This rulemaking is a first step toward achieving two goals: improving side impact protection and reducing the risk of ejection. Both goals have been highlighted in recent agency planning documents. On July 25, 2002, the agency published a notice requesting public comment on a comprehensive multi-year vehicle safety rulemaking and research plan (67 FR 48599; Docket No. NHTSA-2002-212391). Two months later, NHTSA Administrator Jeffrey W. Runge, M.D., formed Integrated Project Teams (IPTs) to conduct an in-depth review of four top priority safety areas. Among them are vehicle compatibility and rollover. Those two areas were selected because they represent the key safety issues presented by the changing composition of the passenger vehicle fleet. The sales and registrations of light trucks, buses and multipurpose passenger vehicles (LTVs) as a percentage of the light vehicle fleet have steadily increased since 1984. In fact, sales of LTVs reached 50 percent of all new light vehicles sold in 2001. The IPTs were chartered to develop comprehensive, science and evidence-based analyses to identify innovative solutions and recommend effective strategies.
Significant progress has been made in addressing these priorities. On June 18, 2003, NHTSA announced the availability of two reports, "Initiatives to Address Vehicle Compatibility," and "Initiatives to Address the Mitigation of Rollovers,"  based on the work of the vehicle compatibility and rollover IPTs (68 FR 36534). Initiatives to upgrade side impact protection and reduce ejection figure prominently in both reports. One month later, the agency announced the availability of its final priority plan, "NHTSA Vehicle Safety Rulemaking and Supporting Research: 2003-2006"  (68 FR 43972; July 18, 2003). The plan, which reflects the results of a comprehensive examination of areas of possible improvements, "outlines the agency's vehicle safety rulemaking actions for the period 2003 to 2006 that offer the greatest potential for saving lives and preventing injury." Upgrading side impact protection is one of the most promising of those actions.
Today's proposal to upgrade the agency's side impact protection standard begins the implementation of the initiatives in the agency's report on improving crash compatibility between passenger cars and LTVs ("Initiatives to Address Vehicle Compatibility," supra.) This proposal would require vehicle manufacturers to assure side impact protection for a wider range of occupant sizes and over a broader range of seating positions. It would likely lead to the installation of new technologies, such as side curtain air bags and torso side air bags capable of improving head and thorax protection to occupants of vehicles that are laterally struck by a higher-riding LTV. (These different side air bag systems are described in a glossary set forth in Appendix A to this preamble.)
II. Executive Summary
In 1990, the agency amended its side impact protection standard, Federal Motor Vehicle Safety Standard (FMVSS) No. 214, "Side Impact Protection," to include a dynamic test, the first anywhere in the world, that assesses occupant protection when a vehicle is struck in the side by another vehicle. A moving deformable barrier is crashed into the side of a vehicle in a manner that simulates a 90-degree side impact between two moving vehicles at an intersection. The standard addresses thoracic and pelvic injuries to struck-side occupants in those vehicle-to-vehicle crashes.
However, the standard does not address side crashes into fixed narrow objects, which account for approximately 20 percent of deaths and serious injuries that occur in side impacts. It also does not address head injuries, which account for 43 percent of the total deaths and serious injuries in the target population addressed by this NPRM. For smaller-statured occupants, head injury represents a higher proportion of the serious injuries than it does for larger occupants as a result of relatively more head contacts with the striking vehicle. 
The current state of knowledge and practicability of measures that could be taken to improve side impact protection are considerably greater than they were just a decade ago. Extensive work by NHTSA, the industry, and others in the safety community have led to substantial progress in dummies, injury criteria and countermeasures. Inflatable side protection systems have become common in current production vehicles. They vary widely in designs, sizes, mounting locations and methods of inflation, and areas of coverage. For example, variations of side impact protection systems include door-mounted thorax bags, seat-mounted thorax bags, seat-mounted head/thorax bags, and head protection systems that deploy from the roof rails (e.g., inflatable tubes and curtains).
Based on this progress and the growing significance of vehicle compatibility issues, NHTSA is proposing to upgrade FMVSS No. 214 substantially by requiring all passenger vehicles with a gross vehicle weight rating (GVWR) of 4,536 kilograms (kg) or less (10,000 lb or less) to protect front seat occupants against head, thoracic and pelvic injuries in a vehicle-to-pole test simulating a vehicle's crashing sideways into narrow fixed objects like telephone poles and trees.  This would be the first time that head injury criteria would need to be met under the standard. The vehicle-to-pole test is similar to the one currently used optionally in FMVSS No. 201, except that NHTSA proposes to change the angle of impact from 90 to 75 degrees and increase the test speed from 29 to 32 kilometers per hour (km/h) (18 to 20 miles per hour (mph) ).
Vehicles would need to meet the injury criteria using new dummies representing mid-size males and small females. Crash data indicate that 35 percent of all serious and fatal injuries to near-side occupants in side impacts occurred to occupants 5 feet 4 inches (or 163 centimeters)(cm) or less, which are better represented by the small female dummy. Thus, the agency believes that use of both dummies, instead of just the mid-size male dummy, will better represent the at-risk population.
For the mid-size or 50th percentile male, NHTSA proposes to adopt a modified version of the European side impact dummy, the ES-2 dummy, for use in the test, since the overall dummy is technically superior to the SID-H3 50th percentile male test dummy currently used in FMVSS No. 201 and to the SID 50th percentile male test dummy currently used in FMVSS No. 214. The modified ES-2 dummy (known as the ES-2re) is superior in that it has improved biofidelity and enhanced injury assessment capability compared to the other dummies. A predecessor dummy, known as EuroSID-1, is currently specified by European governments for use in perpendicular side impact testing and work has been undertaken to replace that dummy with the ES-2re. The non-governmental European New Car Assessment Program (EuroNCAP) on side impact has used the ES-2 dummy since February 2003 in perpendicular MDB side impact tests.
The small or 5th percentile female dummy has been used by Transport Canada in crash tests in the late 1990s and early 2000, and is used by the Insurance Institute for Highway Safety (IIHS), a nonprofit group funded by insurers, in IIHS's side impact consumer information program which ranks vehicles based on performance when impacted perpendicularly by a moving barrier at about 30 mph. The countermeasures that are installed to meet the proposed pole test would need to enable the vehicle to meet the requirements when tested with both dummies, which would ensure protection for shorter drivers who sit closer to the steering wheel than the mid-size occupant.
We anticipate that vehicle manufacturers will install dynamically deploying side air bags to meet the proposed vehicle-to-pole test. The agency estimates that the proposals in this NPRM would prevent 686 fatalities and 880 MAIS 3 to 5 injuries a year when fully implemented throughout the light vehicle fleet.  Those benefits are based on an assumption that manufacturers would use a 2-sensor (per vehicle) combination air bag system. (This system would be the least costly countermeasure that manufacturers could use to achieve compliance. Manufacturers might also install side air curtains or other measures that not only reduce head injuries, but also can help reduce ejections through side windows.) The cost for the 2-sensor combination air bag system is estimated to be $121 per vehicle. We are proposing to provide significant lead time to ensure that the regulatory burden is practicable and feasible.
In addition, this NPRM proposes to upgrade the moving deformable barrier test in several ways. It would enhance the MDB test's existing chest and pelvis protection requirements and require compliance with head injury criteria. It proposes replacing the current 50th percentile male dummy with the new one mentioned above and requiring compliance with the criteria developed for that new dummy. The proposal would also enhance protection for smaller adult occupants by adding the new 5th percentile female dummy mentioned above and require compliance with the injury criteria for that dummy.
Mindful of the magnitude of this rulemaking and the principles for regulatory decision-making set forth in Executive Order 12866, Regulatory Planning and Review, NHTSA examined the benefits and costs of a variety of potential proposals and, based on that analysis, took reasonable steps to limit the scope of this NPRM. First, because rear seat occupants make up a small percentage of the seriously injured occupants in side crashes, NHTSA has focused the proposal for the pole test on the front seat. (We note that some side air curtains cover both front and rear side window openings and thus would also afford some head protection to rear seat occupants in the absence of a test applying to the rear seat.)
Second, the agency is not proposing a limit on chest deflection in tests using the 5th percentile female dummy. The modified SID-IIs dummy appears to require further refinement in measuring chest deflection for oblique loading conditions, such as those present in the oblique pole and MDB tests, and so the agency wishes to further analyze test data before proceeding with a proposal limiting the chest deflection of the dummy in the tests proposed today. However, the agency will continue to monitor the chest deflection performance of vehicles in tests using the modified SID-IIs dummy.
Third, NHTSA is also not proposing changes to the standard's MDB at this time. Initiatives to improve vehicle compatibility between passenger cars and LTVs in side crashes are likely to change the characteristics of striking vehicles in the future, as countermeasures are pursued to reduce the aggressivity of LTVs in side impacts. Once the likely future changes to the fleet have been identified, we can determine how the FMVSS No. 214 barrier should be modified to better represent future striking vehicles in side impacts. We also believe that the countermeasures resulting from today's proposed pole test would encompass and go beyond those that would be likely to be installed as a result of a higher/heavier barrier.
III. Safety Problem
In the 2001 Fatality Analysis Reporting System (FARS), there were 9,088 side impact fatalities. For our target population, we excluded from these side impact fatalities those cases which included rollovers as first event (203), rear seat occupants (732), middle front seat or unknown seat occupants (327), far-side occupants (2,601), children under 12 in the front seat nearside (71), and delta-Vs not in our assumed effectiveness range of 19 to 40 km/h (12 to 25 mph) (2,084). We also made an adjustment based on the estimated benefits that would result from the FMVSS No. 201 upper interior requirements for the A-pillar, B-pillar, and roof side rail (160).  This left us with a target population of 2,910 fatalities and 7,248 non-fatal serious to critical AIS 3-5 injuries.
The 2,910 fatalities were divided into three groups for the analysis: (a) vehicle to pole impacts (599); (b) vehicle to vehicle or other roadside objects impacts, which include partial ejections in these cases (1,715); and (c) complete occupant ejections in non-rollovers (636). In this target population, 40 percent of the total fatalities are caused by head/face injuries, 38 percent by chest injuries and 8 percent by abdominal injuries. In contrast, for the 7,248 non-fatal AIS 3-5 target population, chest injuries are the predominant maximum injury source accounting for 59 percent, head/face injuries account for 13 percent, and abdominal injuries account for 6 percent. Combining all serious to fatal injuries, chest injuries account for 53 percent, head/face injuries account for 20 percent, and abdominal injuries account for 7 percent.
In April 2001, NHTSA analyzed fatalities in the 1991, 1995, and 1999 FARS files using non-rollover, near-side impact data. The fatalities occurred in the first and second rows of seats in light vehicles in side impacts with various objects. The percentage of vehicle-to-rigid narrow object impacts has remained stable at approximately 21 percent of the total number of fatal side impact crashes. The percentage of collisions with LTVs has increased, while the percentage of collisions with passenger cars has decreased over time. The results of the analysis are presented below:
TABLE 1: Occupant Fatality Distribution
(Non-Rollover Near-Side Impacts)
Collisions With Passenger Cars (Percent) Collisions With LTVs (Percent) Collisions With Rigid Narrow Objects (Percent) Collisions With Other Vehicles/Objects (Percent) FARS 1991
MY 1987 and Later Light Vehicles
28.9 26.3 20.1 24.8 FARS 1995
MY 1991 and Later Light Vehicles
24.7 31.8 21.2 21.9 FARS 1999
MY 1995 and Later Light Vehicles
20.5 35.5 21.1 22.9
 Samaha R. S., Elliott D. S., "NHTSA Side Impact Research: Motivation for Upgraded Test Procedures", 18th International Technical Conference on the Enhanced Safety Of Vehicles Conference (ESV), Paper No. 492, 2003.
 The pole test would apply to the driver and front outboard passenger seats, and not to the rear seats. In contrast, the moving deformable barrier test applies to both the front and rear outboard seating positions on the side of the vehicle struck by the barrier.
In the pole and MDB tests, both sides of the vehicle are subject to testing by NHTSA. Manufacturers must certify that the vehicle complies with the standard when either side of the vehicle is tested by NHTSA. The standard does not require NHTSA to test both sides of the vehicle.
 While 20 mph converts to 32.2 km/h, we propose rounding 32.2 km/h to 32 km/h.
 You may inspect the dummies by contacting our Vehicle Research and Test Center in East Liberty, OH.
 The AIS, or Abbreviated Injury Scale, is used to rank injuries by level of severity. An AIS 1 injury is a minor one, while an AIS 6 injury is one that is currently untreatable and fatal. The Maximum Abbreviated Injury Scale, or MAIS, is the maximum injury per occupant.
 NHTSA also adjusted the target population by assuming increased seat belt use based on 2003 use rates.