
adequately address head injuries, and they encouraged NHTSA to use 
either EuroSID-2 and/or the SID-IIs side impact dummy.
    Volvo recommended that the dummies and injury criteria for the NCAP 
side barrier test procedures be the same as they are for FMVSS No. 214. 
Volvo supported the addition of head injury criteria in the NCAP 
evaluation for the side barrier; however, they would prefer that the 
NCAP criteria limits are set more stringent in order to encourage 
manufacturers to exceed the performance standards outlined in the legal 
requirement. BMW recommended that NHTSA use the ES-2re dummy for the 
driver position in the MDB test because the SID-IIs dummy is already 
included in the MDB test conducted by IIHS, and the biofidelity of the 
SID-IIs dummy in these types of impacts is well understood. GM also 
suggested the ES-2re dummy for the driver position since the most 
frequent occupant, and most frequently injured occupant type at the 
driver position is an adult male.
    Autoliv asserted that the ES-2re dummy should be used for the front 
seating position in both the oblique pole and MDB tests, as this dummy 
represents the largest percentage of front seat occupants. They also 
recommended the SID-IIs dummy for the rear seating position to provide 
information on protection for older children and small adults seated in 
the rear. GM also recommended the SID-IIs dummy for the rear seating 
position because more frail persons tend to sit in the rear, the SID-
IIs dummy is tuned for frail occupants, and placement in the rear will 
import safety improvements across the range of occupants.
3. Side NCAP Research
    As a longer term approach, the agency suggested research into the 
moving barrier test protocol to address injuries and fatalities that 
might occur in vehicles equipped with curtain and side impact air bags. 
The agency indicated this research could lead to a new barrier, an 
increased barrier test speed, and a reevaluation of the impact 
configuration.
    The Alliance, AIAM, Honda and Subaru agreed that NHTSA should 
analyze real-world side impact crashes for vehicles with side curtain 
airbags. However, the Alliance recommended that the agency and 
automotive industry should develop more experience with the new pole 
test and test dummies before considering any increase in test speeds. 
In addition, the Alliance asserted that future research should evaluate 
whether it would be beneficial for NCAP to harmonize with the existing 
IIHS barrier.
    Toyota supported additional research efforts to gain a better 
understanding of the potential for and the necessity of changes to the 
test device and configuration for vehicles equipped with side airbags. 
Furthermore, Toyota stated that questions remain relating to barrier 
characteristics, injury criteria and appropriate ATDs that should be 
researched from relevant field data.\9\
---------------------------------------------------------------------------

    \9\ In particular, Toyota recommended continued investigation 
into previously identified concerns with the performance of the SID-
IIs upper arm, which they believed was not biofidelic and affected 
the thoracic rib response.
---------------------------------------------------------------------------

    Autoliv recommended that NHTSA research increasing the test speed 
and develop a single test that would assess both the head and thorax 
injury protection systems installed in newer vehicles. Autoliv also 
suggested that the adoption of the WorldSID dummy would be suitable if 
incorporated into Part 572 and FMVSS No. 214.

[[Page 40021]]

Additionally, Delphi opposed releasing a new regulation under FMVSS No. 
214 and then promoting a different set of barrier protocols, dummy 
types and injury metrics for side NCAP evaluation since that decision 
could cause misdirection for original equipment manufacturers and 
suppliers.

C. Rollover NCAP

    Comments regarding NHTSA's rollover program are grouped into the 
following categories: Rollover Risk Model and Dynamic Rollover 
Structural Test.
1. Rollover Risk Model
    Most commenters supported the development of a new rollover risk 
model. Several commenters agreed that real-world crash data was 
necessary to develop an effective rollover risk model. Specifically, 
the Alliance, AIAM, the National Automobile Dealers Association (NADA), 
and VW each commented that NHTSA should collect new crash data for 
rollover NCAP. In particular, the Alliance and Ford recommended that 
the agency collect crash data on both ESC and non-ESC equipped vehicles 
to develop a new rollover risk model that better describes rollover 
risk for all vehicles, but also accurately reflects the differences 
between ESC and non-ESC vehicles. Toyota believed that the update to 
rollover NCAP should reflect real-world benefits of ESC on rollover 
risk, and that the rollover rating should be combined (with advanced 
technologies) into an overall crash avoidance rating. AIAM suggested 
that NHTSA consider adjusting a vehicle's rollover risk rating to 
reflect the safety benefits of ESC or adopt some other means of 
communicating those benefits to consumers.
    Recognizing that since such a data collection and analysis cannot 
be completed in the near term, Ford, the Alliance and Volvo suggested 
that in the near term, an additional rollover NCAP star should be 
awarded to those vehicles equipped with an ESC system to recognize the 
benefits of ESC. Specifically, the Alliance recommended that NHTSA 
provide additional information in the form of a footnote on the 
agency's Web site and in the Safer Car brochure that explains the 
benefits of ESC and why these benefits warrant an additional star.
2. Dynamic Rollover Structural Test
    Some commenters encouraged NHTSA to develop a test for structural 
integrity to enhance rollover NCAP. Specifically, Consumers Union, 
Public Citizen and ARCCA Incorporated (ARCCA) urged the agency to 
consider a dynamic test to assess body structure, seat belt design 
(including pretension), side curtain airbags, roof strength, door locks 
and retention, and the retention of window glazing. In particular, 
Public Citizen believed that a rollover NCAP rating should be based on 
a vehicle's ability to resist rollover and to protect occupants in a 
rollover crash. They suggested a rating that included ejection as a 
consideration since this would provide valuable information about a 
vehicle's ability to prevent death or serious injury in a rollover 
crash. Additionally, the rating should measure rollover propensity, as 
well as crashworthiness measures of performance in a rollover crash.
    The Center for Injury Research (CIR) recommended that an NCAP 
rollover test be dynamic and somewhat more severe than a dynamic 
compliance standard. According to CIR, a dynamic test for use as both a 
safety compliance standard and as an NCAP test can and should be 
developed simultaneously with action on the roof crush standard. 
Moreover, CFA and CAS recommended adding a rollover test with 
comparative roof crush tests, while IIHS suggested that NHTSA should 
conduct additional research on roof crush. Bidez and Associates stated 
that a meaningful rollover crashworthiness test must include roof 
deformation, seat belt performance, door opening, and window breakage. 
They emphasized that protection should be assessed for front and rear 
passengers, adults and children, and that the Jordan Rollover System 
(JRS) holds great promise. Conversely, the Alliance, Ford and Nissan 
opposed the use of JRS in NCAP. The Alliance commented, and Ford and 
Nissan stated at the public meeting that there has been no JRS tests 
conducted with an instrumented dummy and therefore, the JRS test 
results cannot be related scientifically to the real-world risk of 
injury in a rollover crash.

D. Rear Impact

    Comments regarding NHTSA's rear impact NCAP activity are divided 
into the following categories: Basic Information, Links to the IIHS, 
and Dynamic Test.
1. Basic Information
    Commenters presented similar views on how NHTSA should provide 
consumers with basic information concerning rear impact crashes in an 
NCAP publication. GM, Toyota, Subaru and VW supported the inclusion of 
information on the proper adjustment and utilization of head restraint 
systems. Additionally, GM supported consumer education that included 
material such as safety tips and safe driving practices.
2. Links to the IIHS
    The IIHS endorsed the agency's proposal and offered their head 
restraint evaluation information for posting on the agency's Web site. 
Toyota believes that the IIHS results are only one way to assess rear 
impact performance, and thus the agency should be cautious and thorough 
when determining what rear impact evaluation should be part of a future 
NCAP evaluation. They also stated that ample consideration should be 
given to passive and active head restraint concepts in order to 
maintain benefits from all design types.
    The Alliance felt that NHTSA's proposal did not seem consistent 
with the principle of the Federal government independently generating 
all NCAP data. Rather, they advocated that the agency should 
investigate further the injury mechanism of whiplash and then choose 
which responses to evaluate based on biomechanics. Similarly, GM 
discouraged NHTSA from implementing this option. According to GM, links 
to the IIHS Web site might imply that NHTSA has given full endorsement 
of IIHS methodology and interpretations, and some consumers may even 
conclude that IIHS is a government agency.
3. Dynamic Test
    The Alliance believed that NHTSA should first evaluate potential 
effectiveness and safety benefits prior to incorporating a rear crash 
rating into NCAP. Consumers Union stated that rear impact whiplash 
injuries are debilitating to those involved and cause a large cost to 
society. Consumers Union recommended that NHTSA look at IIHS's work on 
rear impact testing to determine whether developing NCAP ratings for 
rear impact results would be cost effective. Public Citizen suggested 
that the agency develop a rear-impact crash NCAP rating, especially at 
speeds of 35 to 40 mph (56 to 64 kmph) to improve rear-impact occupant 
protection and seat back strength. Furthermore, ARCCA stated that rear 
impact testing for fuel integrity should be utilized, and that this 
type of testing would enable the agency to assess occupant kinematics 
and interactions in rear impacts.
    Nissan recommended that NHTSA harmonize with the global technical 
regulation (GTR) dynamic test

[[Page 40022]]

procedure.\10\ GM stated that the development of a dynamic test by 
NHTSA should be considered only after recent revisions to FMVSS No. 202 
are assessed. According to GM, if the regulatory changes are shown to 
be effective in mitigating injury, a rear impact NCAP could be better 
directed toward areas not fully addressed by the current regulation. 
Similarly, while Subaru did not support new requirements for FMVSS No. 
202a in the short term, they asserted that NHTSA needs to educate 
consumers on the proper use and adjustment of head restraints. However, 
Subaru believed that in the long term, NHTSA should focus on the study 
of whiplash-type injury mechanisms and applicable countermeasures.
---------------------------------------------------------------------------

    \10\ See http://www.unece.org/trans/doc/2007/wp29/WP29-143-23r1e.doc. This is an agreement to begin work on Phase 2 of this 
GTR, which will analyze a revised dynamic test procedure 
incorporating the BioRID-II dummy.
---------------------------------------------------------------------------

E. Crash Avoidance Technologies

    Comments regarding NCAP information on crash avoidance technologies 
are grouped into three categories: Program Implementation, Selected 
Technologies, and Rating System.
1. Program Implementation
    Most commenters encouraged NHTSA to implement a new component into 
NCAP to rate vehicles on the presence of crash avoidance technologies. 
They agreed that such a program would help educate consumers about 
these technologies and encourage manufacturers to include them in more 
vehicles. According to Ford, the first step would be to identify 
promising technologies with measurable real-world safety benefits. 
Next, those items must be assessed using developed performance based 
metrics, and finally, the assessments should be used to develop crash 
avoidance NCAP ratings that balance rating flexibility with stability.
    GM emphasized an overarching principle that crash avoidance NCAP 
should be biased toward including features that have a high likelihood 
of improving safety. GM suggested further that the agency consider a 
wording revision, perhaps to `Collision Avoidance and Post-Crash Safety 
(CAPS)' NCAP so that a technology such as Automatic Collision 
Notification could be considered and included.
    Honda encouraged NHTSA to consider a program that would define the 
various crash avoidance technologies. They stated that these 
definitions should be based on the effect each function of a particular 
system has from the driver's point of view, and include a clear 
explanation of the actions the system can take to enhance safety. 
Honda, along with Delphi, suggested the development of assessment-
weighting coefficients derived from a system's expected benefits and 
the frequency of the crash type (using appropriate U.S. databases) that 
the system is supposed to address.
    BMW suggested a program that would accomplish the agency's goals 
without over-promising consumers on expected performance and avoid 
crediting systems prematurely. They suggested a program that would 
differentiate technologies with real-world effectiveness from those 
whose effectiveness numbers were generated by some other means. They 
also suggested that NHTSA and manufacturers collaborate on ways to 
educate consumers on emerging technologies with promising capabilities 
and proven benefits.
    Mercedes-Benz (Mercedes) recommended that NHTSA work with the 
automotive industry before developing crash avoidance ratings. To 
develop future ratings they, along with Continental Automotive Systems, 
supported the idea of creating an advisory panel that represents the 
viewpoints of all manufacturers competing in the U.S. market.
    Nissan agreed with the agency's desire to implement this new 
program. They also stated that the agency should identify immediately 
its priority technologies through a press release, on the NCAP Web 
site, through the ``Buying a Safer Car'' brochure, and on each 
vehicle's NCAP summary Web page.
    IIHS and NADA were not convinced of the need for NCAP crash 
avoidance ratings at this time. IIHS suggested that NHTSA should not 
rate vehicle crash avoidance technologies, since the agency cannot 
currently identify which systems are most effective.
2. Selected Technologies
    Nissan and Delphi agreed with the three technologies selected by 
the agency. However, GM and Toyota believed that there were additional 
crash avoidance technologies that should be promoted because they would 
provide safety value to consumers. For brevity, we chose not to list 
them all in this document, but they included such things as daytime 
running lights, backover prevention technology, and advanced collision 
notification. GM further believed that there were data for some of 
these crash avoidance technologies and methods by which potential 
benefits could be assessed, and they could be included in the initial 
implementation of a crash avoidance NCAP. GM felt that limiting crash 
avoidance technologies to the three identified by the agency would 
unnecessarily limit the potential safety benefits to consumers.
3. Rating System
a. Cumulative Rating (NHTSA's Approach 1)
    There was little support for NHTSA's proposed Approach 1. In the 
short term, only Nissan supported a simple cumulative rating whereby 
each priority technology would be weighted the same. Both the Alliance 
and GM were opposed to this approach. GM believed that a cumulative 
rating would not discriminate among the three technologies, and they 
would prefer that NHTSA weight appropriately safety-enhancing features 
based on their relative benefits. The Alliance stated that the 
effectiveness of the selected technologies was not equal, and providing 
equal weighting would significantly mislead the consumer as to their 
relative safety benefits.
    Rather than a star rating or the use of a cumulative rating, BMW 
suggested a ``thumbs up'' rating system to assist consumers in quickly 
and intuitively distinguishing among technologies on the basis of 
maturity. BMW believed that this approach would deliver to consumers 
two levels of information: which technologies have the potential for 
success and which technologies have a history of success. Furthermore, 
BMW felt that this approach would reduce the need for NHTSA to 
research, analyze and document the actual benefits of a technology. 
Mercedes believed that NCAP should issue publications that would rank 
the merits of emerging technologies in a manner similar to that used in 
the IIHS status reports, and that NHTSA should communicate with the 
industry so that public safety messages could be coordinated with 
industry advertisements.
b. Effectiveness Rating (NHTSA's Approach 2)
    Nissan, in the long term, along with Toyota, Volvo, Public Citizen, 
AORC, the Alliance, AIAM and GM favored the agency's proposed Approach 
2 of establishing an effectiveness rating for crash avoidance 
technologies. Toyota, however, believed that it would be ideal to 
develop information related to each new technology's safety potential 
and to establish a ``Graduated Comprehensive Crash Avoidance Rating 
System'' concept. They also recommended

[[Page 40023]]

further study to expand the list of technologies beyond ESC, lane 
departure warning and forward collision warning to include systems such 
as rear pre-collision preparation/warning, emergency stop signal, blind 
zone alert, vehicle-to-vehicle and vehicle-to-infrastructure 
communications.

F. Presentation of NCAP Information

    Comments regarding the presentation and dissemination of NCAP 
focused mainly on a combined crashworthiness rating. A few commenters 
offered suggestions on the dissemination of NCAP information. NADA 
suggested that NHTSA develop, maintain and make available a database of 
non-agency sources of credible vehicle safety information. The CAS and 
CFA suggested that the agency implement additional and more 
sophisticated systems that deliver safety information at the point of 
sale. They believed this information should be beyond the agency's new 
NCAP labeling program (no examples were given).
Combined Crashworthiness Rating
    Most responders to the NCAP notice expressed support for an overall 
crashworthiness rating that combined the results from all the crash 
modes (front and side) tested. However, IIHS cautioned that an all-
encompassing single rating may allow some poor performance qualities to 
be hidden under the umbrella rating. Therefore, they urged NHTSA to 
provide consumers with all of the scores in each crash mode to allow 
them to choose which vehicle to purchase. Additionally, Delphi, Public 
Citizen and Bidez and Associates noted that while a single overall 
crashworthiness rating would simplify information for consumers, it 
could also confuse consumers if not based on sound science.
    Toyota believed there is merit to combining ratings for 
crashworthiness evaluations to provide the consumer with a 
comprehensive summary of the crash performance of the vehicle in front 
and side impacts. They recommended weighting the injuries and 
assessment in each impact condition by the distribution of serious 
injuries (AIS3+) and fatalities. After determining the weighting 
factors for each injury, each impact configuration should receive 
similar ``Field Relevance Weighting'' based on frequency, severe injury 
risk, and occupancy. Because of the small number of fatalities in NASS, 
Toyota suggested exploring FARS augmented with the Multiple Cause of 
Death (MCOD) database.
    Honda supported a combined crashworthiness rating that covers a 
wide variety of real-world collisions. Honda recommended compatibility 
testing that assesses performance in crashes between two vehicles with 
different geometries and/or weights. Further, they recommended 
weighting coefficients for each region of the crash test dummy, 
representing specific types of injuries, based on real-world crash and 
injury data.
    The Alliance generally supported the concept of a combined 
crashworthiness rating. They believed that it is possible to combine 
the different body regions into a single star rating for both frontal 
and side. However, they noted that the frontal NCAP ratings are 
vehicle-weight dependent while the side NCAP ratings are generally 
weight independent. Thus, the Alliance asserted that a combined 
crashworthiness rating would be comparable only within vehicle weight 
class. Moreover, AIAM urged NHTSA to ensure that a single rating is 
meaningful in terms of real-world performance to drive safety 
improvements in all crash modes. They recommended that changes to the 
star system be considered only if based on appropriate research 
involving consumer surveys or focus groups, and not on intuitive 
judgments about what data presentation is most effective.
    Public Citizen supported a single rating if it were weighted with 
respect to saving lives and preventing injuries. They also suggested 
that NHTSA use a letter grade rating system instead of ``stars.'' 
Volkswagen believed that the agency should consider a single crash 
rating only until a crash avoidance NCAP rating grows in substance and 
scope. Delphi expressed that a combined crashworthiness rating would 
obscure safety benefits; rather, they supported a Euro NCAP style point 
system and recommended that key performance-based assessments be 
presented as the primary information and that feature-based indicators 
be presented as of secondary importance.

G. Manufacturer Self-Certification (of NCAP Results)

    With regards to manufacturers providing their own NCAP test 
results, GM and Toyota supported the implementation of a type-approval 
program wherein NHTSA would oversee NCAP testing conducted by the 
manufacturer. GM felt that NHTSA's attendance (or the presence of a 
NHTSA representative) would allow appropriate scrutiny of the testing 
and ensure consumer confidence in such a program. Additionally, they 
strongly discouraged implementation of any program that could 
compromise NHTSA-sanctioned vehicle ratings because of results obtained 
through spot-checking (presumably conducted by NHTSA). Bidez and 
Associates, Consumers Union and Public Citizen urged NHTSA to consider 
a manufacturer self-certifying process in which the industry would test 
and rate its own vehicles and undergo spot checking of their test 
results by NHTSA. According to these commenters, the benefit of such a 
program would be to disseminate NCAP test information on newly-
introduced vehicles more rapidly than under the current system.

H. Other Suggestions

    In addition to the approaches that NHTSA had proposed to further 
enhance its NCAP crashworthiness and crash avoidance activities, 
commenters submitted other recommendations to the agency. These 
comments on other possible approaches to improving NCAP are grouped 
into the following categories: Child Restraints and Rear Seat Testing, 
Lighting, and Pedestrians.
1. Child Restraints
    Public Citizen suggested that NHTSA incorporate a dynamic child 
restraint system (CRS) test into NCAP in all crash modes (including 
frontal, rollover, side and rear crashes). They recommended that a six-
year old Hybrid III dummy be restrained in a backless booster and a 5th 
percentile female Hybrid III dummy be placed in a 3-point belt in both 
rear-outboard seating positions. ARCCA recommended adding instrumented 
child dummies to the outboard-designated seating positions in the rear 
to investigate issues associated with accommodations and crash 
performance of rear-seated occupants resulting from cargo.
    Bidez & Associates asserted that the agency should build upon and 
leverage the experience of EuroNCAP in child protection to force design 
innovation in rear seat safety for six to twelve-year olds.\11\ They 
believed there was a need to enhance frontal impact protection of nine 
to twelve-year old children who are properly belted in the rear seat. 
Their research for restrained nine to twelve-year old children 
suggested that rear seat occupants had a risk of serious injury 78 
percent higher than that of front seat occupants. They estimated that 
the overall injury rate for all restrained nine to twelve-year olds in 
all crash types was 38 percent higher in the rear seat than in the 
front seat. As such,

[[Page 40024]]

Bidez & Associates recommended that NHTSA immediately warn consumers, 
retract its message to parents about placing children in the rear, and 
force the automobile industry to upgrade the safety of the rear 
occupant area of the existing and future vehicle fleet.
---------------------------------------------------------------------------

    \11\ The commenter did not provide specific detail as to what 
design innovations have occurred as a result of the EuroNCAP 
activity.
---------------------------------------------------------------------------

    Subaru, GM and the Alliance opposed implementation of a CRS test 
into NCAP. GM asserted that there can be no meaningful dynamic NCAP 
test for CRS until there is a meaningful way to tie a CRS NCAP 
performance rating to real-world performance. They believed that it is 
inappropriate to invent a test and claim correlation to real-world 
safety performance improvements without sound data to back this claim. 
These commenters felt that using child safety seats in NCAP vehicle 
tests would confound the test results and would not lead to a 
meaningful vehicle or CRS rating. Additionally, the Alliance asserted 
that the real-world safety benefits of child restraints demonstrate the 
children are already very well-protected in the rear seat. As such, 
they believed that adding child dummies in child restraints to the rear 
seating position for front or side NCAP testing would not maximize 
advancements in child protection.
    Volvo suggested that if the agency wanted to develop a child 
restraint test, then the test should be performed on a sled, and they 
asserted that there should be improvements in FMVSS No. 213. According 
to Volvo, the restrictions for design and testing of the restraints, as 
set up in this standard, basically prohibit innovative concepts with 
improved performance for reducing misuse and improper installation and 
for improving safety performance in a crash. To improve child safety, 
Consumers Union recommended that NHTSA pursue research toward an NCAP 
rating on (rear) vehicle visibility since they believed that data from 
Kids and Cars and others suggest that children are most at risk from 
poor visibility and blind zones around the vehicle.
2. Rear Seat Testing
    Adding rear seat dummies into the frontal NCAP program was 
encouraged by some commenters. In particular, AORC and Bidez and 
Associates suggested the addition of the 5th percentile female or the 
10-year old dummy. However, AORC asserted that an analysis of field 
data would be needed to determine the most appropriate dummy and 
seating position, and that dummy development may be required in this 
area to better measure abdominal injuries that may be present among 
belted occupants in the rear seat.
    Individual commenter Mr. Todd Saczalski recommended rear seat 
testing with adult and child dummies and child restraints to assess the 
difficulty exiting the vehicle and to examine injuries due to seat back 
failure. The Children's Hospital of Philadelphia (CHOP) stated that the 
agency should place an older belt-restrained dummy, such as the six or 
ten-year old Hybrid III child dummy, in the rear seat of the NCAP 
frontal test to better understand rear restraint systems for child 
occupants. Additionally, they encouraged the use of a belt-positioning 
booster seat with the six-year old Hybrid III dummy.
    Subaru did not support adding dummies to the rear seating position. 
Subaru stated that it might not be possible, with the current front 
seat positioning procedure, to properly position a 50th percentile male 
Hybrid III dummy in the rear seat of some vehicles; the result could be 
inconsistent performance evaluations across all vehicles.
3. Lighting
    Some public commenters expressed concerns about lighting and glare 
related to daytime running lights (DRLs). However, the glare comments 
were focused on the agency's rulemaking activity and not its consumer 
information activity. Therefore, daytime running lights are not 
discussed in this notice. GM stated that numerous field effectiveness 
studies conducted throughout the world show that DRLs could prevent 
some crashes. Citing an analysis of field data suggesting that under 
daytime conditions, daytime running lights can prevent 5 percent of 
opposite direction crashes and 12 percent of pedestrian and 
pedalcyclist crashes, GM encouraged NHTSA to expand the installation of 
DRLs and include this technology in its crash avoidance rating so that 
manufacturers will be encouraged to install them and provide additional 
collision avoidance benefit.
4. Pedestrians
    Consumers Union recommended that NHTSA study the work of auto 
safety researchers in other countries to determine whether a 
pedestrian-friendly NCAP rating would be effective in the United 
States. Consumers Union noted that Honda has taken a leadership role in 
designing a dummy for testing pedestrian safety and designing its 
vehicles with pedestrian safety in mind. They urged NHTSA to consider 
using the Honda pedestrian dummy and to pursue other opportunities to 
improve pedestrian safety. Public Citizen encouraged NHTSA to issue a 
pedestrian NCAP test and an accompanying safety standard. They also 
challenged NHTSA to follow the lead of the rest of the world by taking 
a far more aggressive stand against the dangers vehicles pose to 
pedestrians and to raise the bar for pedestrian safety in its 
discussions for a Global Technical Regulation (GTR) on pedestrian 
safety.

IV. Discussion and Agency Decision

A. Frontal NCAP

    In the comments to the notice and the public hearing concerning 
enhancements to frontal NCAP, most manufacturers and vehicle safety 
advocates supported the retention of the current frontal crash test 
protocol at 35 mph (56 kmph). Additionally, several comments suggested 
that NCAP injury criteria and metrics be consistent with FMVSS No. 208. 
Most responders favored using the KTH injury metric (after additional 
research) but also encouraged the inclusion of other injury criteria 
such as neck and chest deflection. Some commenters suggested that the 
agency immediately evaluate lower leg injuries with the Thor-Lx dummy, 
while others recommended that NHTSA harmonize with Japan and Euro NCAP 
on lower leg assessments. The agency's analysis and decisions on 
frontal NCAP are grouped by categories: Test Dummies, Injury Criteria 
and their associated Risk Curves, and Lower Speed Testing.
Test Dummies
    Comments pertaining to the adoption of additional test dummies 
included wide support for the 5th percentile female Hybrid III dummy, 
including its placement in the right front seating position. Others 
recommended that the agency include a 95th percentile male Hybrid III 
dummy in frontal NCAP. It was also suggested that dummies be placed in 
the rear seat for the purpose of rating vehicles.
    In response to these comments, NHTSA has decided to include the 5th 
percentile female Hybrid III dummy in the right front passenger seating 
position. GM provided the most compelling evidence, and the agency 
reexamined its own data and reached the same conclusion.\12\ That is, 
the real-

[[Page 40025]]

world data suggest that the smaller females were at greater risk and 
more likely to be seated in the right front position in frontal 
crashes. The agency believes that this dummy's incorporation into the 
NCAP frontal program is reflective of real-world crash conditions.
---------------------------------------------------------------------------

    \12\ The agency's analysis found, based on NASS-CDS estimates 
from 1997-2006, that the risk of AIS 2+ injury for smaller belted 
occupants in the right front passenger seating position is 33% 
greater than that of a mid-sized adult belted occupant in the same 
seating position in full frontal crashes (0-40 delta velocities, 
non-rollover cases, age ranges from 13 years old or older, height 
for small adult: Less than 65 inches, and height for mid-sized 
adult: 65-73 inches).
---------------------------------------------------------------------------

    NHTSA has chosen, however, not to include the 95th percentile male 
Hybrid III dummy in frontal NCAP at this time. The 95th percentile male 
Hybrid III dummy has not been evaluated for robustness, 
reproducibility, and repeatability in laboratory impact conditions and 
it has only undergone very limited sled and vehicle testing. As such, 
we believe additional research and testing with this dummy is necessary 
before it can be included into frontal NCAP.
    With regards to placing adult dummies in the rear seating positions 
of frontal NCAP tests, NHTSA believes that more analysis is needed 
before a rating program that includes rear seat occupants can be 
established. The agency has conducted some limited testing with both 
the 50th and 5th percentile Hybrid III adult dummies in the rear seat 
under a full frontal impact condition. However, these preliminary 
results did not correlate to findings in the real-world and additional 
research is necessary to better understand the results.\13\ Similarly, 
none of the commenters that suggested an NCAP rating program for the 
rear seat provided the necessary data to establish how such a program 
would lead to meaningful improvements in safety.
---------------------------------------------------------------------------

    \13\ Kuppa, S., Saunders, J., Fessahaie, O., Rear Seat Occupant 
Protection in Frontal Crashes, Paper No. 05-0212, Nineteenth ESV 
Conference, Washington DC (2005).
---------------------------------------------------------------------------

    The agency has decided not to incorporate the use of the lower legs 
from the Thor dummy to evaluate lower leg injuries into the program at 
this time. The agency is awaiting the completion of research currently 
in progress by an SAE task group. Additionally, this tool has not 
undergone the necessary robustness, reproducibility, and repeatability 
testing that the agency believes is necessary for incorporation into an 
NCAP ratings program.
Injury Criteria and Risk Curves
    With regards to frontal NCAP injury criteria, the agency agrees 
with the commenters and has decided to include all of the FMVSS No. 208 
body regions into the frontal NCAP rating system. As suggested by many 
commenters, the agency believes that their inclusion will not only add 
to the robustness of vehicle evaluations, but it will make the criteria 
used to assign NCAP frontal ratings consistent with those used in FMVSS 
No. 208 and in other frontal-crash vehicle assessment programs. It will 
also allow the agency to incorporate all safety concerns related to 
injury criteria readings into the calculation of the frontal rating 
thus eliminating the need to use the safety concern symbol.\14\ 
However, unlike the current NCAP program which uses chest acceleration 
to assess thoracic injury risk, the new frontal program will focus 
instead on peak chest deflection instead. We believe that the inclusion 
of chest deflection into frontal NCAP will encourage development of 
restraint systems that will further reduce the risk of thoracic 
injuries.\15\ This is especially true given a manufacturer's compliance 
margin with the chest acceleration limit of 60 G's and the fact that 
the FMVSS No. 208 belted test is now conducted at the same speed as the 
frontal NCAP test. Accordingly, frontal NCAP will include the following 
body regions and injury criteria: Head (HIC15), neck (Nij, 
tension, and compression), chest (deflection), and femur (axial force). 
The risk curves that will be used for these criteria are described 
below.
---------------------------------------------------------------------------

    \14\ A safety concern symbol is a test occurrence that is not 
reflected in a vehicle's star rating but that NHTSA feels is of 
significant importance that the event should be communicated to 
consumers.
    \15\ The agency evaluated new MY 2005-2007 tested vehicles and 
found that for acceleration, the standard deviation for risk of 
injury was approximately 3% compared to chest deflection 
which was approximately 4%.
---------------------------------------------------------------------------

    As indicated in our proposal, NHTSA is also adopting AIS 3+ and AIS 
2+ injury risk curves to assess the risk of injury to front seat 
occupants.\16\ This approach is different from the current NCAP rating 
system which uses AIS 4+ (severe) injury risk curves. The new risk 
curves will focus vehicle performance on more frequently occurring 
injuries than severe (AIS 4+) or critical (AIS 5+) injuries.
---------------------------------------------------------------------------

    \16\ Details of these injury risk curves are provided in 
Appendix C, Injury Risk Curves for the NCAP Combined Crashworthiness 
Rating System.
---------------------------------------------------------------------------

    With the exception of chest deflection, the AIS 3+ injury risk 
curves that will be used by the agency in NCAP are the same as those 
used for FMVSS No. 208. The AIS 3+ chest deflection injury risk curve 
that the agency will use in NCAP was developed in 2003 by Laituri et 
al.\17\ The agency chose this risk curve for deflection because, as 
noted by the agency during the FMVSS No. 208 advanced air bag 
rulemaking, the chest deflection risk curve published by the agency was 
not used to establish the performance limits currently in FMVSS No. 
208.
---------------------------------------------------------------------------

    \17\ Laituri, T., Prasad, P., Sullivan, K., Frankstein, M., 
Thomas, R. (2005), Derivation and Evaluation of a Provisional, Age 
Dependent AIS 3+ Thoracic Risk Curve for Belted Adults in Frontal 
Impacts, SAE Paper No. 2005-01-0297.
---------------------------------------------------------------------------

    The agency will be using an AIS 2+ risk curve for the femur because 
most femur fractures are either of the AIS 2 or AIS 3 injury severity. 
Additionally, the AIS 2+ femur risk curve was primarily developed from 
multi-fragmentary patellar fractures, which, like other articular 
surface injuries, are associated with a high level of disability. As 
such, using an AIS 2+ injury risk curve will help ensure that 
debilitating multi-fragmentary patellar fractures are addressed.\18\
---------------------------------------------------------------------------

    \18\ See Ore, L., Tanner, B., States, J. (1993), Accident 
Investigation and Impairment Study of Lower Extremity Injury, SAE 
Paper No. 930096, SAE International Congress and Exposition, 
Detroit, MI, and MacKenzie, E. (1986), The Public Health Impact of 
Lower Extremity Trauma, SAE Paper No. 861932, Symposium on 
Biomechanics and Medical Aspects of Lower Limb Injuries, San Diego.
---------------------------------------------------------------------------

    NHTSA has decided not to incorporate an advanced KTH risk curve 
into frontal NCAP at this time. In consideration of the comments 
received and because this risk curve is undergoing additional 
evaluation, the agency felt it would be premature to include it in 
NCAP. However, we do believe that the inclusion of a femur injury 
criterion, as indicated above, will lead to improved bolster design. 
Similarly, when coupled with the other injury criteria for chest 
deflection and neck, will lead to overall improved restraint system 
designs. NHTSA has also decided not to harmonize its NCAP femur injury 
values with those of EuroNCAP and Japan NCAP. The agency evaluated the 
rating schemes of these international programs along with that from the 
IIHS. These programs use a sliding scale to rate vehicles as opposed to 
injury risk curves. As such, as will be explained later in this 
document, because we have chosen to maintain our current methodology 
for combining injury risk we cannot substitute sliding scales for risk 
curves.\19\
---------------------------------------------------------------------------

    \19\ The sliding scales in these programs relate injury measures 
to point values without equating them to probability of injury. 
However, risk curves equate the injury measures to expected risks of 
injury.
---------------------------------------------------------------------------

    The injury risk curves used in the NCAP frontal crash test program 
for the 50th percentile male Hybrid III and 5th percentile female 
Hybrid III dummies

[[Page 40026]]

are shown below. How these injury risk curves will be combined to 
generate a vehicle's frontal NCAP star rating will be discussed later 
in Section IV-F.
BILLING CODE 4910-13-P
[GRAPHIC] [TIFF OMITTED] TN11JY08.000


[[Page 40027]]


[GRAPHIC] [TIFF OMITTED] TN11JY08.001

BILLING CODE 4910-13-C
Lower Test Speed
    A lower test speed for frontal NCAP was supported by some 
commenters but an almost equal number opposed such an NCAP test. In 
light of the real-world studies conducted by the agency and some of the 
commenters, NHTSA has decided that additional research is necessary to 
fully address the proposal for a lower test speed. At this time, the 
agency has insufficient data with respect to test speed, injury 
mechanisms, dummy biofidelity, and risk curves to proceed.

B. Side NCAP

    Most commenters supported the agency's proposal to incorporate an 
oblique pole test into the program, with several suggesting that this 
test should be adopted after the completion of the FMVSS No. 214 phase-
in. Additionally, several responses encouraged the adoption of new test 
dummies for side NCAP including WorldSID, SID-IIs and ES-2re dummies. 
Commenters also suggested that side impact test procedures and injury 
criteria be consistent with FMVSS No. 214. Finally, IIHS encouraged 
NHTSA to adopt or modify their current moving deformable barrier (MDB). 
The agency's analysis and decisions on side NCAP are grouped into the 
following categories: MDB Design, MDB Test Speed, Oblique Pole Test, 
Test Dummies in the MDB and Oblique Pole Tests, and Injury Criteria and 
their associated Risk Curves.

[[Page 40028]]

MDB Design
    The agency has decided against any modifications to the existing 
moving deformable barrier. Instead, we will evaluate the IIHS MDB 
(including the crabbed vs. perpendicular configuration) as part of a 
more comprehensive approach that is currently underway. This research 
will help the agency decide what properties a new MDB should have. As 
noted in the FMVSS No. 214 Final Rule,\20\ initiatives to improve 
vehicle compatibility between passenger cars and light truck vehicles 
in side crashes are likely to change the characteristics of striking 
vehicles in the future.\21\ As such, we believe these new 
characteristics should be included in any upgraded MDB.
---------------------------------------------------------------------------

    \20\ 72 FR 51908, Docket No. NHTSA-2007-29134.
    \21\ 69 FR at 27992, Docket No. NHTSA-2004-17694.
---------------------------------------------------------------------------

MDB Test Speed
    There was little support for an increased test speed for side NCAP, 
while some urged the agency to maintain or lower the current speed. As 
indicated in our request for comments, the real-world data indicates 
that the current test speed is largely representative of real-world 
crashes in which serious and fatal injuries occur; yet, increasing the 
test speed by 5 mph (8 kmph) would capture approximately 5,000 more 
serious and fatal injuries. No commenters disagreed with this data. 
However, NHTSA has not conducted any testing at this increased test 
speed with the ES-2re or SID-IIs dummies, and we want to better 
understand what countermeasures would be developed if the test speed in 
side NCAP were increased to 43.5 mph (71 kmph) or higher. As such, 
NHTSA has decided to maintain the current test speed and we will 
evaluate the test speed as part of our more comprehensive research work 
that is already underway.
Oblique Pole Test
    Most commenters supported incorporating an oblique pole test into 
NCAP. However, some opposed this proposal, stating that a pole test 
would not add an incentive for manufacturers to provide additional head 
side impact protection beyond the IIHS side impact test. The agency 
does not agree with these commenters. As we stated in the FMVSS No. 214 
Final Rule, we believe that the pole test in conjunction with our 
current MDB will drive better head, chest and pelvis protection than 
conducting the IIHS side impact test alone. Recent pole tests conducted 
on vehicles that were found to have ``Good'' or ``Acceptable'' 
performance in the IIHS barrier test had dummy head and pelvis injury 
readings, for some vehicles, that were significantly higher than the 
IIHS test indicated.\22\ These test results indicate that the use of 
the oblique pole test in NCAP will demand more robust countermeasure 
designs leading to higher levels of safety performance.
---------------------------------------------------------------------------

    \22\ See Appendix A, NCAP and IIHS Pole Test Results.
---------------------------------------------------------------------------

    Because the pole test can evaluate only one seating position at a 
time, most commenters were in support of running one pole test. Several 
stated that conducting multiple side impact pole tests with different 
sizes of dummies would introduce significant test burden. We have 
decided to add the oblique pole test procedure specified in the FMVSS 
No. 214 Final Rule for all vehicles tested by NCAP. Therefore, rather 
than conducting a pole test for each outboard seating position in the 
vehicle, we will conduct only one test to evaluate the front seat 
outboard performance of vehicles. NHTSA believes that a single pole 
test with one dummy will provide consumers with information on side 
pole performance without introducing significant test burden to both 
NHTSA and manufacturers.
Test Dummies in the MDB and Oblique Pole Tests
    Outside of those commenters who suggested use of the World SID, 
most commenters supported the incorporation of the new, recently 
federalized side impact crash test dummies into side NCAP. Some 
specifically proposed that the agency use the 50th percentile male ES-
2re dummy for the driver seating position and the 5th percentile female 
SID-IIs dummy for the rear seating position in the MDB test. For an 
oblique pole test, most encouraged the use of the SID-IIs dummy in the 
driver seating position.
    Several commenters recommended that the agency incorporate the 
WorldSID dummy into Part 572 and side NCAP. For both test 
configurations (pole and MDB), the agency has decided not to 
incorporate this dummy into NCAP at this time. Although the agency has 
been conducting testing and evaluation to determine the suitability of 
incorporating the WorldSID into Part 572 and side impact crash tests, 
further work remains to be completed before its use in NCAP can occur.
    Test dummy selection for the MDB and the pole test are discussed 
below.
a. MDB Test
    NHTSA has decided to incorporate the new 50th percentile male ES-
2re dummy into the driver seating position and the 5th percentile 
female SID-IIs dummy in the rear seating position for the MDB test as 
adopted in the FMVSS No. 214 Final Rule. The agency selected the 50th 
percentile male ES-2re dummy in the driver position because its weight 
and height is more representative of the average driving population 
than is the SID-IIs dummy. The 5th percentile SID-IIs dummy was 
selected for the rear seating position because it is closer in height 
to the average outboard rear seat occupant than the 50th percentile ES-
2re dummy, and its placement in the rear seat will lead to a more 
demanding test.\23\
---------------------------------------------------------------------------

    \23\ In the testing which supported the FMVSS No. 214 upgrade, 
both the 5th and the 50th percentile dummies passed the MDB test but 
the rear was more stringent and difficult for the 5th percentile 
dummy.
---------------------------------------------------------------------------

b. Oblique Pole Test
    NHTSA has decided to conduct only one oblique pole impact test with 
the 5th percentile female SID-IIs dummy in the driver position. As 
stated in our recent FMVSS No. 214 Final Rule, small stature drivers 
(height up to 5 feet 4 inches) comprise approximately 28 percent of 
seriously or fatally injured drivers in narrow object side impacts. In 
addition, real-world crash data suggests that small stature occupants 
have a higher proportion of head, abdominal, and pelvic injuries and a 
lesser proportion of chest injuries than median stature occupants.
    So while we selected the 50th percentile dummy for the front 
seating position in the MDB test (because it represents the average 
driver), for the pole test we are selecting the 5th percentile dummy as 
the driver because in collisions with narrow objects, the 5th 
percentile has the higher risk of injury. Additionally, since we are 
conducting the MDB test with the 50th percentile dummy in the driver 
seating position and the 5th percentile dummy in the driver seating 
position for the pole test, manufacturers will have to encompass a 
broader range of seating positions with their vehicle and restraint 
system designs.
Injury Criteria and Risk Curves
    As with frontal NCAP, several commenters stated that the injury 
metrics used in NCAP should be consistent with the safety standard that 
serves as their basis. In the case of side NCAP, the safety standard is 
FMVSS No. 214. Several commenters stated that

[[Page 40029]]

the adoption of the 50th percentile male ES-2re and 5th percentile 
female SID-IIs dummies and their associated injury criteria from FMVSS 
No. 214 would facilitate a more comprehensive assessment of side impact 
injury. NHTSA agrees with these commenters and has decided to 
incorporate head (HIC36), chest (deflection), abdomen 
(force), and pelvic (force) injury criteria as well as applicable risk 
curves to rate vehicles for the ES-2re and, consistent with the safety 
standard, HIC36 and pelvic (force) for the SID-IIs 
dummy.\24\ NHTSA believes that these criteria and their inclusion in 
side NCAP will lead to a more robust rating. Similarly, it will also 
allow the inclusion of head- and pelvic-related injury criteria in the 
calculation of the side rating without the need for the safety concern 
symbol. Similarly, the injury risk curves that the agency will use in 
side NCAP are the same as those used for the recent upgrade to FMVSS 
No. 214.\25\
---------------------------------------------------------------------------

    \24\ We note that for the SID IIs, we are not incorporating 
spine acceleration at this time. Even though this measure is 
included in the new FMVSS No. 214, we do not have a risk curve that 
has been validated at this time to include in our rating scheme for 
rating vehicles for side impact protection.
    \25\ Details of these injury risk curves are provided in 
Appendix C, Injury Risk Curves for the NCAP Combined Crashworthiness 
Rating System.
---------------------------------------------------------------------------

    The table below presents the applicable injury criteria and 
associated injury risk curves for each dummy that will be used in the 
side NCAP vehicle rating. How these injury risk curves will be combined 
to generate a vehicle's side NCAP star rating will be discussed later 
in Section IV-F.
BILLING CODE 4910-13-P

[[Page 40030]]

[GRAPHIC] [TIFF OMITTED] TN11JY08.002


[[Page 40031]]


[GRAPHIC] [TIFF OMITTED] TN11JY08.003

BILLING CODE 4910-13-C
Lead Time
    While most commenters supported the inclusion of the pole test in 
NCAP, an almost equal number suggested that the test not be 
incorporated until after FMVSS No. 214 is fully phased-in. NHTSA does 
not agree with these commenters. NHTSA believes that some manufacturers 
have begun to design vehicles to meet the pole test and we want 
consumers to be aware of those vehicles. Additionally, we believe that 
conducting the pole test for MY 2010 will provide an incentive for 
others to begin and/or accelerate their processes for improvement as 
well. Finally, rating vehicles on both their performance in the pole 
test and the MDB test, which will now incorporate HIC and other 
criteria, will help foster an environment for vehicle manufacturers to 
design better side impact designs for the head, chest and pelvis, and 
allow consumers to make more informed choices based on these new tests.

C. Rollover NCAP

    Several commenters suggested that the agency add an additional star 
to the Rollover NCAP rating for vehicles equipped with ESC. They 
suggested the extra star be supplemented by a footnote saying, 
``equipped with electronic stability control.'' In addition, one 
commenter suggested that a star be subtracted from vehicles not 
equipped with ESC. Commenters also recommended that NHTSA incorporate a 
new, dynamic structural test into rollover NCAP. The agency's analysis

[[Page 40032]]

and decisions regarding NHTSA's rollover program are grouped into two 
categories: Rollover Risk and Injury Risk Models and Dynamic Rollover 
and Structural Test.
Rollover Risk and Injury Risk Models
    With regards to the agency's proposal to develop a new rollover 
risk model, the agency agrees with commenters' concerns about the 
effects of ESC on the rollover risk model. However, we do not agree 
that is appropriate to add or subtract a star in the rollover rating to 
account for ESC. The current rollover rating is the result of a 
detailed analysis of a vehicle's potential risk of rollover if a crash 
is initiated. Given that the star bands are set at 10 percent, adding a 
star to the rollover risk rating could suggest to consumers that ESC 
would reduce a particular vehicle's risk of rollover by up to 10 
percent in a given crash. This could result in unsupported and 
inaccurate vehicle ratings.
    The current rollover risk model was fit using crash data collected 
several years ago (at a time when ESC was available in relatively few 
vehicles). We are monitoring the fit of the model to newer data and, in 
particular, to data for ESC-equipped vehicles. We have identified 7,000 
single-vehicle crashes with NCAP-tested vehicles equipped with ESC in 
our State Data System (SDS). At this time, the current model appears 
consistent with the newer data, possibly (at least in part) because of 
the sampling variability associated with the relatively small ESC 
subset. A larger sample may produce different results, and we will 
recalibrate the estimates if we determine conclusively (that is, beyond 
the effects of statistical variability) that the current estimates do 
not describe the newer data. In the meantime, we will continue to use 
the risk estimated from the vehicle's Static Stability Factor (SSF) and 
its propensity to tip up in the dynamic rollover ``fishhook'' test as 
described in 68 FR 59250 (October 14, 2003). These are provided below:
[GRAPHIC] [TIFF OMITTED] TN11JY08.007


Where SSF=static stability factor
    This model describes the absolute risk of rollover given a single-
vehicle crash.
    As will be discussed later, we will include ESC in the new NCAP 
Crash Avoidance Rating. We feel this will be much more effective in 
highlighting the importance of ESC and other potentially life-saving 
technologies.
Dynamic Rollover and Structural Test
    In their public hearing testimony, Ford suggested that NCAP dynamic 
rollover protocol be aligned with compliance protocol for ESC to 
minimize the risk of unintended consequences from the program. The 
agency does not agree with this suggestion. These tests have 
significantly different performance requirements and are intended to 
measure different dynamic vehicle responses. In the future, it may be 
possible to address the likelihood of aligning the new ESC compliance 
test with the NCAP dynamic rollover ``fish-hook'' test, but additional 
research is needed before these two tests can be combined. Neither test 
measures the responses from the other test; therefore, neither test 
could be used as a substitute for the other.
    Some commenters suggested a structural rollover test; in 
particular, NHTSA received comments regarding the Jordan Rollover 
System (JRS) test device.\26\ Some commenters believe that the JRS test 
can be conducted with dummies to demonstrate whether vehicle roof 
performance meets objective injury and ejection criteria for belted and 
unbelted occupants. As part of our roof crush upgrade, the agency has 
received numerous comments regarding the JRS device.\27\ The JRS and 
other dynamic rollover procedures are being addressed as a part of the 
roof crush rulemaking currently underway. Therefore, a decision on its 
appropriateness for incorporation into NCAP would be premature at this 
time.
---------------------------------------------------------------------------

    \26\ The JRS device rotates a vehicle body structure on a 
rotating apparatus (``spit'') while the road surface moves along the 
track and contacts the roof structure.
    \27\ See Docket No. NHTSA-2005-22143.
---------------------------------------------------------------------------

D. Rear Impact

    With regards to rear impact NCAP, some commenters urged the agency 
to include a rear impact crash test rating and/or the IIHS test results 
in NCAP. Others indicated that linkage to IIHS could appear to be an 
agency endorsement of the IIHS testing and that it would be premature 
to incorporate a new rear impact dynamic test into NCAP since the 
effect of the new FMVSS No. 202a requirements is unknown at this 
time.\28\ Rather, they suggested that NHTSA educate consumers on the 
proper use and adjustment of head restraints.
---------------------------------------------------------------------------

    \28\ By MY 2012, 100% of front and rear seats will have to meet 
the upgraded FMVSS No. 202a.
---------------------------------------------------------------------------

    NHTSA does not agree that a dynamic test would be premature at this 
time since such an option exists in our FMVSS No. 202a. However, we do 
agree with the commenters that providing the IIHS results on our Web 
site could lead to consumers believing that the agency has approved, in 
particular, their dynamic test procedure. In addition, we note that the 
test dummy used by IIHS has not been approved for regulatory use, and 
some of the injury criteria used for this assessment have not been 
correlated with real-world injury.
    We also see very little benefit to consumers in publishing IIHS's 
static head restraint ratings of Good, Acceptable, Marginal, etc. on 
http://www.safercar.gov. The agency's upgraded head restraint 
regulation (FMVSS No. 202a) will begin an 80% phase-in for front seats 
in MY 2010. Any manufacturer certifying their head restraint to the 
static option of FMVSS No. 202a, according to IIHS's current scheme, 
would be placed in the Good or Acceptable category. Most of those not 
achieving a Good rating will be adjustable head restraints that IIHS 
downgrades by one category simply because they are adjustable. Thus, 
there would be very little meaningful difference in the rating.
    For those manufacturers certifying their head restraints to the 
dynamic option in FMVSS No. 202a, the static IIHS rating would not 
provide a meaningful metric of performance. The agency also 
contemplated publishing the actual numerical values of static height 
and backset that the IIHS measures but have decided against this 
course. We believe that consumers would find this information confusing

[[Page 40033]]

and difficult to interpret. As such, rather than providing the IIHS 
data on our Web site, we have decided to update http://www.safercar.gov 
to include information related to proper head restraint adjustment.

E. Crash Avoidance Technologies

    Most commenters supported the agency's proposal to implement a 
crash avoidance ratings program. However, there were two commenters who 
did not believe that a crash avoidance rating program was needed at 
this time. Two commenters suggested that NHTSA work with the automotive 
industry to create an advisory panel to develop a crash avoidance 
rating system. Additionally, most responses did not favor a cumulative 
rating system; instead, several commenters emphasized the importance of 
selecting advanced technologies and developing a rating system based on 
real-world effectiveness. Furthermore, several commenters recommended 
that the agency consider other advanced technologies beyond ESC, FCW 
and LDW.
    NHTSA agrees that a rating system that incorporates a crash 
avoidance system's estimated benefit is ideal. We also believe that we 
should establish this new program quickly for two reasons. First, we 
want to draw a greater distinction for consumers regarding vehicles 
that are being equipped with ESC during the phase-in period. Second, in 
addition to ESC, there are other new safety technologies which exist 
today that can assist a driver in preventing severe and frequently 
occurring crashes. We believe that through NCAP, we can provide an 
incentive to encourage accelerated deployment of these new, advanced 
technologies. The agency's analysis and decisions on new crash 
avoidance ratings program are grouped into the following categories: 
Selected Technologies and Rating System.
Selected Technologies
    Those commenters who supported establishment of a program that 
would promote crash avoidance technologies agreed with the agency's 
selection of ESC, FCW and LDW as beneficial technologies. Others 
believed that the agency should expand its list to encompass crash 
avoidance, crashworthiness and post-crash technologies so as not to 
limit the potential safety information that could be provided to 
consumers. NHTSA believes that ESC, FCW and LDW are the only 
technologies that meet the agency's criteria and are mature enough for 
inclusion in a crash avoidance rating program. That is, all three have 
available benefits data and performance test procedures to be included 
in a rating program.
    We believe that both FCW and LDW will address major crash problems 
seen on U.S. roadways. FCW is designed to address primarily rear-end 
crashes, which account for approximately 30 percent of all crashes, 
while LDW is designed to address crashes due to unintended lane drift. 
Crash types that may result from lane drift include road departure and 
opposite direction crashes. The NCAP report showed that rear-end road 
departure, and opposite direction crashes represent a significant 
amount of the total maximum AIS 3+ injuries.\29\ Results from large 
scale field tests for FCW and LDW provided effectiveness and benefit 
information for each technology and suggest that FCW and LDW have the 
potential to significantly reduce the number of crashes that occur in 
the U.S.\30\
---------------------------------------------------------------------------

    \29\ See http://www.safercar.gov/newcarassessmentenhancements-2007.pdf at page 18, Table 6.
    \30\ LDW effectiveness estimated from data included in NHTSA 
Report No. DOT HS 810 854, Evaluation of a Road Departure Crash 
Warning System, December 2007. FCW effectiveness estimated from data 
included in NHTSA Report No. DOT HS 810 569, Evaluation of an 
Automotive Rear-End Collision Avoidance System, March 2006.
---------------------------------------------------------------------------

    Additionally, NHTSA used data from these field operational tests 
(FOTs), as well as additional agency research, to finalize performance 
tests establishing minimum performance criteria for FCW and LDW so that 
vehicles can be rated on their presence.\31\ For ESC, because it had 
been in the field for some time, we used real-world data to establish 
effectiveness and then used the test procedure which accompanied the 
Final Rule (FMVSS No. 126) to develop a performance test and minimum 
performance criteria.\32\ The table below presents NHTSA's 
effectiveness estimate values for ESC, FCW, and LDW.\33\ A range was 
used for LDW to reflect potential system availability variation due to 
lane marking quality.
---------------------------------------------------------------------------

    \31\ See Docket No. NHTSA-2007-27662 for ESC, LDW, and FCW test 
procedures.
    \32\ See NHTSA Report No. DOT HS 810 794, The Statistical 
Analysis of the Effectiveness of Electronic Stability Control (ESC) 
Systems-Final Report, July 2007. See also 72 FR 17236, Docket No. 
NHTSA-2007-27662.
    \33\ See Appendix B, Effectiveness Estimates for ESC, FCW and 
LDW for a summary explanation of how overall effectiveness estimate 
values were generated.

              Effectiveness Estimates for ESC, FCW, and LDW
------------------------------------------------------------------------
                                                          Effectiveness
                        System                              (percent)
------------------------------------------------------------------------
ESC...................................................                59
FCW...................................................                15
LDW...................................................              6-11
------------------------------------------------------------------------

    NHTSA believes that the FOT results for FCW and LDW are applicable 
for estimating real-world safety benefits since these technologies were 
evaluated in the same real-world driving environment in which they 
would be deployed. In general, in an FOT, the major variables impacting 
a technology's safety benefits, including differences in individual 
driving styles and behavior, system performance, and driver acceptance, 
are taken into account. Likewise, critical safety incidents (i.e. near-
crash incidents that occur during the FOT) data are recorded and 
evaluated to determine if the technology provided a safety benefit in 
terms of critical incident reduction. Assuming a proportional 
relationship between near-crash events and actual crashes, critical 
incident data are further evaluated using statistical methods to 
estimate crash reduction benefits. In the field tests for FCW and LDW 
systems, NHTSA provided technical management and the Volpe National 
Transportation Systems Center performed an independent evaluation to 
estimate safety benefits which included rigorous statistical analysis.
    NHTSA believes that ESC, FCW and LDW are the only crash avoidance 
technologies that meet the agency's criteria for inclusion in a crash 
avoidance rating program at this time. That is, all three address a 
major crash problem, safety benefit projections have been assessed, and 
performance tests and procedures are available to ensure an acceptable 
performance level. The agency acknowledges that many other technologies 
were identified by commenters such as collision mitigation braking 
systems, lane keeping assist systems, and side object detection 
technologies. However, at this time the agency does not have enough 
data to estimate the safety benefits of these systems, and therefore 
will not promote these other technologies at this time.
    Through our current research activities and/or information obtained 
from the automotive industry and the public, the agency anticipates 
that it will gain information on the benefits and performance 
capabilities of other advanced safety technologies. If the agency 
anticipates making changes to the rating system or the technologies 
that the agency has chosen to promote

[[Page 40034]]

as that information is gathered, the agency will seek public input on 
the appropriateness of such changes. At this time, we anticipate using 
similar criteria (addresses a major crash problem, assessed safety 
benefits, and established performance tests and procedures) to 
determine technologies for future program inclusion.
Rating System
    Generally, there was little support for a crash avoidance rating 
system based on a cumulative concept (e.g., the more technology you 
have; the higher the rating). Instead, several commenters preferred 
that the agency develop a rating system based on a computation of 
benefits to be expected from the crash avoidance technologies of a 
rated vehicle. Regardless of approach, these commenters all suggested 
that the agency use a star rating system to inform consumers about the 
presence of advanced technologies. BMW and Mercedes suggested a simpler 
approach whereby technologies would essentially be listed without 
regards to their effectiveness and without summing them into an overall 
rating crash avoidance rating. BMW offered an approach where all 
technologies would all be treated equally but where those technologies 
that had been proven beneficial by real world studies would somehow (in 
their scheme solid green and hollow thumbs were used) be denoted 
differently. Similarly, Mercedes suggested a simple ranking system for 
technologies.
    To gauge consumer understanding and acceptance of these various 
systems, NHTSA tested the cumulative approach, the effectiveness 
approach, and the list approach with groups of consumers.\34\ NHTSA 
conducted four focus group sessions in the DC area with participants 
who had to qualify as either a primary or shared decision maker with 
respect to automobile purchases for their household and intended to 
purchase a new or used automobile in the next two years. Participants 
in both groups were also screened to ensure they had some level of 
concern about the safety of automobiles and the groups represented a 
mix of age, education, and income. The agency tested letters, stars, 
words, check marks, and color schemes (for standard and optional 
availability) depending on which one of the three approaches was being 
tested. The agency also tested a subset of these treatments in an on-
line forum.
---------------------------------------------------------------------------

    \34\ The full study report is available http://www.regulations.gov in Docket No. NHTSA-02004-19104.
---------------------------------------------------------------------------

    With regards to what type of rating system should be used, 
participants overwhelmingly preferred a rating system that was a simple 
list approach. Additionally, focus group participants unanimously 
agreed that the use of colors is not visually appealing to fully 
comprehend what they are viewing. In the treatments tested by the 
agency, single check marks as opposed to multiple check marks to 
indicate a technologies importance were preferred by most participants. 
Additionally, to display and communicate the information, consumers 
stated that a single check mark or the use of text (indicating standard 
or optional) is the most understandable way to illustrate the presence 
of crash prevention technologies, though neither marking was 
overwhelmingly preferred.
    Participants overwhelmingly objected to the multiple checks, star 
markings and A-D grading scale, saying they were very difficult to 
understand, despite having an associated key. Several participants also 
stated that if there were a technology or several technologies that 
were more important than the others, than that should be specifically 
communicated or noted on the layout and inferred, not the use of stars, 
individual letter grades, or multiple check marks.
    The agency believes that the preference for the use of check marks 
or text over the use of an effectiveness approach may be rooted in the 
fact that participants (and to the extent that they are reflective in 
general of new car buyers) may not fully grasp the importance of these 
features. For example, participants generally stated that they think of 
these features as ``nice to haves'' rather than ``must haves'' because 
they are not yet aware of how the features can reduce fatalities. As 
such, the agency intends to continue monitor the public's understanding 
of this new rating program and if necessary change the way in which 
ratings are communicated to the public. For now, based on these focus 
group results, the agency will use text to communicate the standard or 
optional presence of ESC, LDW, and FCW on vehicles.

F. Presentation and Dissemination of Safety Information

    Some commenters encouraged the agency to disseminate additional and 
more sophisticated consumer information but no specific examples were 
given. Most commenters discussed and supported the agency's proposal 
for a combined crashworthiness rating. The agency's analysis and 
decisions on the presentation and dissemination of safety information 
are divided into the following categories: Presentation of Safety 
Information and Combined Crashworthiness Rating.
Presentation of Safety Information
    Some commenters supported consumer education materials such as 
safety tips and safe driving practices. Others suggested that NHTSA 
develop, maintain and make available a database of non-agency sources 
of credible vehicle safety information. Finally, some commenters 
suggested that the agency provide additional information at the point 
of sale (beyond that required by the new labeling program). NHTSA 
agrees with many of these suggestions. NHTSA continuously investigates 
ways to improve marketing the NCAP vehicle ratings program. We will 
place the results of our enhanced marketing studies in Docket No. 
NHTSA-02004-19104, as they are completed.
Combined Crashworthiness Rating
    Most commenters supported an overall crashworthiness rating that 
combined the results from all test conditions. Honda and Toyota 
provided some details but GM and Ford provided very specific 
information on how this new rating could be calculated. Some commenters 
cautioned that an overall rating would overly simplify information for 
consumers, and that it could mislead consumers if poor performance were 
hidden under an umbrella rating. Given the general support for an 
overall rating and the public's desire for simpler information, NHTSA 
is implementing a new overall crashworthiness rating that combines the 
results of the front, side and rollover programs.
    NHTSA will provide a summary crashworthiness rating for each 
vehicle (which we will call the Vehicle Safety Score) plus individual 
scores for each occupant in each crash condition for that vehicle (as a 
set of relative risk measures). This is in accordance with comments 
from Delphi, Public Citizen, Bidez and Associates, and the IIHS who 
expressed concern over individual test results being masked and that 
individual scores in each crash mode should continue to be provided to 
the consumer. Scores for vehicles will be provided to the consumer via 
a star rating system where the new bands for 1 to 5 stars were 
determined by the mean and dispersion of the risk of injury in each 
crash test condition (front and side) and the risk of rollover.
    Although NHTSA's previous proposal did not suggest including the 
rollover risk rating in the crashworthiness rating, the agency has now 
decided to do so.

[[Page 40035]]

The agency's decision to include the rollover rating in the combined 
rating is consistent with the 1996 Transportation Research Board 
recommendation,\35\ and we believe that its inclusion provides a more 
complete summary rating. Below, we describe how the frontal and side 
scores are developed and how these scores are combined with the 
rollover score to create an overall score.
---------------------------------------------------------------------------

    \35\ See Transportation Research Board, Shopping For Safety: 
Providing Consumer Automotive Safety Information. TRB Special Report 
248 (1996).
---------------------------------------------------------------------------

    Consistent with what has already been presented, NHTSA has selected 
the following test conditions, test dummies and injury criteria to 
develop its combined rating:
     One frontal impact crash test (full frontal rigid barrier 
crash test at 35 mph (56 kmph)) with a 50th percentile male Hybrid III 
dummy in the driver position and a 5th percentile female Hybrid III 
dummy in the front passenger seating position.
     One side impact crash test (38.5 mph (62 kmph) with 
NHTSA's moving deformable barrier (MDB) crabbed at 27 degrees into the 
side of vehicle) with an ES-2re dummy in the front seating position and 
a SID-IIs dummy in the rear seating position on the struck side of the 
vehicle.
     An oblique pole impact test (20 mph (32 kmph)) at 75 
degrees into a 25 cm diameter pole including the SID-IIs dummy in the 
front seating position.
     Dynamic maneuvering (fish-hook) rollover test and static 
stability factor (SSF).
     All applicable injury criteria.
     Use of injury risk curves.
a. Combining Injury Risk From Different Body Regions
    The agency has chosen to maintain its current method for combining 
injury metrics for any seating position in its test. That is, the risk 
of injury to each body region are assumed to be independent events and 
can be statistically combined to determine the joint probability of 
injury to the occupant using the following equation: p(A or B) = 
p(A)+p(B)-p(A)*p(B) where A and B are the independent events. Using 
injury risk curves for different body regions, this method results in 
an overall risk of injury for the occupant. For the two adult Hybrid 
III dummies there are four independent events to combine, which are 
injury risk to the head, neck, chest, and femur/knee. For the ES-2re 
dummy, there are also four independent events, which are injury risk to 
the head, chest, abdomen, and pelvis, while for the SID-IIs dummy, 
there are only 2 independent events which are injury risk to the head 
and pelvis.
    In GM's proposal, the normalized injury measures for different body 
regions are combined by weighting each by the proportion of injuries 
associated with each injury measure. The result of this method does not 
represent either an absolute injury risk or a relative injury risk (as 
in NHTSA's method). Therefore, the risk levels of different vehicles 
are not quantifiable. In addition, Ford stated that GM's proposal 
assumes a linear relationship between the dummy response and injury 
risk, when generally the relationship is non-linear. Therefore, Ford 
expressed that GM's proposal could result in an inaccurate estimation 
of the relative vehicle safety performance. NHTSA agrees with this 
assessment and has chosen to use the joint probability of injury 
formula, as it does now, to combine injury risks to different body 
regions for an occupant. However, the agency notes that computation of 
the joint probability requires there to be quality data available for 
all of the injury risks being combined. Similarly, to compute the 
overall summary rating, data must also be available from all of the 
tests to prevent a model from not being rated. As such, the agency has 
included redundant sensor measurement capability in the test dummies 
(where possible), grouped tests (front, side, and rollover) together, 
and worked with our test labs to ensure that they are using the most up 
to date calibration procedures. In this way, we hope to alleviate the 
potential loss of data and subsequently, vehicles with incomplete 
ratings.
b. Risk of Injury by Seating Position and Test Condition
    For each vehicle, the risk of injury is estimated from six test 
results, which are: (1) Driver in frontal crash, (2) passenger in 
frontal crash, (3) driver in side MDB crash, (4) rear seat passenger in 
side MDB crash, (5) driver in oblique pole impact, and (6) rollover 
potential in single-vehicle crashes using rollover test results. Ford 
suggested that the agency combine results using a simple average, but 
GM suggested a weighted approach to combine results.
    To combine the risk of injury by occupant seating position, GM 
suggested weighting based on occupant demographics and the relative 
frequency of exposure by seating position. Ford commented that this 
approach would undervalue NCAP test results for passengers since the 
proportion of drivers is far greater than that of passengers. Ford 
asserted that this method of obtaining the overall injury risk might 
confuse consumers who seek a broader assessment of safety performance 
than one limited to the driver. Ford proposed using the straight 
average of the risks of injury for the driver and the passenger to 
obtain the overall injury risk. NHTSA agrees with Ford's suggested 
approach.
    However, rather than use the percentages calculated from the 
probability of injury results (as is currently done), NHTSA will be 
computing the relative risk for each seating position and each test 
condition. This relative risk measure provides an estimate of an 
occupant's risk of injury compared to a baseline injury risk. The score 
for each occupant in each test condition is computed by dividing the 
overall risk of injury in each test condition by a baseline risk of 
injury. As will be explained below, the baseline risk of injury in each 
test condition is an approximation of the fleet average injury risk for 
that test condition. The baseline risk of injury is set once and reused 
for subsequent model years. This allows cross-year comparisons with 
future fleets.\36\ This operation results in six summary scores for 
each vehicle representing the relative risk of injury for the driver 
and passenger in the frontal crash test and side MDB test, the driver 
in the oblique pole test, and the relative risk for all occupants in 
rollovers with respect to a baseline injury risk. As such, the scores 
indicate how a particular vehicle compares to a baseline risk and these 
are the scores (star ratings) that will be presented to consumers on 
the Web site and in agency publications.
---------------------------------------------------------------------------

    \36\ In the future, the baseline could be adjusted to reflect 
vehicle designs. However, the agency would seek public input on the 
issue before such an adjustment would occur.
---------------------------------------------------------------------------

    To compute a vehicle's overall risk of injury in frontal crash 
tests, NHTSA has decided to use the simple average of the probability 
of injury to the driver and front passenger. The risk of injury to the 
driver in side crashes is calculated as the weighted average of the 
combined probability of injury of the driver in the MDB test (weighted 
by 80 percent) and that of the driver in the oblique pole test 
(weighted by 20 percent). The weights reflect the proportion of belted 
driver fatalities in real-world crashes represented by the MDB and pole 
tests in MY 1999 and newer vehicles (FMVSS No. 214 Final Rule, Docket 
No. NHTSA-2007-29134). The overall risk of injury in side crashes is 
then computed as the average of the risk of injury to the driver in 
side impacts (weighted average from MDB and pole test results) and the 
probability of injury to the rear seat

[[Page 40036]]

passenger in the MDB test. For rollover, in order to combine the risk 
from the rollover test with the risk of injuries obtained from the 
crash test, the agency has assumed that a belted occupant in a single-
vehicle crash p(roll) has the same relative risk of injury as the risk 
of rollover given a single vehicle crash.
    As suggested in Ford's proposal, NHTSA is adopting this method of 
averaging the risk of injury between the driver and the passenger to 
obtain an overall injury risk for each crash mode to ensure equal 
weighting for all seating positions. This is unlike GM's approach of 
applying significantly higher weight to the driver than the passenger 
based on occupancy rates in each seating position. NHTSA believes that 
GM's proposal would not encourage manufacturers to offer advanced 
safety systems to all seating positions, thereby resulting in reduced 
protection to some. This is especially significant in the side MDB 
crash test where the SID-IIs dummy in the rear seat generally 
demonstrates a higher risk of injury than the driver. Under GM's 
approach, the rear seating position would have far less value than the 
driver seating position because the rear seat has a relatively low 
occupancy rate. However, when combining the pole test results with the 
MDB results for the front seat, we do believe that weighting by crash 
test condition is appropriate. In this way, the results from the pole 
tests are proportional to their occurrence and do not mask a vehicles 
performance in the MDB test, possibly providing an inaccurate portrayal 
of the vehicle.
    The figure below graphically illustrates the method of combining 
the different risks.
BILLING CODE 4910-13-P

[[Page 40037]]

[GRAPHIC] [TIFF OMITTED] TN11JY08.004

BILLING CODE 4910-59-C
c. Combined Crashworthiness Rating
    The agency's combined crashworthiness rating, the Vehicle Safety 
Score (VSS), is computed as the weighted average of the three summary 
scores for front, side, and rollover. The weight factors applied (\5/
12\ for frontal crashes, \4/12\ for side crashes, and \3/12\ rollovers) 
reflect the proportion of injuries for belted occupants (in vehicles of 
model year 1999 and later) in each crash mode.\37\ This approach is 
similar to GM's proposal of combining the crash test results using a 
weighted average.
---------------------------------------------------------------------------

    \37\ These model years were chosen to reflect newer vehicle 
designs and to obtain a statistically robust trend from the NASS/CDS 
data.
---------------------------------------------------------------------------

    Since the NCAP frontal crash test involves a vehicle with a fixed 
rigid barrier, it represents a crash between two vehicles of the same 
weight. Therefore, the safety rating from the NCAP frontal crash test 
and the combined crashworthiness rating (which includes the frontal 
crash test results) depends on vehicle mass, and cannot be compared 
across vehicle weight classes. In contrast, on an individual basis, the 
side crash (pole and MDB) test results and the rollover results can be 
compared across vehicle classes.

[[Page 40038]]

d. Determination of Baseline Risk and Star Bands
    NHTSA will continue to use the star rating system to provide an 
individual crashworthiness rating for each seating position, each crash 
mode, and their combination. However under the new system, stars will 
be interpreted differently. Bands for 1 to 5 stars were determined by 
the mean and dispersion of the risk of injury in all three test 
conditions (front, side, and rollover).
    In the NCAP frontal tests, the average risk of injury to the driver 
in all 2008 model year vehicles is 15 percent  5 percent. 
Based on our NCAP injury data for the 50th percentile male seated in 
the right front passenger seat, we expect that a 5th percentile seated 
in that same seating position would have a similar distribution. 
Therefore, the agency selected a baseline injury risk of 15 percent to 
compute the frontal relative risk scores. A relationship between 
relative risk of injury and the number of stars assigned was developed 
using the existing NCAP frontal crash test data for the 50th percentile 
male Hybrid III dummy in the driver seating position.
    To determine the star bands for frontal NCAP, NHTSA selected a 
baseline risk of 15 percent (representing the average risk of injury to 
the driver in MY 2008 vehicles in the NCAP frontal crash test) to serve 
as the break point for the 4 star and 3 star rating. Other criteria 
used to determine the star bands were (1) vehicles performing 
exceptionally well (At 0-15 percentile of vehicles tested) are assigned 
a five star rating, and (2) Vehicles performing very poorly (greater 
than 4 standard deviations from mean) would be assigned a one star. 
Attempts were also made to maintain equidistant star band boundaries. 
Based on these criteria and the distribution of the relative risk of 
injury scores of MY 2008 vehicles, the relationship between the 
Relative Risk Score (RRS) and the number of stars was established, and 
is presented below. The RRS is computed by (1) rounding the injury risk 
to the nearest tenth of a percent in accordance with the rounding-off 
method of ASTM Standard Practice E 29 for Using Significant Digits in 
Test Data to Determine Conformance with Specifications, (2) dividing 
the injury risk by 0.15 (15.0 percent baseline injury risk), (3) and 
finally rounding the result to the nearest one hundredth in accordance 
to ASTM Standard E 29.
    As with frontal NCAP, this same methodology was applied to the 
scores in the side MDB and oblique pole tests as well as the combined 
crashworthiness Vehicle Safety Score. The agency found, for a limited 
number of newer vehicles tested to both the MDB and Pole test, that 
when the MDB test results were combined with the pole test, the average 
risk was 15%. As such, for side NCAP, the combined crashworthiness 
rating also represents the relative risk of injury with respect to an 
injury risk of 15 percent.

      Relationship Between the Relative Risk and the Star Bands for Front and Side Crash Tests Using 15 Percent Risk of Injury as the Fleet Average
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                             5 stars                 4 stars                3 stars                2 stars                 1 star
--------------------------------------------------------------------------------------------------------------------------------------------------------
RRS Values.........................  RRS < 0.67............  0.67 <= RRS < 1.00      1.00 <= RRS < 1.33     1.33 <= RRS < 2.67.    RRS >= 2.67.
Probability........................  P < 0.100.............  0.100 <= P < 0.150      0.150 <= P < 0.200     0.200 <= P < 0.400     P >= 0.400.
--------------------------------------------------------------------------------------------------------------------------------------------------------

    Similarly for rollover, we selected a baseline risk of 15 percent 
for the risk of rollover, which produces the relative risk measures 
shown below.\38\
---------------------------------------------------------------------------

    \38\ See Appendix D, Probability of Injury, Vehicle Safety 
Score, and the Star Rating System.

  Current NCAP Star Rating in Rollover and Its Relationship With the Relative Risk in Rollover Using 15 Percent
                                         Risk of Injury as the Baseline
----------------------------------------------------------------------------------------------------------------
          Number of stars                       Risk of rollover               Relative risk score in rollover
----------------------------------------------------------------------------------------------------------------
1 star.............................  P >=0.40.............................  RRS >=2.67.
2 stars............................  0.30 <=P <0.40.......................  2.00 <=RRS <2.67.
3 stars............................  0.20 <=P <0.30.......................  1.33 <=RRS <2.00.
4 stars............................  0.10 <=P <0.20.......................  0.67 <=RRS <1.33.
5 stars............................  P <0.10..............................  P <0.67.
----------------------------------------------------------------------------------------------------------------

G. Manufacturer Self-Certification

    Several commenters suggested that NHTSA consider a self-
certification process in which NHTSA would oversee the testing 
conducted by the manufacturer. However, it seems possible that 
manufacturers could run several tests and report only the best results; 
or because manufacturers would know exactly what vehicle was being 
tested, the vehicle's star ratings might not be indicative of a random 
sample (as currently done by the agency). Additionally, because NHTSA 
does not currently have the resources to conduct oversight over a 
manufacturer's test facility, dummy certification and test setup, a 
manufacturer's facilities might take more liberty than agency contract 
laboratories in their testing procedures.
    These issues do not affect a manufacturer's self-certification of 
compliance with the Federal motor vehicle safety standards. A 
manufacturer had a legal duty to report any non-compliance promptly to 
NHTSA. They must also recall and remedy without charge to the purchaser 
any vehicle that fails to comply with an applicable safety standard. 
The manufacturer also is subject to additional penalties if it cannot 
demonstrate that it had no reason to know, despite exercising 
reasonable care, that the vehicle did not comply with the standard. 
These are all express provisions of Title 49, Chapter 301 of the United 
States Code. There are no

[[Page 40039]]

parallel provisions for the New Car Assessment Program.
    In addition, one of the primary reasons for allowing manufacturer 
self-certification in NCAP was to allow information about new vehicles 
to be provided more quickly. In this case, NHTSA has had an optional 
NCAP test program in place for nearly 20 years. This allows 
manufacturers to request a test of new or redesigned vehicles and get 
the NCAP information out quickly to the public. Given these 
considerations, NHTSA is not adopting the suggestions to permit 
manufacturer self-certification of NCAP results.

H. Other Recommendations

    Several commenters, in their responses to the notice and at the 
public hearing, presented other recommendations for the agency's 
consideration. NHTSA has decided not to adopt any of these 
recommendations at this time for the reasons outlined below.
Compatibility Assessment
    Some commenters recommended front-to-front compatibility 
assessments, while others suggested vehicle aggressivity evaluations 
for frontal NCAP. These commenters did not provide (and NHTSA is not 
aware of) any data that would support an NCAP compatibility evaluation 
at this time. The agency has a research program in this area and should 
a valid compatibility metric emerge from that research, the agency will 
consider it at that time.
Child Restraints
    Some commenters suggested that the agency test and rate child 
restraints either in the vehicle and/or on a sled test. NHTSA has 
examined this in the past and at that time concluded that: (1) A 
dynamic rating for a child restraint system (CRS) was not feasible; (2) 
the agency wanted to focus on ease of use ratings; and (3) limited in-
vehicle testing with a six-year old dummy did not correlate with real-
world data.\39\ However, the agency has continued to investigate CRS 
and child dummy performance in the current NCAP test environment, and 
their correlation to injury risks for children in real-world crashes. 
The agency will take actions at such time as the test results and 
analyses can be used to support such a rating program.
---------------------------------------------------------------------------

    \39\ See 70 FR 29815, Docket No. NHTSA-2004-18682.
---------------------------------------------------------------------------

Increased Test Speed
    Two commenters and most automobile manufacturers stated that 
increased test speeds in frontal NCAP would promote stiffer vehicle 
designs and more aggressive restraints. NHTSA agrees that without an 
appropriate measure of vehicle stiffness, a higher speed test could 
lead to more aggressive vehicle designs. Therefore, NHTSA has decided 
not to adopt a 40 mph (64 kmph) frontal NCAP test because of concerns 
about vehicle compatibility, the lack of test data, and no clear 
understanding of potential countermeasures that could be used by 
manufacturers to achieve the top rating. In addition, the agency notes 
that the current frontal NCAP test speed represents 99 percent of all 
crashes, and increasing the test speed would not address a large 
portion of real-world crashes.
Lighting
    Some commenters recommended that NHTSA incorporate a lighting/
visibility program into NCAP to address vehicle blind spots and glare. 
The commenters did not provide (and NHTSA does not believe that there 
is) sufficient data to justify incorporating a lighting or visibility 
measure into NCAP at this time. The agency is conducting research in 
both of these areas to better assess the safety problem and explore 
what approaches and/or countermeasures should be considered. Therefore, 
NHTSA has decided not to incorporate an NCAP rating for lighting or 
visibility at this time.
Frontal Offset Test
    Some commenters encouraged the incorporation of a frontal offset 
test into frontal NCAP. However, others did not support an offset test 
stating that such a test did not provide sufficient benefit to 
consumers or that it was already being done by others (e.g., IIHS). 
NHTSA has been studying the offset test procedure, but we continue to 
believe that further research and analysis is needed to ensure that 
improved occupant protection is provided by such a test without 
potential unintended consequences such as increased vehicle stiffness 
and aggressivity.
Pedestrians
    Some commenters encouraged NHTSA to pursue opportunities to improve 
pedestrian safety through NCAP. The agency has no pedestrian standard 
at this time. While NHTSA is actively engaged in the development of a 
Global Technical Regulation on pedestrian safety, we feel it would be 
premature to develop a rating program before the details, test protocol 
and potential benefits of this activity have been resolved. Therefore, 
we are not incorporating pedestrian rating into NCAP at this time.
Frontal Pole Test
    A frontal pole test was suggested by two commenters and 
specifically opposed by one. While the real-world data presented by the 
IIHS seems to imply that a number of fatalities and injuries are 
occurring in narrow object frontal impacts, at this time NHTSA is 
unclear as to what countermeasures might be developed. Similarly, a 
significant amount of research would need to be conducted to establish 
a new frontal impact pole test for NCAP. Accordingly, the agency is not 
adopting this proposal at this time.

I. Monroney Label

    On August 10, 2005, the President signed into law the Safe, 
Accountable, Flexible, Efficient Transportation Equity Act: A Legacy 
for Users (SAFETEA-LU). Section 10307 of the Act requires new passenger 
automobiles to have NCAP safety ratings displayed on the price sticker, 
known as the Monroney label. As required by SAFETEA-LU, on September 
12, 2006 (71 FR 53572), NHTSA published a final rule implementing this 
statutory requirement, including prescribing the form, required 
information, and layout of the label. The rule, set forth at 49 CFR 
part 575.301, applied to covered vehicles manufactured on or after 
September 1, 2007.
    Regulation 575.301 specifies the required information for the NCAP 
front, side and rollover tests. For the frontal crash, there are two 
separate ratings, one for the driver and one for the right front 
passenger. Similarly, two separate ratings are established for the side 
crash, one for the front seat and one for the rear seat. One rating is 
provided for rollover.
    Under our regulation, front, side and rollover NCAP ratings must be 
placed on new vehicles manufactured 30 or more days after the 
manufacturer receives notification from NHTSA of the ratings. As 
explained earlier in this notice, in addition to any overall rating, 
the agency will still make available on http://www.safercar.gov the 
individual seating position results for each crash condition (front, 
side pole, and side MDB) and for side NCAP, the front seat and rear 
seat score developed from the combination of the pole and MDB test 
results. However, the agency is using this notice to inform 
manufacturers and other interested persons of our intent to use the new 
combined side impact score

[[Page 40040]]

developed from the pole and MDB tests for the Monroney label. In 
addition, we will initiate rulemaking to change the format and/or the 
layout of the Monroney label to incorporate the new overall combined 
crashworthiness rating. We believe that the combined rating and the new 
side impact score will provide consumers with the information they need 
to make comparative judgments on new vehicles.
    When we issue the notice of proposed rulemaking, we will address 
relevant issues including changing the layout and format of the label 
to incorporate this new, additional information and to address other 
labeling issues such as the lead time necessary for the manufacturers 
to update their labeling operations.

V. Conclusion

    NHTSA will implement these decisions regarding enhancements to NCAP 
beginning with MY 2010 vehicles. For that model year, the agency will 
make changes to its existing front and side testing activities 
requiring all vehicles to be rated using these new protocols. With 
regards to the frontal crash test program, NHTSA will maintain the 35 
mph (56 kmph) full frontal barrier test protocol but will incorporate 
the following body injury criteria: Head (HIC15), neck (Nij, 
tension, and compression), chest (deflection), and femur (axial force). 
The agency will also add the 5th percentile female Hybrid III dummy in 
the right front seating position. For side impact, NHTSA will maintain 
the current moving deformable barrier test at 38.5 mph (63 kmph) but 
will update that test to include head (HIC36), chest 
(deflection), abdomen (force), and pelvic (force) injury criteria for 
the ES-2re and, consistent with the safety standard, HIC36 
and pelvic (force) for the SID-IIs dummy. For the MDB test, the 50th 
percentile male ES-2re dummy will be used for the driver position and 
the 5th percentile SID-IIs dummy for the rear seated passenger 
position. Additionally, vehicles will also be assessed using a new 
oblique pole test and a 5th percentile female dummy in the driver 
position, using HIC36 and pelvic (force). For rollover, the 
agency will continue to rate vehicles for rollover propensity, but will 
wait to update its rollover risk model to allow for more real-world 
crash data of vehicles equipped with electronic stability control.
    For MY 2010, the agency will also implement a new crash avoidance 
program that will rate vehicles on the presence of select advanced 
technologies and a new overall Vehicle Safety Score that will combine 
the star ratings from the front, side, and rollover programs.

Appendix A

NCAP and IIHS Pole Test Results

                                                                          NHTSA
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                               Combined
                                                                                                                                Lower spine   acetabulum
               Vehicle                       Vehicle class               SAB type            Driver test dummy        HIC36      accel (Gs)    & iliac
                                                                                                                                              force (N)
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                          IARV Limits............         1000           82         5525
2007 Honda Pilot.....................  SUV.....................  Curtain + Torso........  SIDIIs.................         3464           68         6649
2007 Nissan Quest....................  Van.....................  Curtain................  SIDIIs.................         5694           79         5786
2007 Ford Escape.....................  SUV.....................  Curtain + Torso........  SIDIIs.................          407           65         6515
2006 VW Passat.......................  Medium PC...............  Curtain + Torso........  SIDIIs.................          323           40         3778
2006 Subaru Impreza..................  Medium PC...............  Combo..................  SIDIIs.................          184           58         4377
2007 Toyota Avalon...................  Heavy PC................  Curtain + Torso........  SIDIIs.................          642           62         6672
--------------------------------------------------------------------------------------------------------------------------------------------------------


------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                   Combined
                                                                        Driver test               acetabulum     Overall                                                  Head       Structure/
            Vehicle                Vehicle class        SAB type           dummy         HIC15      & iliac      rating       Head/neck       Torso      Pelvis/leg    protection    safety cage
                                                                                                   force (N)
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
2007 Honda Pilot...............  SUV.............  Curtain + Torso...  SID-IIs......         167        4700  G             G             G             G             G             A
2007 Nissan Quest..............  Van.............  Curtain + Torso...  SID-IIs......         207        2900  G             G             G             G             G             A
2007 Ford Escape...............  SUV.............  Curtain + Torso...  SID-IIs......         216        5600  G             G             G             A             G             A
2006 VW Passat.................  Medium PC.......  Curtain + Torso...  SID-IIs......         168        3300  G             G             G             G             G             G
2006 Subaru Impreza............  Medium PC.......  Combo.............  SID-IIs......         325        5100  G             G             G             G             G             A
2007 Toyota Avalon.............  Heavy PC........  Curtain + Torso...  SID-IIs......         350        4100  G             G             A             G             G             A
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Appendix B

Effectiveness Estimates for ESC, FCW and LDW

Electronic Stability Control (ESC)

    This effectiveness estimate comes from the report: Statistical 
Analysis of the Effectiveness of Electronic Stability Control (ESC) 
Systems--Final Report. Report No. DOT HS 810 794, July 2007.
    From the Executive Summary, page vii, for Road Departure--Police 
Reported Crashes:
     The effectiveness of ESC for passenger cars = 45% 
(weighting for the difference in crash reporting among the States).
     The effectiveness of ESC for Light Trucks and Vans 
(LTV's) = 72% (weighting for the difference in crash reporting among 
the States).
     Assuming an equal weighting between passenger cars and 
LTVs, the average effectiveness = 59% for Road Departure Crashes.
    59% was assumed to be a best overall effectiveness estimate for 
road departure crashes.

Forward Collision Warning (FCW)

    Based on field operational test (FOT) data from the Automotive 
Rear-End Collision Avoidance FOT (ACAS FOT) collected from 66 
participants who each drove an FCW-equipped vehicle for 3 weeks, it 
was estimated that the FCW system has the potential to reduce about 
15% of all rear-end crashes. The FCW system integrated rear-end 
crash warning function with adaptive cruise control function. This 
system becomes operational when vehicle speed exceeds 25 mph and 
disengages when the speed falls below 20 mph. The participants 
accumulated 98,000 miles of driving data. The FCW system operated in 
the background during the first week of the FOT, providing 
information about baseline driving. The final 2 weeks of the FOT 
generated information

[[Page 40041]]

about driver performance with the FCW system while it operated in 
the foreground.
    FCW system effectiveness was estimated separately in each of 
nine driving conditions based on FOT data, which combined three 
driving states (lead vehicle stopped, lead vehicle decelerating, and 
slower constant-speed lead vehicle) and three travel speed bins 
(<25, between 25 and 35, and >=35 mph). Total system effectiveness 
was derived by integrating individual system effectiveness estimates 
in the nine driving conditions using corresponding rear-end crash 
data from the GES (see Equation (6) in Section 4.2.2.3 on page 4-
70). Based on available FOT data, the FCW has shown crash prevention 
potential in lead vehicle stopped at speeds over 25 mph, slower 
constant-speed lead vehicle at speeds below 25 and over 35 mph, and 
lead vehicle decelerating at speeds over 35 mph (see Table 4-32 on 
page 4-73). Using corresponding crash data by travel speed only (not 
taking into account crash data by attempted avoidance maneuver), 
total system effectiveness was estimated at 95% of all 
rear-end crashes (see Figure 4-42 on page 4-74). However, GES crash 
data on travel speed are unreliable since the travel speed variable 
is coded as ``unknown'' in over 70% of the rear-end crash cases. As 
an alternative to travel speed, it is recommended that the posted 
limit data be used to break down the rear-end crash data. Thus by 
using corresponding crash data by posted speed limit, total system 
effectiveness was estimated at 1511% of all rear-end 
crashes assuming that crash-involved vehicles were traveling at the 
posted speed limits reported in the crash database (see Figure 4-42 
on page 4-74). This safety benefit also assumes 100% system 
deployment in the vehicle fleet.
    15% was assumed to be a best overall effectiveness estimate for 
rear-end crash prevention.

Reference

    Najm, W.G., Stearns, M.D., Howarth, H., Koopmann, J., and Hitz, 
J., ``Evaluation of an Automotive Rear-End Collision Avoidance 
System''. U.S. Department of Transportation, National Highway 
Traffic Safety Administration, DOT HS 810 569, March 2006.

Lane Departure Warning (LDW)

    The overall average crash reduction estimate range (6% to 11%) 
for Lane Departure Warning was obtained from data collected during a 
Road Departure Collision Warning (RDCW) System Field operational 
test (FOT). The system merged and arbitrated warnings between a lane 
departure warning system (referred to as a lateral drift warning 
function in the study) and Curve speed warning (CSW) function. LDW 
monitored the vehicle's lane position, lateral speed and available 
maneuvering room. The CSW monitored the vehicle's speed and upcoming 
road curvature.
    The RDCW Evaluation Final Report \1\ discusses numerous safety-
related benefits that resulted during the treatment period, when the 
RDCW alerts were enabled. Most safety benefits were accrued by the 
LDW portion of the RDCW system. These benefits include increased 
turn signal use, improved lane keeping, and fewer crossings of a 
solid lane marker at speeds above 55 mph. However, only one of these 
benefits--fewer crossing of a solid lane marker--was used to 
forecast a reduction in road-departure crashes. Solid lane markers 
serve as the road boundary. During the treatment period and at 
speeds above 55 mph, drivers crossed solid lane markers 44 percent 
less often than they did in the baseline period, when RDCW alerts 
were not enabled. This reduction, weighted by the national departure 
crash counts at this speed range, resulted in a forecasted reduction 
in road-departure crashes.
    Road-departure crash statistics presented in Section 4.1 of the 
RDCW Evaluation Report.\1\

                       Table 4-1.--Road-Departure Precrash Scenarios (Thousands) GES 2003
                                                [Critical event]
----------------------------------------------------------------------------------------------------------------
                                                                Departed       Lost
                                         Vehicle  movement     road edge     control       Other      Row totals
----------------------------------------------------------------------------------------------------------------
Count...............................  ......................          261          208  ...........          469
Row Percent.........................  Going Straight........         55.7         44.3  ...........  ...........
Percent.............................  ......................         25.4         20.3  ...........         45.7
Count...............................  ......................          116          172  ...........          288
Row Percent.........................  Negotiating a Curve...         40.3         59.7  ...........  ...........
Percent.............................  ......................         11.3         16.7  ...........         28.0
Count...............................  ......................           65           55  ...........          120
Row Percent.........................  Initiating a Maneuver.         54.2         45.8  ...........  ...........
Percent.............................  ......................          6.3          5.4  ...........         11.7
Count...............................  ......................  ...........  ...........          150          150
Percent.............................  Other.................  ...........  ...........  ...........         14.6
Count...............................  ......................          442          435          150        1,027
Percent.............................  All Groups............         43.0         42.4         14.6  ...........
----------------------------------------------------------------------------------------------------------------

    From section 4.4.1, this results in an estimated 9,372 to 74,844 
fewer road-departure crashes each year. The average of this range 
equals 42,108. This range is based on full LDW availability.

Effectiveness = collisions avoided/collision population

    Collision population originates from two departure road edge 
cells in Table 4-1, and equals 377,000 crashes. With full 
availability, the effectiveness equals:
[GRAPHIC] [TIFF OMITTED] TN11JY08.008

    With the 56% availability observed in the FOT, the estimated 
effectiveness estimated is (.56)(.11) = 6%.
    Since system availability may vary depending on the quality of 
lane markings, a range of 6 to 11% was assumed to be the best 
overall effectiveness estimate for crashes caused by lane drift.

Reference

    [1] Wilson, B.H., Stearns, M.D., Koopman, J., Yang, D., 
``Evaluation of a Road Departure Crash Warning System''. U.S. 
Department of Transportation, National Highway Traffic Safety 
Administration, DOT HS 810 854, December 2007.

Appendix C

Injury Risk Curves for the NCAP Combined Crashworthiness Rating System

    This Appendix presents the injury risk curves for various body 
regions applicable to the Hybrid III 50th percentile male (HIII 50M) 
and the Hybrid III 5th percentile female (HIII 5F) dummies in 
frontal crash tests and the ES-2re and the SID-IIs side impact 
dummies in lateral crash tests.

Injury Risk Curves for Frontal NCAP Head

    The head injury criterion (HIC15) as a metric for assessing head 
injury risk is well established and in use in FMVSS No. 208 
(Eppinger et al., 1999).

[[Page 40042]]

[GRAPHIC] [TIFF OMITTED] TN11JY08.009

Where [Phi] = cumulative normal distribution

    The AIS 3+ head injury risk curve from the FMVSS No. 208 
Advanced Airbag Final Economic Assessment was extended from the 
Hertz (1993) AIS 2+ head injury risk curve using real-world data to 
determine the relative incidence of different severity brain 
injuries. Since NHTSA will assess the risk of serious or more severe 
head injuries, this equation has been selected for use in NCAP 
(Equation 1). Due to the uncertainty in the scaling methods, NHTSA 
took the conservative approach in estimating head injury assessment 
reference values for the HIII 5F dummy. As such, this equation will 
also be used to assess the risk of AIS 3+ head injury for the HIII 
5F dummies.

Neck

    The risk of AIS 3+ neck injury is assessed using Nij (Equation 
2) as described in Eppinger et al. (1999, 2000) and currently used 
in FMVSS No. 208. The equation below presents the Nij formulation 
and Table 1 presents the intercept values (from FMVSS No. 208) of 
Fint and Mint used in Nij.
[GRAPHIC] [TIFF OMITTED] TN11JY08.010

Where Fz is the axial force and My is the flexion/extension moment 
measured in the upper neck load cell.

       Table 1.--Nij Intercept Values and Tension/Compression Limits for In-Position 50th Percentile Adult Male and 5th Percentile Female Dummies
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                      Nij intercepts
              Dummy              -----------------------------------------------------------------------------------------------------------------------
                                        Tension           Compression           Tension           Compression           Flexion            Extension
--------------------------------------------------------------------------------------------------------------------------------------------------------
HIII 50M........................  4170 N............  4000 N............  6806 N............  6160 N............  310 Nm............  135 Nm.
HIII 5F.........................  2620 N............  2520 N............  4287 N............  3880 N............  155 Nm............  67 Nm.
--------------------------------------------------------------------------------------------------------------------------------------------------------

    In general, neck injuries occur due to combination loading to 
in-position occupants. As such, the Nij injury risk curve is 
applicable and the agency has selected the risk curve used in the 
establishment of the Advanced Air Bag rule for FMVSS No. 208 from 
Eppinger. The neck tension injury risk curve was developed using the 
same paired pig and dummy test data used for the development of Nij. 
NHTSA assumed that the tensile neck tolerance is approximately equal 
to the compressive neck tolerance. Therefore, the injury risk curve 
for neck tension can also be applied to obtain neck injury risk due 
to neck compression. Equations 3-5 present the risk of AIS 3+ neck 
injury as a function of Nij, neck tension, and neck compression for 
the HIII 50M and HIII 5F dummies.
[GRAPHIC] [TIFF OMITTED] TN11JY08.011

Where tension--or--compression is in kV.

    The risk of AIS 3+ neck injury in the NCAP frontal crash test is 
the greater of the injury risk for Nij, neck tension, and neck 
compression. In general, the risk of injury obtained from Nij is 
higher than that for neck tension or compression in frontal NCAP 
tests.

Chest

    Eppinger et al. (1999) developed injury risk curves for chest 
deflection. However, the derived injury risk curve was independent 
of occupant age and was not adequately adjusted to reflect real-
world chest injury risk. As such, we have chosen to use a more 
recent, peer reviewed thoracic injury risk curve using chest 
deflection. Laituri et al. (2003, 2005) developed AIS 3+ thoracic 
injury risk curves by analyzing published cadaveric sled test data 
and then developing a transfer function between dummy chest 
deflection measurements and cadaveric chest deflection under similar 
impact conditions. The resulting thoracic injury risk curve is based 
on dummy measured chest deflection and occupant age and was 
evaluated against real world injury risk in frontal crashes. In 
order to apply this AIS 3+ thoracic injury risk curve in NCAP, it 
was normalized to the average age of the driving population which is 
approximately 35 years. The injury risk curve based on this 
evaluation for assessing risk of AIS 3+ chest injury is presented in 
Equation 6 for the Hybrid III 50th percentile male dummy. The injury 
risk curve as a function of chest deflection (Equation 7) for the 
HIII 5th percentile female dummy (HIII 5F) is obtained by scaling 
the risk curves for the HIII 50M using the scale factor for chest 
deflection (=0.817) which is the ratio of the chest depth of a 5th 
percentile female to that of a 50th percentile male (Eppinger (1999) 
and Mertz (2003)).

[[Page 40043]]

[GRAPHIC] [TIFF OMITTED] TN11JY08.012

Knee-Thigh-Hip

    The injury risk curve that the agency will use for the Knee-
Thigh-Hip (KTH) is the same as that reported by Eppinger et al. 
(1999) in support of FMVSS No. 208 (Equation 8). The injury risk 
curves represent femur and knee injury risk since most of the 
injuries in the datasets that were used to develop these injury risk 
curves were to the distal femur and knee and only four of the 126 
tests used to develop these risks curves produced a hip fracture. In 
addition, the knee injuries in this dataset were primarily 
multifragmentary patellar fractures, which, like other articular 
surface injuries, are associated with a high level of long-term 
disability.
    The femur injury risk curve as a function of femur axial force 
for the HIII 5th percentile female dummy (HIII 5F) was developed by 
scaling the risk curves for the HIII 50M using a scale factor of 
0.68 (Equation 9). This scale factor was proposed by Eppinger (1999) 
and later by Mertz (2003) and is based on the ratio of the thigh 
circumference of a 5th percentile female to that of a 50th 
percentile male.
[GRAPHIC] [TIFF OMITTED] TN11JY08.013

Joint Probability of Injury

    The joint probability of injury to an occupant is obtained by 
combining the risk of injury to each body region assuming the injury 
to different body regions are independent events. Therefore the 
probability of serious injury, Pjoint, is given by:
[GRAPHIC] [TIFF OMITTED] TN11JY08.014

Injury Risk Curves for Side NCAP

    The injury risk curves for the side impact dummies, ES-2re and 
SID-IIs (Kuppa, 2006), were developed from biomechanical tests 
involving human cadaveric subjects and detailed in NHTSA docket 
(NHTSA-2007-29134).

Head

    The Head Injury Criterion (HIC), used for assessing injury risk 
in frontal impacts is based on repeated drop tests of embalmed human 
cadavers onto rigid and padded surfaces where the impact area was 
the forehead (Lissner et al. 1960, Hodgson et al. 1972). Though 
forehead impacts are representative of a frontal impact scenario, 
the ECE R95 directive and Euro NCAP continue to apply HIC for head 
injury assessment in lateral impact scenarios, implicitly assuming 
that the head/brain injury tolerance is independent of loading 
direction and impact location. Similarly, NHTSA applied HIC36 to 
assess head/brain injuries in lateral crashes in the upgrade to 
FMVSS No. 214 so as to harmonize with the existing FMVSS No. 201 
optional pole impact test.
    Therefore, the FMVSS No. 208 AIS 3+ injury risk function 
presented above for the HIII 50M and HIII 5F dummies will be used in 
the NCAP side impact tests with the ES-2re and SID-IIs dummies. 
However, in order to be consistent with FMVSS No. 214, HIC36 will be 
used rather than HIC15 (Equation 10).
[GRAPHIC] [TIFF OMITTED] TN11JY08.015

Where [Phi] = cumulative normal distribution

Chest

    The risk of AIS 3+ and AIS 4+ thoracic injury for a 45 year old 
(average age of the driving population involved in side impacts) 
50th percentile adult male occupant as a function of maximum rib 
deflection of the ES-2re side impact dummy was developed by Kuppa 
(2006) by considering the injury severity to be a polychotomous 
variable (AIS<3, AIS=3, AIS>3). However, this AIS 3+ injury risk 
curve has a finite risk of injury even at zero mm of rib deflection. 
The same cadaver and dummy test data reported by Kuppa (2006) were 
reanalyzed considering the injury severity to be dichotomous (AIS<3 
and AIS>=3 or AIS<4 and AIS>=4) to develop new AIS 3+ and AIS 4+ 
injury risk curves. Since the injury risk curves have not been 
adjusted to represent the average risk of injury in real world side 
crashes, NHTSA will use the AIS 4+ injury risk curve as the 
corresponding AIS 3+ injury risk in NCAP. The risk of AIS 3+ 
thoracic injury for a 45 year old (average age of the driving 
population involved in side impacts) 50th percentile adult male 
occupant as a function of maximum rib deflection of the ES-2re for 
use in NCAP is presented in Equation 11.
[GRAPHIC] [TIFF OMITTED] TN11JY08.016


[[Page 40044]]


    FMVSS 214 final rule does not utilize rib deflection measures of 
the SID IIs dummy and so they are not considered in NCAP at this 
time. Additionally, because the agency does not have a valid risk 
curve at this time for spine acceleration, it is also not included.

Abdomen

    The AIS 3+ abdominal injury risk curve using the total force in 
the ES-2re abdomen reported by Kuppa (2006) is utilized in NCAP and 
is presented in Equation 12.
[GRAPHIC] [TIFF OMITTED] TN11JY08.017

Where F is the total force in the ES-2re abdomen in Newtons.

    Since FMVSS No. 214 does not utilize the abdominal rib 
deflection measures of the SID-IIs dummy for injury assessment, no 
abdominal injury risk assessment will be applied to the NCAP side 
MDB test and the oblique pole test using the SID IIs dummy.

Pelvis

    NHTSA will utilize the AIS 3+ pelvic injury risk curve (Equation 
13) reported by Kuppa (2006) for injury assessment with the ES-2re 
driver in the side MDB NCAP test.
[GRAPHIC] [TIFF OMITTED] TN11JY08.018

Where F is the pubic force in the ES-2re dummy in Newtons

    Kuppa (2006) developed the risk curve for AIS 2+ pelvic fracture 
as a function of the sum of iliac wing and acetabular force in the 
SID-IIs by scaling the normalized 50th percentile male data to that 
of a 5th percentile female, accounting for older subject age, 
adjusting for lower bone tolerance among female occupants, and 
transforming the applied force on the cadaver to the sum of 
acetabular and iliac force measured in the SID-IIs dummy. This 
pelvic injury risk function for the SID-IIs is presented in Equation 
14.
[GRAPHIC] [TIFF OMITTED] TN11JY08.019

Where F is the sum of the acetabular and iliac force in the SID-IIs 
dummy in Newtons

    In developing the pelvis injury criteria for the SID-IIs, an 
occupant age of 56 years was considered to correspond to the average 
age of AIS 3+ injured occupants (of height less than 5 ft 4 inches) 
involved in side crashes. Research has indicated that pelvic 
injuries to older occupants are associated with increased mortality 
(O'Brien et al. 2002; Henry et al. 2002). During a 5-year period, O' 
Brien et al. and Henry et al. examined patients who sustained a 
pelvic fracture and found that patients 55 years and older were more 
likely to sustain a lateral compression fracture pattern and had a 
higher frequency of mortality due to the injury than younger 
patients (<55 years old). Due to the higher mortality rate 
associated with the elderly, an AIS 2+ injury risk curve is used in 
NCAP for the SID-IIs representing a 56 year old small female rather 
than the AIS 3+ injury risk specified for the ES-2re dummy

Joint Probability of Injury

    The joint risk of injury to an occupant is obtained by combining 
the risk of injury to the head, chest, abdomen and pelvis assuming 
the injury to different body regions are independent events (as was 
done for frontal impact). Note that for the SID-IIs, the risk of 
chest and abdomen injury is omitted and only the risk of injury to 
the head and pelvis are combined.
[GRAPHIC] [TIFF OMITTED] TN11JY08.020

Injury Risk In Rollover Crashes

    The Static Stability Factor (SSF) of a vehicle is defined as 
one-half the track width, t, divided by h, the height of the center 
of gravity above the road (SSF = t/(2 x h)). Since 2004, the NCAP 
vehicle rollover rating has been calculated as a function of the 
vehicle's static stability factor and its propensity to tip up in 
the dynamic rollover ``fishhook'' test (68 FR 59250). The risk of 
rollover in single-vehicle crashes as a function of the static 
stability factor and the results of the dynamic rollover test was 
estimated from the State Data System and is presented below in 
Equations 15 and 16.
[GRAPHIC] [TIFF OMITTED] TN11JY08.021

Where SSF=static stability factor

    This model describes the absolute risk of rollover given a 
single-vehicle crash. We can also describe the risk of rollover 
relative to an ``average'' vehicle. For example, we could use a 
``typical'' SSF (which is about 1.35 for the current fleet) for 
vehicles that did not tip up in the dynamic test (which reflects the 
future in the sense that when all vehicles are equipped with ESC 
there will be essentially no tip-ups in the dynamic test). The risk 
of rollover for a subject vehicle compared to the risk of rollover 
for this baseline case describes how much more or less likely the 
subject vehicle is to roll over compared to the baseline. Thus, for 
example, a relative risk of rollover of 0.80 means that the subject 
vehicle is 20 percent less likely to roll over than the baseline; a 
relative risk of 1.25 means that the subject vehicle is 25 percent 
more likely to roll over than the baseline. For certain purposes 
(specifically, in producing the Vehicle Safety Score as described 
elsewhere in this Notice), we treat this as equivalent to the 
relative risk that a belted occupant is injured in a rollover crash 
given a single-vehicle crash. This is not strictly true, but our 
review of the SDS data for belted drivers indicates that it is 
approximately true. Therefore, the relative risks of injury to a 
belted driver in a rollover crash conditional on being involved in a 
single-vehicle crash are approximately proportional to the risks of 
rollover outlined above.

REFERENCES

    AAM Association of Automobile Manufacturers (1999), Comments to 
the Supplemental Notice of Proposed Rulemaking FMVSS No. 208 
Occupant Crash Protection--Air bags, NHTSA Docket No. NHTSA-1999-
6407-40.
    Backaitis SH and St. Laurent A. (1986) Chest Deflection 
Characteristics of Volunteers and Hybrid III Dummies. Proceedings of 
the Thirtieth Stapp Car Crash Conference, pp 157-166. SAE Paper No. 
861884.
    Bouquet, R., Ramet, M., Bermond, F., Vyes, C. (1998) Pelvic 
Human Response to Lateral Impact, 16th International Technical 
Conference on the Enhanced Safety of Vehicles, Paper No. 98-S7-W-16, 
National Highway Traffic Administration, Windsor, 1998.
    Cesari D and Bouquet R. (1990) Behavior of Human Surrogates 
under Belt Loading. Proceedings of the Thirty-Fourth Stapp Car Crash 
Conference, pp 73-82. SAE Paper No. 902310.
    Eppinger et al., (1999) Development of Improved Injury Criteria 
for the Assessment of Advanced Automotive Restraint Systems II, 
NHTSA Docket No. NHTSA-1999-6407-5.
    Eppinger et al. (2000) Supplement: Development of Improved 
Injury Criteria for the Assessment of Advanced Automotive

[[Page 40045]]

Restraint Systems II, NHTSA Docket No. NHTSA-2000-7013-3.
    Hertz E. (1993) A Note on the Head Injury Criteria (HIC) as a 
Predictor of the Risk of Skull Fracture. 37th Annual Proceedings of 
the Association for the Advancement of Automotive Medicine.
    Hodgson, V.R. and Thomas, L. M. (1972) Effect of Long Duration 
Impact on Head, SAE 72096, Sixteenth Stapp Car Crash Conference, 
Society of Automotive Engineers, Warrendale, PA.
    Horsch JD, et al. (1991) Thoracic Injury Assessment of Belt 
Restraint Systems Based on Hybrid III Chest Compression. SAE Paper 
No. 912895, Thirty-Fifth Stapp car Crash Conference, pp 85-108.
    International Standards Organization (ISO) Working Group 6, ISO 
TC 22/SC 12/ WG 6, Road Vehicles--Injury Risk Curves to Evaluate 
Occupant Protection in Side Impact, ISO/TR 12350:2002(E).
    Kent, R., Patrie, J., Benson, N. (2003) The Hybrid III dummy as 
a discriminator of injurious and non-injurious restraint loading, 
Forty0Seventh Annual Proceedings, Association for the Advancement of 
Automotive Medicine.
    Kuppa, S., Eppinger, R. (1998) ``Development of an Improved 
Thoracic Injury Criterion,'' Proceedings of the 42nd Stapp Car Crash 
Conference, SAE No. 983153.
    Kuppa, S. (2001) Lower Extremity Injuries and Associated Injury 
Criteria, Proceedings of the Seventeenth International Technical 
Conference on the Enhanced Safety of Vehicles, Amsterdam, June, 
2001.
    Kuppa, S. (2006) Injury Criteria for Side Impact Dummies, Docket 
No. NHTSA-2007-29134-0001.
    Laituri, T., Prasad, P., Kachnowski, B., Sullivan, K., Przybylo, 
P. (2003) Prediction of AIS 3+ Thoracic Risks for Belted Occupants 
in Full Engagements, Real World Frontal Impacts: Sensitivity to 
Various Theoretical Risk Curves, SAE Paper No. 2003-01-1355, 2003 
SAE World Congress.
    Laituri, T., Prasad, P., Sullivan, K., Frankstein, M., Thomas, 
R. (2005) Derivation and Evaluation of a Provisional, Age Dependent 
AIS 3+ Thoracic Risk Curve for Belted Adults in Frontal Impacts, SAE 
Paper No. 2005-01-0297.
    Laituri, T., Henry, S., Sullivan, K., Prasad, P. (2006) 
Derivation and Theoretical Assessment of a Set of Biomechanics-
based, AIS 2+ Risk Equations for the Knee-Thigh-Hip Complex, Stapp 
Car Crash journal, Vol. 50, November 2006.
    Lissner, H. R. et al. (1960) Experimental Studies on the 
Relation between Acceleration and Intracranial Pressure Changes in 
Man, Surgery Gynecology and Obstetrics, pp.329-338.
    Mertz, H., Weber, D. (1982) Interpretations of the Impact 
Responses of a 3-year-old child dummy relative to child injury 
potential. Ninth Experimental Safety Vehicle Conference, pp. 368-
376, Kyoto, Japan.
    Mertz, H., Irwin, A., Prasad, P. (2003) Biomechanical and 
Scaling Bases for Frontal and Side Impact Injury Assessment 
Reference Values. Stapp Car Crash Journal, Vol. 47, pp. 155-188.
    Mertz H. (1993) Anthropomorphic Test Devices, Accidental Injury, 
Biomechanics, and Prevention, edited by Nahum, A., Melvin, J., 
Springer-Verlag.
    Morgan, et al. (1990) Human Cadaver and Hybrid III Responses to 
Axial Impacts of the Femur, Proceedings of the 1990 International 
IRCOBI Conference on the Biomechanics of Impacts, 1990.
    NHTSA, Final Economic Assessment FMVSS No. 208 Advanced Air Bag 
(2000), NHTSA Docket No. NHTSA-2000-7013-2.
    Nusholtz, G., Domenico, L., Shi, Y., Eagle, P. (2003) Studies of 
Neck Injury Criteria Based on Existing Biomechanical Test Data, 
Accident Analysis and Prevention, Vol. 35, pp. 777-786.
    O' Brien, D., Luchette, F., Pereira, S., Lim, E., Seeskin, C., 
James, L., Miller, S., Davis, K., Hurst, J., Johannigman, J., Frame, 
S. (2002) Pelvic Fracture in the Elderly is Associated with 
Increased Mortality, Surgery, Volume 132, pp. 710-715.
    Prasad, P., Daniel, R., (1984) A biomechanical analysis of head, 
neck, and torso injuries to child surrogates due to sudden torso 
acceleration, SAE Paper No. 841656.
    Rupp, J., Reed, M., Kuppa, S., Wang, S., Goulet, J., Schneider, 
L., ``The Tolerance of the Human Hip to Dynamic Knee Loading,'' 
Stapp Car Crash Journal, Vol. 46, pp. 211-228, 2002.
    Rupp, J., Reed, M., Jeffreys, Y., Schneider, L. (2003). Effects 
of Hip Posture on the Frontal Impact Tolerance of the Human Hip 
Joint. Stapp Car Crash Journal 47:21-33.
    Rupp J., Reed, M., Madura, N., Miller, C., Kuppa, S., Schneider, 
L. (2005). Comparison of the Inertial Response of the Thor-NT, 
Hybrid III, and Unembalmed Cadaver to Simulated Knee-to-Knee-Bolster 
Impacts. Proceedings of the 19th International Technical Conference 
on the Enhanced Safety of Vehicles, Paper 05-0086. National Highway 
Traffic Safety Administration, Washington DC.
    Rupp, J. (2006). Biomechanics of Hip Fractures in Frontal Motor 
Vehicle Crashes. Ph.D. Dissertation. The University of Michigan, Ann 
Arbor, MI.
    Viano, et al. (1977) ``Considerations for a Femur Injury 
Criterion,'' Proceedings of the Twenty-First Stapp Car Crash 
Conference, SAE Paper No. 770925.
    Viano, D. (1989) Biomechanical Responses and Injuries in Blunt 
Lateral Impact. Proc. Thirty-third Stapp Car Crash Conference, pp. 
113-142, Society of Automotive Engineers, Warrendale, PA.
    Viano, D., Lau, I., Asbury, C., King, A., Begeman, P. (1989) 
Biomechanics of the Human Chest, Abdomen, and Pelvis in Lateral 
Impact, Accident Analysis and Prevention, Vol. 21, No. 6, pp. 553-
574.
    Walfisch, G., Fayon, C., Terriere, J., et al., ``Designing of a 
Dummy's Abdomen for Detecting Injuries in Side Impact Collisions,'' 
5th International IRCOBI Conference, 1980.

Appendix D

Relative Risk of Injury, Vehicle Safety Score, and the Star Rating 
System

Introduction

    The risk of injury to each occupant in NHTSA's Crashworthiness 
Rating System is the joint probability of injury to each body region 
considered for that occupant. The overall risk of injury in frontal 
crashes is the average of the injury risk to the driver and 
passenger in the frontal crash test. The risk of injury to the 
driver in side crashes is the weighted average of the risk to the 
driver in the MDB test (weight=0.8) and the pole test (weight=0.2). 
The overall risk of injury in side crashes is the average of the 
injury risk to the driver in side crashes (MDB and Pole) and the 
injury risk to the rear seat passenger in the MDB test.
    The crashworthiness rating system provides relative risk of 
injury for each occupant in each crash test condition (driver and 
front outboard passenger in the frontal crash test, driver and near 
side rear seat passenger in the side MDB test, driver in the oblique 
pole impact test, and rollover test) and a Combined Crashworthiness 
Rating Vehicle Safety Score. The relative risk of injury in each 
test condition for a vehicle is computed by dividing the overall 
risk of injury in each crash mode by an average baseline risk (for 
example, the average risk of serious injury in the fleet or that of 
a group of select vehicles in the fleet for a certain model year). 
The Combined Crashworthiness Rating Vehicle Safety Score (VSS) is 
obtained as a weighted average of the individual Relative Risk Score 
(RRS) in each test condition.
    The RRS for each test condition and the VSS represent the risk 
of injury to occupants of the vehicle relative to a baseline risk of 
injury. For example, a VSS of 1.15 for a vehicle implies that the 
occupants in that vehicle are 15 percent more likely to sustain 
serious injury than a vehicle representing the baseline risk.

Frontal Crash Test Rating

    The historical frontal NCAP crash test data for the driver from 
the model years 1995 through 2008 were examined using the injury 
risk curves presented in Appendix C.
    The average risk of injury to the head, neck, chest, and femur 
of the driver, computed using the injury risk curves from Appendix 
C, for each vehicle of model years 2004 to 2008 is presented in 
Figure 1.

[[Page 40046]]

[GRAPHIC] [TIFF OMITTED] TN11JY08.005

    When compared to data from 1995, these data indicate that the 
average risk of injury to the driver by model year has been reduced 
since 1995 and is less than 0.2 after MY 2002 (Table 2). If the 
average performance of all the vehicles tested in NCAP each year is 
used to represent the fleet of new cars, then for MY 2008, the 
average risk of serious injury in the fleet is approximately 0.15. 
Therefore, the baseline injury risk of 0.15 was used to compute the 
relative risk of injury in frontal crashes for each vehicle (Table 
3).

                            Table 1.--Probability of Injury Statistics for Drivers in NCAP Frontal Crash Tests by Model Year
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                 Average     Prob Std.                   P 25%                     P 75%
                              MY                                   prob      deviation    Minimum P     quartile     P Median     quartile    Maximum P
--------------------------------------------------------------------------------------------------------------------------------------------------------
1995.........................................................         0.30         0.12         0.10         0.21         0.27         0.35         0.62
1996.........................................................         0.32         0.18         0.13         0.18         0.28         0.40         0.86
1997.........................................................         0.26         0.14         0.12         0.17         0.22         0.28         0.69
1998.........................................................         0.26         0.11         0.11         0.20         0.24         0.30         0.63
1999.........................................................         0.29         0.18         0.09         0.17         0.23         0.36         0.71
2000.........................................................         0.25         0.15         0.11         0.15         0.22         0.28         0.64
2001.........................................................         0.23         0.12         0.09         0.17         0.19         0.26         0.63
2002.........................................................         0.20         0.09         0.09         0.14         0.17         0.22         0.61
2003.........................................................         0.18         0.09         0.08         0.12         0.15         0.18         0.45
2004.........................................................         0.15         0.07         0.08         0.11         0.14         0.18         0.46
2005.........................................................         0.17         0.11         0.09         0.11         0.14         0.19         0.57
2006.........................................................         0.17         0.06         0.08         0.13         0.15         0.22         0.31
2007.........................................................         0.15         0.05         0.09         0.12         0.14         0.17         0.38
2008.........................................................         0.15         0.04         0.09         0.12         0.14         0.18         0.24
                                  Average MY 1995-2008                                          0.10         0.15         0.19         0.25         0.56
                                  Average MY 2004-2008                                          0.08         0.12         0.14         0.19         0.39
--------------------------------------------------------------------------------------------------------------------------------------------------------


[[Page 40047]]

[GRAPHIC] [TIFF OMITTED] TN11JY08.022

    The average, minimum, maximum, and the quartiles presented in 
Table 3 provide an estimate of the dispersion of Relative Risk Score 
(RRS) in different model years. Since most of the current vehicles 
receive four or five stars in the NCAP frontal crash tests, NHTSA 
prescribed the baseline risk of 15 percent (representing the average 
risk of injury to the driver in MY 2007 and MY 2008 vehicles in the 
NCAP frontal crash test) to be at the border of the 4 star and 3 
star rating. Other criteria used to determine the star bands were 
(1) vehicles performing exceptionally well (at 0-15 percentile of 
vehicles tested) are assigned a five star rating, and (2) vehicles 
performing very poorly (greater than 4 standard deviations from 
mean) would be assigned a one star. Attempts were also made to 
maintain equidistant star band boundaries. Based on these criteria 
and the distribution of relative risk of injury scores presented in 
Table 3, the relationship between RRS and the number of stars was 
established as presented in Table 4. The RRS is computed by (1) 
rounding the injury risk to the nearest tenth of a percent in 
accordance with the rounding-off method of ASTM Standard Practice E 
29 for Using Significant Digits in Test Data to Determine 
Conformance with Specifications, (2) dividing the injury risk by 
0.15 (15.0 percent baseline injury risk), (3) and finally rounding 
the result to the nearest one hundredth in accordance to ASTM 
Standard E 29. It should be noted that a vehicle which passes 
compliance (with a 20 percent compliance margin) would have an 
injury risk of 52.1 percent corresponding to a RRS value of 3.47.
BILLING CODE 4910-59-P

[[Page 40048]]

[GRAPHIC] [TIFF OMITTED] TN11JY08.023

BILLING CODE 4910-59-C

[[Page 40049]]

Side Crash Test Rating

    Because the agency did not have test data using the ES 2re or 
SID IIs dummies at the NCAP test speed for the MDB test, the agency 
computed the average risk of serious injury derived from relevant 
MDB tests and oblique pole impact tests done in support of the FMVSS 
214 side impact protection upgrade. The MDB test is conducted with 
the ES-2re dummy in the front driver seat and the SID-IIs in the 
rear passenger seat. The pole impact test is conduced with the SID-
IIs in the driver's seat.
    The injury risk curves for side impact reported in Appendix C 
are applied to side MDB tests and oblique pole tests. These tests 
were part of NHTSA's fleet evaluation for the FMVSS 214 side impact 
upgrade and details and thorough analysis of these tests are 
available in the NHTSA docket number NHTSA-2007-25441.
    There were six vehicles which were tested in the FMVSS 214 test 
conditions (MDB impact at 53 km/h rather than the NCAP 62 km/h) as 
well as the oblique pole impact with the SID-IIs dummies. The dummy 
injury measures in the paired crash tests of these vehicles with the 
ES-2re and SID-IIs dummies were used to determine risk of injury in 
side crashes and a Relative Risk Score (RRS) for side crashes. Table 
4 presents the statistics for the risk of injury (average, standard 
deviation, minimum, maximum, median, and 25 and 75 percentile injury 
risk values) for each dummy in the MDB and oblique pole tests using 
the injury risk curves from Appendix C.
    The overall risk of injury to the driver for each vehicle is the 
weighted average of the driver injury risk in the MDB test 
(multiplied by 0.8) and that in the oblique pole test (multiplied by 
0.2). The risk of injuries in side crashes for a vehicle is the 
simple average of the injury risk of the rear seat passenger in the 
MDB test and the overall driver injury risk. Table 4 also presents 
the statistics for the overall risk of injury to the driver and the 
risk of injury in side crashes.

                 Table 4.--Probability of Injury (P) Statistics for Different Occupants in the Side MDB and the Oblique Pole Crash Tests
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                          25%                       75%
                          Crash type                            Average P   Std. Dev. P     Min P      quartile P    Median P    quartile P     Max P
--------------------------------------------------------------------------------------------------------------------------------------------------------
MDB Driver...................................................         0.09         0.04         0.04         0.06         0.09         0.12         0.13
MDB Pass.....................................................         0.13         0.21         0.03         0.03         0.04         0.07         0.55
Pole Driver..................................................         0.64         0.39         0.13         0.32         0.79         0.93         0.98
Overall Driver...............................................         0.20         0.11         0.06         0.12         0.23         0.28         0.30
Side Impact..................................................         0.16         0.16         0.05         0.07         0.14         0.18         0.43
--------------------------------------------------------------------------------------------------------------------------------------------------------
 The overall risk of injury to the driver is computed as the weighted average of the risk of driver injury in the MDB test (multiplied by 0.8)
  and the risk of driver injury in the pole test (multiplied by 0.2).
 The risk of injury in side impact is the average of the overall driver risk and the risk of rear passenger in the MDB test.

    The average risk of injury from the six MDB tests for the driver 
and the rear passenger is 0.09 and 0.13, respectively. The average 
risk of injury to the driver in the six oblique pole tests is 0.64 
and the average overall risk of injury to the driver (combining the 
MDB and pole test results) is 0.20. For these six vehicles, the 
average risk of injury in side crashes is 0.16.
    In order to promote improvement in side impact safety in all the 
vehicles, the baseline risk of injury to compute Relative Risk 
Scores (RRS) in side crashes is taken to be 15 percent. As in 
frontal crash tests, the RRS in side MDB and pole crash tests is 
computed by 1) rounding the injury risk to the nearest tenth of a 
percent in accordance with the rounding-off method of ASTM Standard 
Practice E 29 for Using Significant Digits in Test Data to Determine 
Conformance with Specifications, 2) dividing the injury risk by 0.15 
(15.0 percent baseline injury risk), 3) and finally rounding the 
result to the nearest one hundredth in accordance to ASTM Standard E 
29. Table 5 presents the RRS statistics corresponding to the injury 
risk presented in Table 4 using a baseline injury risk of 15 
percent.

                 Table 5.--Relative Risk Score (RRS) Statistics for Different Occupants in the Side MDB and the Oblique Pole Crash Tests
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                          25%                       75%
                          Crash type                           Average RRS   Std. Dev.     Min RRS      quartile    Median RRS    quartile     Max RRS
                                                                                RRS                       RRS                       RRS
--------------------------------------------------------------------------------------------------------------------------------------------------------
MDB Driver...................................................         0.60         0.25         0.28         0.42         0.59         0.80         0.87
MDB Pass.....................................................         0.86         1.39         0.20         0.21         0.28         0.45         3.69
Pole Driver..................................................         4.27         2.57         0.89         2.15         5.24         6.23         6.54
Overall Driver...............................................         1.33         0.71         0.40         0.77         1.52         1.89         2.00
Side Impact..................................................         1.09         1.05         0.30         0.49         0.90         1.17         2.84
--------------------------------------------------------------------------------------------------------------------------------------------------------
 The Relative Risk Score for MDB tests, pole tests, and side impacts is obtained by dividing the risk of injury in each side crash mode listed
  in Table 4 by 0.15 which represents the baseline risk of injury in side impacts.

    Vehicles for which all the dummy injury measures (for the ES-2re 
and SID-IIs) in the MDB and pole tests just meet the compliance 
limits, the risk of injury is 0.70 for the ES-2re and 0.42 for the 
SID IIs dummies resulting in an overall risk of injury in side 
crashes of 0.532, a RRS of 3.54.

Rollover Rating

    Since the proposed rollover rating is the same as that currently 
used in NCAP, the current relationship between the risk of rollover 
and star rating used in NCAP is applied here and is shown in Table 
11. If 15 percent risk (corresponding to a 4 star rating) is used as 
the baseline risk (as that in front and side crash test rating), 
then the relationship between the vehicle safety score in rollover 
is as shown in Table 11.

[[Page 40050]]



                Table 11.--Star Rating, Risk of Rollover, and the Relative Risk score in Rollover
                                      [Using a baseline risk of 15 percent]
----------------------------------------------------------------------------------------------------------------
             Number of stars                    Risk of rollover            Relative risk score in rollover
----------------------------------------------------------------------------------------------------------------
1 star..................................  P >= 40 percent............  RRS >= 2.67
2 stars.................................  30 <= P < 40 percent.......  2.0 <= RRS < 2.67
3 stars.................................  20 <= P < 30 percent.......  1.33 <= RRS < 2.0
4 stars.................................  10 <= P < 20 percent.......  0.67 <= RRS < 1.33
5 stars.................................  P < 10 percent.............  RRS < 0.67
----------------------------------------------------------------------------------------------------------------

Combined Crashworthiness Rating Vehicle Safety Score

    The weighted average of the Relative Risk Scores (RRS) in front, 
side, and rollover crashes is the combined crashworthiness rating 
Vehicle Safety Score (VSS). The weight applied to each crash mode 
represents the proportion of injury associated with that crash mode. 
Since the baseline injury risk used to compute RRS in each crash 
mode is 15 percent, the combined crashworthiness rating also 
represents the relative risk of injury with respect to a baseline of 
15 percent. The Vehicle Safety Score for the Combined 
Crashworthiness Rating is computed below:

Combined Rating = (5/12) x RRS(front) + (4/12) x RRS(side) + (3/12) 
x RRS(roll)

    The final VSS value is obtained by rounding the result from the 
above equation to the nearest one hundredth in accordance to ASTM 
Standard E 29. The star bands used for rating frontal and side 
impacts are applied to the combined crashworthiness rating using VSS 
and is presented in Table 12.

                                        Table 12.--Relationship Between Vehicle Safety Score and the Star Rating
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                       5 stars                  4 stars                  3 stars                  2 stars                 1 star
--------------------------------------------------------------------------------------------------------------------------------------------------------
VSS Values...................  VSS <0.67..............  0.67 <= VSS <1.00......  1.00 <= VSS < 1.33.....  1.33 <= VSS < 2.67....  VSS >= 2.67
Probability..................  P < 0.100..............  0.100 <= P < 0.150.....  0.150 <= P < 0.200.....  0.200 <= P < 0.400....  P >= 0.400
--------------------------------------------------------------------------------------------------------------------------------------------------------


    Authority: 49 U.S.C. Sec. Sec.  32302, 30111, 30115, 30117, 
30166, and 30168, and Pub. L. 106-414, 114 Stat. 1800; delegation of 
authority at 49 CFR 1.50.

    Issued on: July 3, 2008.
Nicole R. Nason,
Administrator.

 [FR Doc. E8-15620 Filed 7-10-08; 8:45 am]
BILLING CODE 4910-59-P