[Federal Register: December 14, 2006 (Volume 71, Number 240)] [Rules and Regulations] [Page 75303-75342] From the Federal Register Online via GPO Access [wais.access.gpo.gov] [DOCID:fr14de06-12] [[Page 75303]] ----------------------------------------------------------------------- Part II Department of Transportation ----------------------------------------------------------------------- National Highway Traffic Safety Administration ----------------------------------------------------------------------- 49 CFR Part 572 Anthropomorphic Test Devices; ES-2re Side Impact Crash Test Dummy 50th Percentile Adult Male and SID-IIs Side Impact Crash Test Dummy 5th Percentile Adult Female; Final Rules [[Page 75304]] ----------------------------------------------------------------------- DEPARTMENT OF TRANSPORTATION National Highway Traffic Safety Administration 49 CFR Part 572 Docket No. NHTSA-2004-25441 RIN 2127-AI89 Anthropomorphic Test Devices; ES-2re Side Impact Crash Test Dummy 50th Percentile Adult Male AGENCY: National Highway Traffic Safety Administration (NHTSA), Department of Transportation (DOT). ACTION: Final rule. ----------------------------------------------------------------------- SUMMARY: This final rule amends the agency's regulation on anthropomorphic test devices to add specifications and qualification requirements for a new mid-size adult male crash test dummy, called the ``ES-2re'' test dummy. The ES-2re dummy has enhanced injury assessment capabilities compared to devices existing today, which allows for a fuller assessment of the types and magnitudes of the injuries occurring in side impacts and of the efficacy of countermeasures in improving occupant protection. The agency plans to use the ES-2re dummy in an upgraded Federal Motor Vehicle Safety Standard on side impact protection. DATES: This final rule is effective June 12, 2007. The incorporation by reference of certain publications listed in the regulations is approved by the Director of the Federal Register as of June 12, 2007. If you wish to petition for reconsideration of this rule, your petition must be received by January 29, 2007. ADDRESSES: If you wish to petition for reconsideration of this rule, you should refer in your petition to the docket number of this document and submit your petition to: Administrator, Room 5220, National Highway Traffic Safety Administration, 400 Seventh Street, SW., Washington, DC 20590. The petition will be placed in the docket. Anyone is able to search the electronic form of all documents received into any of our dockets by the name of the individual submitting the comment (or signing the comment, if submitted on behalf of an association, business, labor union, etc.). You may review DOT's complete Privacy Act Statement in the Federal Register published on April 11, 2000 (Volume 65, Number 70; Pages 19477-78) or you may visit http://dms.dot.gov. FOR FURTHER INFORMATION CONTACT: For non-legal issues, you may call Stan Backaitis, NHTSA Office of Crashworthiness Standards (telephone 202-366-4912). For legal issues, you may call Deirdre Fujita, NHTSA Office of Chief Counsel (telephone 202-366-2992) (fax 202-366-3820). You may send mail to these officials at the National Highway Traffic Safety Administration, 400 Seventh St., SW., Washington, DC 20590. SUPPLEMENTARY INFORMATION: Table of Contents I. The ES-2re Dummy Generally Described a. Development of the Rib Extensions b. The Reference Materials for the Dummy II. Notice of Proposed Rulemaking (NPRM) III. Overview of Comments IV. Response to the Comments a. Biofidelity 1. ISO Technical Report 9790 Methodology 2. NHTSA Biofidelity Ranking System b. Other Issues Relating to How Humanlike the Dummy Is 1. Anthropometry of Abdominal and Pelvic Regions 2. Sitting Height 3. ES-2re's Representation of Large Male Population 4. Abdominal Instrumentation 5. Shoulder Design 6. Rib Deflections 7. Rib Extensions c. Repeatability and Reproducibility 1. Sample Size 2. Reproducibility of Pelvic Load Measurements 3. Sensitivity to Initial Conditions 4. Rib Acceleration Response d. Directional Impact Sensitivity 1. Impact Direction 2. Rib Binding in ISO 9790 Tests 3. ISO 9790 Ratings for Lateral and Oblique Impacts e. Durability f. Symmetry g. Using the ES-2 Test Dummy h. Test Dummy Drawing Package 1. 3-D Shape Definitions 2. Material Specifications 3. Dummy Drawing Changes i. Certification Procedures and Response Corridors 1. Overview of the Comments 2. Head Drop Test 3. Neck Flexion Test i. Neck Response Corridors ii. Neck Pendulum Aluminum Honeycomb iii. Neck Pendulum Deceleration Filter Class iv. Nodding Block Configuration v. Adjusting Half-Spherical Neck Screws 4. Thorax i. Full-Body Systems Test ii. Specifying Impact Speed in Rib Module Drop Test iii. Recovery Time Between Successive Tests 5. Lumbar Spine i. Response Corridors ii. Lumbar Cable Nut Adjustment 6. Shoulder i. Shoulder Cord Tension ii. Pendulum Configuration 7. Abdomen 8. Pelvis 9. Other Issues i. Test Probe Suspension Cables and Attachments ii. Pelvis and Abdomen Pendulum Filter Requirements iii. Temperature V. NHTSA Crash Test Experience a. MDB Tests b. Oblique Pole Tests c. Rib Responses d. Torso Back Plate Responses e. Durability VI. Conclusions Rulemaking Analyses and Notices Appendix A to Final Rule Preamble: Specific Drawing Comments and Agency Responses to Those Comments NHTSA published a notice of proposed rulemaking (NPRM) that proposed to upgrade Federal Motor Vehicle Safety Standard (FMVSS) No. 214, ``Side Impact Protection'' (49 CFR 571.214) by, among other things, adopting a dynamic pole test into the standard (May 17, 2004; 69 FR 27990; Docket 17694; reopening of comment period, January 12, 2005, 70 FR 2105). The proposed pole test is similar to, but more demanding than, the one currently used optionally in FMVSS No. 201, ``Occupant Protection in Interior Impact'' (49 CFR 571.201). In the proposed pole test, a vehicle is propelled sideways into a rigid pole at an angle of 75 degrees, at any speed up to 32 km/h (20 mph). The NPRM proposed that compliance with the pole test would be determined in two test configurations, one using a test dummy representing mid-size adult males and the other using a test dummy representing small adult females. The NPRM proposed to require vehicles to protect against head, thoracic and other injuries as measured by the two test dummies. The agency also proposed using the dummies in FMVSS No. 214's existing moving deformable barrier (MDB) test, which simulates a vehicle-to- vehicle ``T-bone'' type intersection crash.\1\ --------------------------------------------------------------------------- \1\ On August 10, 2005, the President signed the ``Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users,'' (SAFETEA-LU), Pub. L. 109-59 (Aug. 10, 2005; 119 Stat. 1144), to authorize funds for Federal-aid highways, highway safety programs, and transit programs, and for other purposes. Section 10302(a) of SAFETEA-LU provides: Sec. 10302. Side-Impact Crash Protection Rulemaking. (a) Rulemaking.--The Secretary shall complete a rulemaking proceeding under chapter 301 of title 49, United States Code, to establish a standard designed to enhance passenger motor vehicle occupant protection, in all seating positions, in side impact crashes. The Secretary shall issue a final rule by July 1, 2008. At the time of the enactment of Sec. 10302(a), the agency's notice of proposed rulemaking to upgrade FMVSS No. 214 was pending. The final rule completing the rulemaking proceeding will be issued in the near future. --------------------------------------------------------------------------- [[Page 75305]] This document establishes the specifications and qualification requirements for the new mid-size adult male crash test dummy, called the ``ES-2re'' test dummy, for use in FMVSS No. 214. The NPRM preceding this Part 572 final rule on the ES-2re dummy was published on September 15, 2004 (69 FR 55550; Docket 18864; reopening of comment period, January 12, 2005, 70 FR 2105).\2\ --------------------------------------------------------------------------- \2\ NHTSA published an NPRM proposing to amend 49 CFR Part 572 to add the specifications for the small female dummy to Part 572 on December 8, 2004 (69 FR 70947; Docket 18865; extension of comment period, March 8, 2005; 70 FR 11189). --------------------------------------------------------------------------- I. The ES-2re Dummy Generally Described The ES-2re can be instrumented with a wide array of sensors to better predict a wider range of injury potential than any other currently available mid-size male side impact test dummy. The ES-2re is technically superior to both the SID-H3 50th percentile male test dummy (49 CFR Part 572, subpart M) currently used in the optional pole test of FMVSS No. 201 and the SID 50th percentile adult male test dummy (49 CFR Part 572, subpart F) now used in the MDB test of FMVSS No. 214. It can assess the potential for head, neck, thoracic, abdominal, pelvic, and other injuries. It can assess the potential for head injury (measuring the resultant head acceleration, which is used to calculate the Head Injury Criterion (HIC)); thoracic injuries in terms of spine and rib accelerations and rib deflections (chest deflection has been shown to be the best predictor of thoracic injuries in low-speed side impacts); abdominal injuries through three load cells to assess the magnitude of lateral and oblique forces; and pelvic injuries.\3\ --------------------------------------------------------------------------- \3\ The ES-2re can also assess load transfer between the upper and the lower torso, torso interaction with the vehicle seat back, neck injuries via upper and lower neck load cells; and the impact severity of the vehicle structure on the legs by way of a femur load cell. In addition, a clavicle load cell is available to assess shoulder loading. --------------------------------------------------------------------------- Its improved biofidelity and enhanced injury assessment capability allows for a fuller assessment of the types and magnitudes of the injuries occurring in side impacts and a more penetrating evaluation of the efficacy of vehicle countermeasures installed to improve side impact protection than now possible using other existing side impact dummies. In the May 17, 2004 NPRM concerning FMVSS No. 214, NHTSA proposed injury criteria for the ES-2re's injury measuring instrumentation of the dummy's head, thorax, abdomen and pelvis. HIC would be limited to 1000 measured in a 36 millisecond time interval (HIC36). Chest deflection would be limited to not greater than 42 millimeters (mm) (1.65 inch (in)) for any rib. Abdominal loads would be limited to 2,500 Newtons (N) (562 pounds). For pelvic injury, pubic symphysis force would be limited to 6,000 N (1,349 pounds). (See, ``Injury Criteria for Side Impact Dummies,'' Docket 17694.) The ES-2re consists of a metallic ``skeleton'' which is covered by ``soft tissue'' consisting of rubber, plastic and foam. The dummy does not have lower arms because researchers concluded that lower arms on the side crash test dummy could interfere with the interaction of the side structure of a vehicle and the dummy's measurement of potential harm to the thoracic and pelvic regions. The ES-2re has a mass of 72 kilograms (kg) (158.8 pounds), which is the mass of a 50th percentile adult male without lower arms.\4\ --------------------------------------------------------------------------- \4\ A 50th percentile adult male with lower arms has a mass of approximately 78 kg (172 pounds). If the ES-2re had arms, its mass would be equivalent. --------------------------------------------------------------------------- The 90.0 cm seated height of the ES-2re is representative of adult males mid-size and taller. The dummy will provide valuable data on the interaction of these occupants with the vehicle's interior in FMVSS No. 214's side impact tests. a. Development of the Rib Extensions The ES-2re is a modified version of a European ES-2 side impact dummy, which was originally developed in Europe as the EuroSID-1 dummy in the late 1980s and early 1990s. The EuroSID-1 dummy is used in European Directive 96/27/EC. The EuroSID-1 dummy was redesigned and reevaluated during the late 1990s and early 2000 to address some problems with dummy performance, and was renamed the ES-2. The ES-2re dummy is the result of a modification of the ES-2. Although the ES-2 has a better design than the EuroSID-1, the ES-2 has a back plate that causes a part of it to ``grab'' parts of a vehicle seat back in a crash test, which alters some of the dummy response measurements. To address the problem, which has also been observed in the EuroSID-1, the ES-2 dummy manufacturer redesigned the rib module by adding rib extensions to the dummy. The extended ribs provide a continuous loading surface that nearly encircles the thorax of the dummy and encloses the posterior gap of the ES-2 ribcage that was thought to be responsible for the seat back grabbing effect. The modified dummy is referred to as the ES-2re, with the ``-re'' suffix indicating the use of the rib extensions on the dummy. The agency's evaluation of the ES-2re dummy indicates that the rib extensions successfully addressed the back plate grabbing problem in the environments in which grabbing had occurred with the ES-2 dummy. b. The Reference Materials for the Dummy A technical report and other materials describing the ES-2re in detail have been placed in the following NHTSA dockets: the docket for the September 15, 2004 NPRM on the ES-2re (Docket 18864); the docket for the May 17, 2004 NPRM proposing the pole test upgrade to FMVSS No. 214 (Docket 17694); and the docket for today's final rule (Docket 25441). When we refer in this preamble to a docket item, we will identify by docket number where the item is filed. The specifications for the ES-2re consist of: (a) A drawing package containing all of the technical details of the dummy; (b) a parts list; and (c) a user manual containing instructions for inspection, assembly, disassembly, use, and adjustments of dummy components. These drawings and specifications ensure that ES-2re dummies will be the same in their design and construction. The drawings, parts list and user manual are available for examination in the NHTSA docket section for this final rule (Docket 25441). Copies of those materials may also be obtained from Leet-Melbrook, Division of New RT, 18810 Woodfield Road, Gaithersburg, Maryland 20879, telephone (301) 670-0090. II. Notice of Proposed Rulemaking (NPRM) The NPRM preceding this Part 572 final rule on the ES-2re dummy was published on September 15, 2004 (69 FR 55550; Docket 18864). On January 12, 2005, in response to a petition from the Alliance of Automobile Manufacturers, NHTSA reopened the comment period for the NPRM until April 12, 2005 (70 FR 2105). The September 15, 2004 NPRM discussed NHTSA's tentative findings that the ES-2re was commercially available, was sufficiently biofidelic, had good repeatability and reproducibility of its impact responses, performed well in vehicle crash tests, and had good durability in evaluation programs. NHTSA believed that the ES-2re could be used for both left- and right-side impacts. The agency also discussed in the NPRM that the dummy's responses did not show sensitivity to oblique impacts in full- [[Page 75306]] scale crash tests. The agency also discussed in the NPRM proposed calibration test specifications and procedures. III. Overview of Comments The agency received comments from 5 different organizations: Autoliv, Denton ATD (DATD), First Technology Safety Systems (FTSS), Ferrari, and the Alliance of Automobile Manufacturers (Alliance). These comments, summarized below, are discussed in detail in the next section of this preamble. Autoliv generally supported the agency's proposal. DATD and FTSS were supportive, but suggested changes to the drawing package, certification corridors, and other technical matters of the NPRM. Ferrari stated that it observed ``anomalous'' peaks in the rib acceleration curves occurring between 67 and 73 ms after barrier impact with the vehicle, which Ferrari believed were caused by insufficient rebound damping in the rib modules. The Alliance did not support the agency's proposal. The Alliance was concerned about matters including: the biofidelity of the dummy (the commenter believed that there are shortcomings in the ES-2re's shoulder, abdominal and pelvic regions, particularly when compared to the performance of the ES-2 and the WorldSID \5\ in full-vehicle tests); the repeatability and reproducibility of the ES-2re; the directional impact sensitivity of the dummy; and miscellaneous issues, such as the symmetry of abdomen response when impacted on the right and left sides and the durability of the ES-2re. The Alliance also had comments regarding the proposed certification procedures and corridors. The Alliance submitted a petition for rulemaking (Docket 17252) asking NHTSA to initiate rulemaking to incorporate WorldSID into 49 CFR Part 572 and to use WorldSID in the upgrade of FMVSS No. 214 rather than the ES-2re.\6\ The Alliance further suggested that, prior to the incorporation of WorldSID into 49 CFR Part 572, the ES-2 dummy should be used rather than the ES-2re, and only to the extent of using the dummy to measure responses relating to the head injury criterion (HIC). --------------------------------------------------------------------------- \5\ WorldSID is the next-generation 50th percentile male side impact dummy developed by industry representatives from the U.S., Europe and Japan, with the support of the European and Japanese governments (see Docket No. 2000-17252). This future dummy is believed by its developers to have better biofidelity than existing dummies, and is intended to better predict a wider range of injury potential in side impact testing than current dummies. \6\ The agency's response to the petition will be issued in rulemaking documents relating to the FMVSS No. 214 rulemaking. --------------------------------------------------------------------------- IV. Response to the Comments a. Biofidelity Biofidelity is a measure of how well a test device duplicates the responses of a human in an impact. As discussed in the NPRM, two methods are currently available for assessing the biofidelity of a dummy in side impact testing. These are: (a) An International Organization of Standardization (ISO) procedure, referred to as ISO Technical Report (TR) 9790, which determines the biofidelity of a dummy by how well the dummy's body segment and/or subsystem impact responses replicate cadaver responses in defined impact environments; and (b) a NHTSA Biofidelity Ranking System.\7\ The latter method determines the dummy's biofidelity based on two assessment measures: the ability of a dummy to load a vehicle or some other type of an impact surface as a cadaver does, termed ``External Biofidelity''; and the ability of a dummy to replicate those cadaver responses that best predict injury potential, termed ``Internal Biofidelity.'' The NPRM explained that the ES-2re's biofidelity was evaluated under both of these methodologies. --------------------------------------------------------------------------- \7\ The NHTSA Biofidelity Ranking System method was reported by Rhule H., et al., in a technical paper in the 2002 Stapp Car Crash Journal, Vol. 46, p. 477, ``Development of a New Biofidelity Ranking System for Anthropomorphic Test Devices.'' --------------------------------------------------------------------------- 1. ISO Technical Report 9790 Methodology The ISO rating system is based on a scale of 0 to 10, with 0 signifying total lack of biofidelity and 10 signifying that the body segment has a biofidelic response much like that of a human subject. Once the ratings are established for each body segment, the overall dummy's biofidelity is calculated and its ranking determined using the following classification scale: 0 to 2.6 (Unacceptable); 2.6 to 4.4 (Marginal); 4.4 to 6.5 (Fair); 6.5 to 8.6 (Good); 8.6 to 10 (Excellent). The agency had tentatively assessed in the NPRM that the ISO-based biofidelity assessment of 4.6 would generally be the same for the ES- 2re as the ES-2. The Occupant Safety Research Partnership (OSRP) and Transport Canada conducted biomechanical testing on the ES-2 dummy using the ISO-specified methodology and test procedures. The results of these tests were reported by Byrnes et al. in the 2002 Stapp Car Crash Journal, Vol. 46, in Paper No. 2002-22-0014. Because the ES-2re dummy's backplate modifications were developed with the express objective not to alter in any way the ES-2 dummy's impact response, and because the ES-2re conformed to the same calibration levels as the ES-2, the agency believed that the rib extension modifications to the ES-2 would not affect the ISO based biofidelity assessment. (Moreover, as reported in the NPRM, the findings of the NHTSA Biofidelity Ranking System tests appeared to confirm this assessment, as it was established that under that ranking system both the ES-2 and the ES-2re dummies had nearly identical biofidelity levels.) In the NPRM, the agency stated that a biofidelity rating of ``fair,'' at 4.6, would be an improvement over the SID and EuroSID-1, which received ratings of 2.3 and 4.4, respectively (Byrnes, et al., ``ES-2 Dummy Biomechanical Responses,'' 2002, Stapp Car Crash Journal, Vol. 46, 2002-22-0014, p. 353). The agency believed that the ES-2 (ES-2re) ISO biofidelity rating also compared favorably to that of the SID/HIII, which received an overall rating of 3.8.\8\ --------------------------------------------------------------------------- \8\ The biofidelity rating for the SID dummy used in FMVSS No. 214 is 2.3. The rating for the SID/HIII of 3.8, using the ISO method, reflects use of the special purpose side impact HIII head and neck as noted in 63 FR 41468, August 4, 1998. --------------------------------------------------------------------------- Comment: In its comment, the Alliance disagreed with NHTSA's statement that the rib modifications made to the ES-2 and resulting in the ES-2re configuration had no effect on the dummy's ISO-based biofidelity assessment. The Alliance stated that testing conducted by the OSRP resulted in an overall ISO score of 4.3 for the ES-2re, as compared to a 4.6 score for the ES-2. Agency response: The Alliance neither provided a reference to a published report nor provided supporting data related to the assertion that the overall ISO score for the ES-2re is 4.3. The absence of substantiation of the comment limits our ability to respond. Even so, assuming the accuracy of the comment that the rib extensions reduced the ISO-based biofidelity assessment of the ES-2 from 4.6 to 4.3, or from ``fair'' to ``marginal,'' we nonetheless conclude that a 4.3 rating of the ES-2re is acceptable. NHTSA believes that the side impact dummy used in FMVSS No. 214 should measure the risk of thoracic and abdominal injuries, since these injuries are the most prevalent injuries in side crashes. The ES-2 (which does not have the rib extensions) is not suitable for use in our compliance testing, because of its back plate design and the problem that can occur with the back plate loading some seat backs and influencing the [[Page 75307]] dummy's rib deflection measurements. The rib extensions of the ES-2re allow for more accuracy in the measurement of rib deflections. Although the dummy with the extensions has a slightly lower, yet acceptable, ISO biofidelity ranking than a dummy without the rib extensions, the ES-2re is preferable over the ES-2 because it allows the agency to measure fully the risk of thoracic and abdominal injury in side crashes. We note also that a 4.3 ISO rating is an improvement over the biofidelity rating of SID, which received a rating of 2.3 (Byrnes, et al., ``ES-2 Dummy Biomechanical Responses,'' 2002, Stapp Car Crash Journal, Vol. 46, 2002-22-0014, p. 353). The ES-2re biofidelity rating also compares favorably to that of the SID/HIII, which received an overall rating of 3.8. Both the SID and SID/HIII have performed well in facilitating the installation of life-saving countermeasures that have substantially improved the safety of occupants in side crashes. 2. NHTSA Biofidelity Ranking System Further, under the NHTSA biofidelity ranking system, the biofidelity rankings for the ES-2 and ES-2re are nearly identical. The biofidelity ranking system developed by Rhule, H., et al., supra, includes an assessment of the dummy's External Biofidelity and Internal Biofidelity. The Overall External and Internal Biofidelity ranks are an average of each of the external and internal body region ranks, respectively. A lower biofidelity rank indicates a more biofidelic dummy by this NHTSA ranking method. A dummy with an External and/or Internal Biofidelity rank of less than 2.0 is considered to respond much like a human subject. The NHTSA ranking system is based on a variety of cadaver and dummy exposures, such as head drop tests, thorax and shoulder drop tests, thorax and shoulder pendulum tests, and whole body sled tests. The NHTSA ranking system also includes abdominal and pelvic offset sled test conditions. Each test condition has a response corridor derived from human cadavers and assigned a weight factor based upon the robustness of the particular test and its similarity to full scale crash conditions. For each response requirement, the cumulative variance of the dummy response relative to the mean cadaver response (DCV) and the cumulative variance of the mean cadaver response relative to the mean plus one standard deviation (CCV) are calculated. The ratio of DCV/CCV expresses how well the dummy response duplicates the mean cadaver response: A smaller ratio indicating better biofidelity. Although this method does not establish an ``absolute'' ranking scale, the ranks provide a relative sense of the ``number of standard deviations away'' the dummy's responses are from the mean human cadaver response. Rhule conducted an analysis and found that if the dummy's biofidelity ranking is below two, then the dummy is behaving similar to the human cadaver. The evaluation methodology provides a comparison of both dummy response to cadaver response as well as a comparison of two or more dummies. Rhule et al., supra, determined external and internal biofidelity rankings for the ES-2 dummy. NHTSA later repeated the tests for the ES- 2re to determine that dummy's biofidelity rankings. Tables 1 and 2, below, provide a summary of External Biofidelity and Internal Biofidelity rankings, respectively, for the ES-2 and the ES-2re. The results of NHTSA's Biofidelity Ranking System tests indicate that the ES-2 and ES-2re dummies have essentially the same external and internal biofidelity assessment values, and that the rib extensions have had no effect on the biofidelity of the ES-2. The overall external biofidelity scores were 2.7 and 2.6 for the ES-2 and ES-2re, respectively, while the overall internal biofidelity scores for both were 1.6. The testing conducted for the ranking indicates that there exists no significant difference in the response characteristics of the ES-2 and ES-2re dummies. Table 1.--External Biofidelity Rankings of the ES-2 and ES-2re ------------------------------------------------------------------------ External biofidelity rank ES-2 ES-2re ------------------------------------------------------------------------ Overall........................................... 2.7 2.6 Head/Neck......................................... 3.7 3.7 Shoulder.......................................... 1.4 1.4 Thorax............................................ 3.2 2.9 Abdomen........................................... 2.5 2.6 Pelvis............................................ 2.7 2.7 ------------------------------------------------------------------------ Table 2.--Internal Biofidelity Rankings of the ES-2 and ES-2re ------------------------------------------------------------------------ Internal biofidelity rank ES-2 ES-2re ------------------------------------------------------------------------ Overall with T1 (w/o abdomen)..................... ......... 1.5 Overall with Defl. (w/o abdomen).................. 1.6 1.6 Overall with TTI (w/o abdomen).................... n/a 1.6 Head*............................................. 1.0 1.0 Thorax--T1........................................ n/a 1.5 Thorax--Delft..................................... 1.7 1.8 Thorax--TTI....................................... ......... 1.8 Abdomen........................................... n/a n/a Pelvis............................................ 2.1 2.0 ------------------------------------------------------------------------ * In its comment, the Alliance pointed out an error in the internal biofidelity score for the ES-2 head, contained in Table 5 of the NPRM (69 FR at 55554, column 3). Table 5 indicated that the ES-2re head received a score of 1.0 while the ES-2 scored a 1.6. As shown in this corrected Table 2, both dummies scored a 1.0 for head internal biofidelity using the NHTSA ranking system. Conclusion: Back plate loading is an undesirable feature of the ES- 2 dummy (see NHTSA Technical Report, ``Design, Development, and Evaluation of the ES-2re Side Crash Test Dummy,'' May 2004, NHTSA Docket No. 2004-17694-11). The rib extensions of the ES-2re have proven to reduce the likelihood of the dummy's spine and back plate to interact with the vehicle's seat back. NHTSA believes that the rib extensions are a necessary component of the dummy and their inclusion has minimal effect on the dummy's response biofidelity. Accordingly, we conclude that the ES-2re test dummy, with rib extensions, will suitably duplicate the responses of a human in FMVSS No. 214 side impact tests. b. Other Issues Relating to How Humanlike the Dummy Is Commenters, primarily the Alliance, raised other issues relating to the humanlike qualities of the ES-2re. The Alliance's comment included a discussion of full-vehicle tests conducted by the OSRP, Toyota, and Transport Canada. The OSRP conducted matched-pair full-scale vehicle tests to compare the responses of the ES-2re, ES-2, and WorldSID in two conditions: (a) FMVSS No. 214 MDB tests at 33.5 mph of a 4-door, mid- size sedan, no air bag and a 4-door, small sedan, head/torso side air bag (SAB); and (b) oblique pole test at 20 mph, 15[deg] impact angle, of a 4-door, small sedan, head/torso SAB. The majority of the Alliance's comments regarding the OSRP tests compared the ES-2re responses to those of the WorldSID, to support the commenter's opinion that the ES-2re is not as humanlike as the WorldSID. We respond in this section to the issues raised by the commenters relating to the acceptability of the ES-2re as a test device for FMVSS No. 214. We will not discuss whether WorldSID is a more humanlike device than the ES-2re because the WorldSID dummy is still [[Page 75308]] under development. As recently as the spring of 2006, the WorldSID design was changing and has not been assessed for its suitability as a compliance test instrument. In short, WorldSID will not be ready for some time to attain the advancements in side impact occupant protection that the agency can achieve today with the ES-2re test dummy. 1. Anthropometry of Abdominal and Pelvic Regions The Alliance believed that the EuroSID family, including the ES-2 and the ES-2re test dummies, is too narrow in the abdominal and pelvic regions as compared to ``the UMTRI anthropometry,'' whereas, the commenter believed, WorldSID is representative of the United States and world populations. Agency Response: In support of its comment, the Alliance references a figure in its submission that provides a coronal-plane view of the ES-2 dummy and the WorldSID. The figure identifies the ES-2 pelvis breadth as 364 mm and the abdominal breadth as 282 mm, while the WorldSID's corresponding dimensions are labeled as 420 mm and 240 mm. (NHTSA believes that the Alliance made an error in its label and that the correct WorldSID abdomen dimension should be 340 mm.) In its submission, the Alliance states: ``The anthropometry of the U.S. population is detailed in a study by UMTRI (1985)\1\. [Footnote in text.]'' The footnote only states ``UMTRI 1985'' without a complete bibliographic reference. NHTSA believes that the Alliance is referring to the University of Michigan Transportation Research Institute (UMTRI) document ``Anthropometry of Motor Vehicle Occupants,'' Volume 1, 1983, performed under NHTSA contract DTNH-80-C-07502. In this UMTRI study, the pelvis and abdominal breadths of the mid-sized adult male are reported to be 385 and 325 mm, respectively. Table 3 below, ``UMTRI, ES-2re and WorldSID Dimensions,'' summarizes the UMTRI dimensions and compares them to the corresponding dimensions in the ES-2re and WorldSID. Table 3.--UMTRI, ES-2re and WorldSID Dimensions -------------------------------------------------------------------------------------------------------------------------------------------------------- Delta, UMTRI vs Dimension UMTRI ES-2re* Delta, UMTRI vs ES-2re WorldSID WorldSID -------------------------------------------------------------------------------------------------------------------------------------------------------- Abdomen breadth................... 325 mm................ 282 mm................ -43 mm................ 340 mm............... +15 mm Pelvis breadth.................... 385 mm................ 366 mm................ -19 mm................ 420 mm............... +35 mm -------------------------------------------------------------------------------------------------------------------------------------------------------- * The ES-2re dimensions are based on the Eurosid specifications derived from European anthropometric studies. From the table, it is observed that the ES-2re does have an abdomen and pelvis that are slightly narrower than the UMTRI target dimension. However, to our knowledge this is of no consequence. Discrepancies relative to the anthropometry targets are often necessary to balance a number of design issues, such as the need to fit the dummy with electronic instrumentation for injury assessment capabilities, component durability, and repeatability of the responses.\9\ The Alliance did not provide any information regarding potential adverse effects that might result from the abdomen and pelvis being slightly narrower in the coronal plane and NHTSA is not aware of any adverse effects associated with the commenter's claim. Accordingly, NHTSA believes that the current dimensional properties of the ES-2re abdomen and pelvis are satisfactory for their intended purpose. --------------------------------------------------------------------------- \9\ We note that the WorldSID's abdomen and pelvis are slightly wider than the UMTRI dimension, which may also be inconsequential. --------------------------------------------------------------------------- 2. Sitting Height The Alliance commented that the pelvis of the ES-2re does not account for compression of soft tissue that occurs when a person is seated in a vehicle seat, and results in a seating height difference between the ES-2re and WorldSID of 58 mm, with the ES-2re seated higher. Agency Response: The comment did not provide any information as to why the seating height of the ES-2re is not adequate for the dummy's intended application.\10\ It appeared that the commenter assumed that the WorldSID seating height is accurate and the ES-2re's seating height is erroneous because it does not match that of the WorldSID. --------------------------------------------------------------------------- \10\ Also, no data was provided regarding what type of vehicle was used or what seating procedure was applied that resulted in the alleged 58 mm difference. Different vehicle seat configurations and materials will play an important role in the seating height of the dummy and, in the absence of any detailed information, it was not possible for us to further examine the assertion. --------------------------------------------------------------------------- NHTSA's review of sitting height anthropometry shows that the mean value of the erect sitting height of the 50th percentile male is 911 mm (reference UMTRI-83-53-1). The designed erect sitting height of the ES- 2 is 909 mm (reference E/ECE/324, Regulation No. 95, October 1, 2004). Comparable design targets for the WorldSID are not yet published. NHTSA attempted to measure the erect seating height of a sample WorldSID dummy, however, making a comparable measurement proved to be somewhat problematic. The WorldSID's pelvis is designed to have an automotive- seated posture and is somewhat resistant to being placed into an erect posture. We measured the WorldSID to have a sitting height of 850 mm. While we do not have data for an average seated occupant height, the UMTRI data indicate that the ES-2re for the intended application is representative of the seated height of real people. 3. ES-2re's Representation of Large Male Population In the September 15, 2004 NPRM (Docket 18864), NHTSA presented injury and fatality statistics in Tables 1 and 2 of that document. Table 1 represented the entire U.S. motor vehicle population. The NPRM stated, ``Of these [statistics in Table 1], approximately 35 percent are small stature occupants. The remaining occupants fall into the midsize and large segments of the population. The ES-2re dummy would address the risk of injury of these occupants in side impacts.'' The Alliance disagreed with NHTSA's assertion that the ES-2re would address the risk of injury for the large-sized segment of the population. The Alliance stated, ``[T]he ES-2re dummy anthropometry and weight are not representative of a large male.'' Agency Response: The agency has assigned benefits to the 50th percentile adult male and 5th percentile adult female dummies in a similar manner as that conducted in the advanced air bag final rule of FMVSS No. 208 (65 FR 30680; May 12, 2000). The countermeasures developed for the 50th percentile male are likely to benefit the 95th percentile adult male. Differences in height between a midsize male and large male occupants in the UMTRI [[Page 75309]] contoured seat study is 2.6 cm (approximately 1 inch), and in standardized normal driving posture is 5 cm (1.96 inches) (UMTRI-83-53- 1). The above data indicate that in a vehicle, the head of an ES-2re dummy would be lower than that of a large (95th percentile) male occupant by approximately 1 to 2 inches. FMVSS No. 214 pole test data indicate that curtain bags, at an inflated stage, come down far enough to cover the head of the ES-2re. Since the head of the seated 95th percentile male is higher than that of the ES-2re 50th percentile adult male dummy, the countermeasures developed to meet the test using the ES-2re 50th percentile adult male dummy are likely to provide similar benefits to the 95th percentile adult male occupant. 4. Abdominal Instrumentation The Alliance stated that OSRP reported that the ES-2re measured abdominal forces below an injury assessment reference value (IARV) in full-scale tests, whereas WorldSID measured abdominal deflections above an IARV.\11\ The commenter also stated that an upcoming research paper will report that the ES-2re is inadequately instrumented in the abdominal region, allowing it to miss important vehicle interactions. The Alliance stated that, in contrast to the ES-2re, the WorldSID presents a continuous surface through the thorax and abdomen up to the pelvis region, that is fully instrumented in the thorax and abdomen regions to ensure that all dummy to vehicle interactions are measured. --------------------------------------------------------------------------- \11\ The Alliance did not provide any data to substantiate a basis for comparison among tests, such as equivalency of vehicle crash pulses or intrusion patterns. --------------------------------------------------------------------------- Agency Response: The ES-2re makes possible a more complete assessment of vehicle performance in side impacts than the SID or the SID/HIII, which will lead to greater side impact protection for occupants. In a NASS study of side impact crashes, it was estimated that between 8.5 percent and 21.8 percent of all AIS 3+ injuries are to the abdomen of restrained near side front seat occupants.\12\ The abdominal load cells are sufficiently sensitive to measure the potential for injury. In an FMVSS No. 214 moving deformable barrier (MDB) test described in the May 2004 NPRM (69 FR at 28010, Docket 17694), the ES-2re detected a high abdominal force in the Chevrolet Impala at the dummy's abdominal area that was caused by an intruding armrest. In full-scale vehicle oblique pole tests conducted by the agency (see ``NHTSA Fleet Testing for FMVSS No. 214 Upgrade MY 2004- 2005,'' discussed in Section V of this preamble), three vehicles exhibited loads which exceeded the IARV for the abdomen: the Ford 500, Chevy Colorado, and Ford Expedition. Because the current side impact dummy used in FMVSS No. 214 does not measure abdominal force, this potential injury risk will be newly detected by the ES-2re. --------------------------------------------------------------------------- \12\ Samaha, R.S., Elliot, D., ``NHTSA Side Impact Research: Motivation for Upgraded Test Procedures,'' supra. --------------------------------------------------------------------------- The commenter failed to show that the abdominal measurements of the ES-2re are problematic or deficient. The injury measuring capabilities of the ES-2re and the WorldSID are different. The WorldSID IARV for abdomen is based on abdomen rib deflection, while the ES-2re's IARV used in the FMVSS No. 214 final rule is based on loads measured at the abdomen (abdominal force limit of 2,500 N). Limiting the load to the abdomen will lead to important gains in occupant protection. The agency also believes that the ES-2re is well instrumented in the abdomen region. The abdomen instrumentation is appropriately located and sensitive to lateral loading in the region above the pelvis and below the ribs. ES-2re drawing number 175-0000, sheet 4 of 5, provides information regarding the location of the abdominal load cells with respect to the pelvis and the lower rib of the thorax. The abdominal load cell extends from just below the upper surface of the pelvis, upward across the abdominal region, and ends approximately 50 mm below the lower surface of the lower thoracic rib. The load cell provides adequate coverage for measuring loads imparted to the abdominal region. 5. Shoulder Design The Alliance referred to matched pair full-scale oblique pole tests that the commenter said Transport Canada (TC) conducted with the WorldSID and ES-2re. The Alliance stated that visual observations made in the TC study indicated that the ES-2re shoulder ``rotated significantly'' while the WorldSID shoulder ``deflected laterally inward towards the spine of the dummy.'' ``This [WorldSID's] motion is similar to the human shoulder tests run by Compigne et al,'' which, the Alliance stated, showed that ``the human shoulder deflects in oblique impact instead of rotating away from the impact'' or ``compresses inward and moves slightly backwards during loading from the front or directly from the side.'' The Alliance stated that the ES-2re dummy's shoulder rotates away from intruding structures, which can lead to a ``reduced excursion of the head when compared to WorldSID head kinematics'' and ``lower rib deflections [compared to WorldSID] that were evenly distributed across the ribs.'' To illustrate its comment, the Alliance referenced a Figure 18 in its submission, which depicted several camera images from tests on an Audi vehicle with thorax and window curtain side air bags using the ES-2re dummy and the WorldSID. The commenter also stated that in full-scale vehicle crash tests, ``The components of force measured at the shoulder of the ES-2re describe a combined loading characterized by equivalent longitudinal and lateral forces whereas the WorldSID forces are purely lateral.'' Agency Response: Test data indicate that the ES-2re's shoulder is fully acceptable. There is no indication of any detrimental effects in vehicle crash tests relating to the ES-2re's shoulder design, such as rib flat-topping which might occur when the shoulder has reached its limit for range of motion. Further, upon examination of the Alliance's Figure 18, we observe that: (1) The ES-2re's shoulder and head appear to be higher relative to the vehicle interior than that of the WorldSID; (2) the ES-2re's shoulder interacts substantially with the side curtain air bag, whereas the WorldSID's shoulder does not appear to contact the window curtain air bag; (3) the ES-2re's head contacts the window curtain air bag higher than does the WorldSID's head, and possibly makes contact with the upper portion of the door trim. These observations indicate that the ES-2re and WorldSID dummies experienced different loading patterns, consistent with the lower seated height of the WorldSID. To the extent that the WorldSID development has not yet been completed, any assessment about differences in kinematics and impact responses between the two dummies is premature. Also, scientific information is not available at this time to support a determination as to whether the ES-2re or the WorldSID has a better shoulder design. We believe the commenter's reference to the Compigne study is not relevant. The Compigne research studied localized pendulum impacts to the shoulder in a controlled test environment, whereas the full-scale oblique pole crashes conducted by TC resulted in loading over a much broader area of the dummy, with no controls on the direction or magnitude of the loading. With regard to internal shoulder loading, the scientific literature on this subject has not characterized internal shoulder loads recorded during lateral and oblique shoulder impacts. In the studies, only [[Page 75310]] pendulum impact loads, an external load, have been recorded. In the absence of such data, it is not possible to establish a biofidelic basis for internal shoulder loads or to determine whether the ES-2re's or the WorldSID's internal shoulder responses better represent those of a human shoulder. 6. Rib Deflections The Alliance's comment included a discussion of full-vehicle tests conducted by the OSRP, Toyota, and Transport Canada. The OSRP conducted matched-pair full-scale vehicle tests to compare the responses of the ES-2re, ES-2, and WorldSID in two conditions: (a) FMVSS No. 214 MDB tests at 33.5 mph of a 4-door, mid-size sedan, no air bag and a 4-door, small sedan, head/torso side air bag (SAB); and (b) oblique pole test at 20 mph, 15[deg] impact angle, of a 4-door, small sedan, head/torso SAB. The majority of the Alliance's comments regarding the OSRP study are comparisons of the ES-2re responses to those of the WorldSID and ES-2. A. Rib Deflections of ES-2re vs. WorldSID in Perpendicular Impacts. The Alliance believed that in perpendicular impacts, the ES-2re exhibited higher rib deflections than either the WorldSID or ES-2. Agency Response: We note that the Alliance did not provide any data to substantiate a basis for comparison among tests, such as equivalency of vehicle crash pulses or intrusion patterns. Rib deflection response variation could be attributed to variations in crash pulse or intrusion patterns, which were not quantified in the Alliance's submission. Further, with regard to the comparison between the ES-2 and the ES- 2re, an increase in rib deflection is not unexpected or surprising. The ES-2re's rib extensions and modified back plate prevent the spine box from interacting with the vehicle seat. That interaction had limited the lateral torso translation of the ES-2 and provided an unrealistic load path in the dummy. Loads that would be absorbed by the spine box of the ES-2 are directed to other body segments in the ES-2re, such as the thorax, and thus a greater rib deflection in the ES-2re is anticipated. With regard to the comparison of ES-2re rib deflections with those of the WorldSID, the observation that the ES-2re exhibited a different amount of rib deflection than that of the WorldSID does not indicate a shortcoming with the ES-2re. To the extent that the WorldSID development has not been completed, specific comments about differences in rib deflections in vehicle crash tests or comparative biofidelity between the two dummies are premature. B. Rib Deflections of ES-2re vs. WorldSID in Oblique Loading. The Alliance stated that the OSRP tests showed that the ES-2re exhibits lower rib deflections than either the WorldSID or ES-2 when subjected to oblique loading in FMVSS No. 214 MDB tests, and that Transport Canada observed ``under oblique loading conditions, the range of WorldSID rib deflections was much greater than the range of the ES-2re rib deflections. * * * Therefore, WorldSID appears to be more sensitive to differences in loading along the torso and better able to discriminate different loading conditions than the ES-2re.'' Agency Response: The observation that the ES-2re exhibited a different amount of rib deflection than that of the WorldSID and ES-2 does not indicate a shortcoming with the ES-2re.\13\ The ability of the ES-2re to measure rib deflections in a meaningful way in a vehicle crash test is discussed in the section, ``Directional Impact Sensitivity,'' infra. Inasmuch as the WorldSID development has not been completed, specific comments about differences in rib deflections in oblique vehicle crash tests are premature. While the agency remains committed to proposing the incorporation of the WorldSID when the dummy is fully developed and shown to be suitable, gains in occupant protection will result from use of the ES-2re in today's side impact testing. --------------------------------------------------------------------------- \13\ Furthermore, rib deflection response variation could be attributed to variation in crash pulse or intrusion patterns, which were not quantified in the Alliance's submission. We note also that the validity of the WorldSID's rib deflection responses in a vehicle crash test has not been established. --------------------------------------------------------------------------- 7. Rib Extensions A. Back Plate Loads. The Alliance stated that the ES-2re back plate displayed reduced lateral loads and increased longitudinal loads as compared to the ES-2 when tested in FMVSS No. 214 MDB tests. Agency Response: The ``no rib grab'' modifications made to the ES-2 dummy are intended to preclude the dummy's spine from acting directly as a lateral load path. Thus, it is reasonable to expect reduced lateral loads in the backplate of the ES-2re and somewhat increased front-to-back loading as the dummy interacts with the curvature of the seatback. The Alliance did not offer any supporting evidence that would indicate that the increase in longitudinal loads was unrealistic or that it resulted in any type of detrimental effect. NHTSA is unaware of detrimental effects that would arise due to increased longitudinal loading of the back plate. B. Load Path. The Alliance also provided comments on Toyota full- scale vehicle tests in which the performance of the ES-2 and ES-2re were compared for oblique pole impacts. The commenter stated that during the oblique pole test, the door trim separated from the back of the door and struck the dummy's torso obliquely from the rear. The commenter believed that the rib extensions in the ES-2re provide a load path not found in the ES-2, and thus rib deflections for the ES-2re were greater than that observed in the ES-2. Agency Response: NHTSA believes that the rib extensions found in the ES-2re represent a more humanlike continuous loading surface configuration than that of the ES-2. Since the ES-2 does not have structural elements at the oblique posterior location, there is nothing to impact, and so it is reasonable to expect lower rib deflections for oblique rear loading conditions than would occur for either the ES-2re, or in humans, under similar loading. c. Repeatability and Reproducibility A dummy's repeatability and reproducibility is typically based on the results of component tests and sled tests. (Repeatability is the similarity of responses of a single dummy measured under multiple identical test conditions. Reproducibility is the smallness of response variability between different dummies of the same design under identical test conditions.) In the tests, the impact inputs as well as the test equipment are carefully controlled to minimize external effects on the dummy's response. Component tests are typically better controlled than sled and vehicle tests, and thus produce more reliable estimates of the dummy's repeatability and reproducibility than is possible in the latter-type tests. Component tests are used to establish the dummy's component performance relative to the biomechanical corridors to which each major body segment must correctly respond. That is, if the dummy's component is or becomes deficient, the component test will identify to the user that the component will not respond properly in impact tests. Sled tests offer a method of evaluating the dummy as a complete system in an environment more like a vehicle test. Sled tests establish the consistency of the dummy's kinematics, its impact response as an assembly, and the integrity of the dummy's structure and instrumentation under controlled and [[Page 75311]] representative crash environment test conditions. NPRM The NPRM stated that the agency's component and sled repeatability and reproducibility tests were based on two dummies. (See ``Technical Report--Design, Development and Evaluation of the ES-2re Side Crash Test Dummy,'' Docket 17694.) Component Tests The component tests were conducted on head, neck, shoulder, upper rib, middle rib, lower rib, abdomen, lumbar spine and pelvis body regions. The repeatability assessment was made in terms of percent CV (Coefficient of Variance). A CV value of less than 5 percent is considered excellent, 5-8 percent good, 8-10 percent acceptable, and above 10 percent poor.\14\ The repeatability of the dummies was assessed in two separate series of tests. In the first series, the dummy calibrations were performed between sled or vehicle crash tests. In the second series, the calibration tests were performed consecutively without any other intermittent tests. In the first series, nine tests were performed with one of the dummies, and seven tests with the other. In the second series, two newly acquired dummies were exposed to five sets of calibration tests each. Reproducibility was assessed by comparing the average responses of both dummies. --------------------------------------------------------------------------- \14\ ISO/TC22/SC12/WG5. --------------------------------------------------------------------------- The results of the component repeatability tests indicated ``excellent'' and good repeatability for the ES-2re dummy for all components except for the pelvis, which had a rating classification of ``good,'' and the shoulder with a rating of ``acceptable.'' The reproducibility assessment was made in terms of response differences between each of the two sets of dummies with respect to the mean. The rating for reproducibility takes into account the cumulative variabilities of two or more dummies and is primarily indicative of the repeatability of the manufacturing process of the same type of dummy and to some extent the repeatability of design specifications, inspection, and test methodology. The reproducibility assessment does not serve the purposes of accepting or rejecting the dummy; rather it is an indication of how far the responses of different dummies could vary under identical test conditions. The results of the pooled component tests indicate that the neck, thorax, lumbar spine and pelvis responses are well below the 5% level and the head, shoulder and abdomen response below the 7% level. These levels are quite acceptable and consistent with the repeatability norms. Sled Tests To reduce test-to-test variation of sled pulse parameters, NHTSA tested two ES-2re dummies (designated ``dummy 070'' and ``dummy 071'') simultaneously on a dual occupant side impact Hyge sled buck developed by the agency. The sled pulse was an approximate half-sine wave, with the peak acceleration of 12.7 g's and duration of approximately 80 ms. The impact speed was 6.7 meters per second (m/s) (22 ft/s). Two test conditions were used for the repeatability and reproducibility assessment: a flat rigid wall; and a rigid wall with abdomen offset (simulating a vehicle armrest). The two ES-2re dummies were exposed to two series of five Hyge sled tests, for a total of 10 test exposures per dummy. For the flat wall test condition, the wall was 374 mm (14.7 in) high from the front edge of the seat, and 368 mm (14.5 in) long from the back of the seat. For the abdomen offset test condition, the same flat wall was used, with a protruding 305 mm (12 in) long, 76 mm (3 in) thick and 83 mm (3.3 in) wide wooden offset block attached to the wall. The offset block, simulating an armrest, was oriented such that it would impact the abdomen only, above the pelvis and below the lower rib. The objective of the abdomen offset tests was to provide a test environment with severe loading of the abdominal region. The sled buck incorporated a Teflon-covered bench seat with two Teflon-covered rails to support the seated dummies from behind. As the sled buck was accelerated, the buck slid beneath the dummies until the dummies' left side impacted the rigid wall. High-speed digital video cameras were positioned in front of each dummy in order to capture head motion for use in performing motion analysis of the head translation. The dummies were instrumented with sensors to record principal injury indicators such as head, resultant lower spine (T12) and pelvis accelerations, rib deflections, abdominal, lumbar and pubic symphysis loads, and other parameters. A contact switch was positioned on the side of each dummy and on the load wall at the location of first contact to indicate the precise instant of dummy contact with the wall. Flat Rigid Wall Test Results Using the dummy rating practice set forth in ISO/TC22/SC12/WG5, generally the responses in the flat wall tests displayed either excellent or good repeatability, except for the lumbar Y (shear) force repeatability of dummy Serial Number (S/N) 070 falling outside the CV acceptability boundary at 14.8%. This elevated CV value for dummy 070 also was responsible for a reproducibility assessment at 17.5%. While these CV values are relatively high, the agency is not considering an injury assessment associated with this response. Moreover, this response is not considered to be of importance since it did not have an effect on either the magnitude of the loading or the variability of the adjacent structure responses, such as pubic symphysis, the abdomen and the T12. HIC responses exhibited excellent repeatability of each dummy and reproducibility of both dummies. In all tests, the rib displacement time history provided a smooth response, with no indications of the flat topping phenomena that had been a shortcoming of previous versions of the EuroSID, EuroSID-1, and the prototype ES-2 dummies. Rigid Wall With Abdomen Offset Test Results The responses for the abdomen offset sled tests \15\ provided either excellent or good repeatability and reproducibility, except for one test in which the lumbar moment reproducibility response had a CV value of 16.7, which is only by 1.7% into the poor range. While this CV value is high, this measurement is not considered for injury assessment with the EuroSID, EuroSID-1 and ES-2re dummies. Furthermore, this slightly elevated response appears not to affect either the magnitude of the loading or the variability of the adjacent structure responses, such as pubic symphysis, the abdomen, the T12 moment and the rib displacement time history, without any indications of flat topping. --------------------------------------------------------------------------- \15\ The first test in the series with dummy S/N 070 was excluded. Upon review of the response traces after the test series was completed, it was noted that this test resulted in significantly lower abdominal and lumbar loads and larger rib displacements than in the remaining four tests. (See Appendix C, Figures C.10 through .18 of the Technical Report, Docket 18864-12, supra). Upon review, the data for that test indicated that impact contact with the abdominal offset block appear to have slightly favored the proximity of the lower rib rather than the middle of the abdomen, as had been the case in the subsequent four tests. This could have been caused either by a slight variation in the set-up of the dummy for the test or a slight posture realignment during the dummy's movement while approaching the impact surface. Inasmuch as the seating procedure was not varied and this aberration did not reoccur in the four subsequent tests, this test was considered to be a legitimate outlier. --------------------------------------------------------------------------- Based on the above, the agency tentatively concluded that the repeatability and reproducibility of the [[Page 75312]] ES-2re responses in flat wall and abdominal offset impacts are acceptable (generally in the order of ``excellent''). 1. Sample Size Both the Alliance and Autoliv expressed concerns with the small sample size (n=2) of dummies used to establish repeatability and reproducibility of the ES-2re. The Alliance was concerned that only one dummy manufacturer was represented in the sample. The Alliance stated: ``In order to get a reasonable assessment of dummy repeatability and reproducibility, it is necessary to subject six dummies, of each combination, to the same series of tests.'' Agency Response: At the time NHTSA conducted its evaluation of the ES-2re, only one dummy manufacturer could provide NHTSA with production-ready samples of the dummy. That said, the agency nonetheless believes that the sample size (n=2) used for the NPRM was sufficient. The repeatability and reproducibility studies of the ES-2re described in the NPRM complemented the repeatability and reproducibility work previously conducted on the ES-2 dummy. The ES-2 has been used for testing and research purposes in Europe, the United States and elsewhere for years and has proven repeatable and reproducible performance. The repeatability and reproducibility work on the ES-2re built on those earlier assessments of the ES-2 and showed that the ES-2 with the rib extensions had good to excellent repeatability and reproducibility. At the same time, we recognize that valid data can potentially be gleaned from tests of additional ES-2re test dummies with regard to the certification corridors used to assess performance of the dummy. As explained later in this preamble, the agency has therefore used ES-2re performance data submitted by the Alliance and the SAE in determining the certification corridors of this final rule, since the inclusive database is based on a larger sample size of ES-2re tests. 2. Reproducibility of Pelvic Load Measurements The Alliance expressed concerns ``relative to the reproducibility of the pelvic load measurement of the EuroSID-family of dummies.'' The commenter stated that it analyzed pelvic certification data provided by the SAE, studying the correlation between the internal and external loads measured in the pelvic impact certification test. The Alliance plotted the pubic symphysis load (internal loads) against the impactor force (external loads) and computed the coefficient of determination (R \2\) for the relationship. The Alliance stated that the resulting R \2\ values were low and therefore expressed concern ``* * *that the pelvic load measurement of the ES-families of dummies has a reproducibility issue.* * * '' Agency Response: We disagree with the implication that there is a need to establish controls on the ratio of force input to output for the pelvis certification test. The R \2\ is not a meaningful assessment in this case, because the external loads account for impact inputs through several portions of the dummy, such as friction of the dummy with the seat, lumbar spine shear, and compression of the flesh, whereas the pubic symphysis loading reflects internal loads between the two pelvis halves. Furthermore, the agency conducted an evaluation of the repeatability and reproducibility of the pelvis response in both certification and sled test environments (reference NHTSA-2004-18864-15 and -16, respectively). In certification testing (pendulum testing), the ES-2re dummies exhibited excellent repeatability and reproducibility for all response criteria. In the sled testing portion of the evaluation, both dummies displayed excellent repeatability and reproducibility when exposed to the flat-wall test condition. In the abdomen offset sled test condition, one dummy exhibited excellent repeatability, while the second dummy scored a good rating for repeatability. Taking the certification and sled test results both into account, the dummies' pelvis response provided excellent reproducibility. Given these findings, the agency has concluded that the reproducibility of the pelvic load measurement of the ES-2re test dummy is acceptable. 3. Sensitivity to Initial Conditions The Alliance believed that the ES-2re exhibited ``an unacceptable sensitivity to initial conditions.''