V. POTENTIAL BENEFITS OF DEPOWERING BASED ON THE HOLDEN BAG

In the AAMA petition, a discussion was provided of an air bag designed in Australia to provide benefits for belted occupants. This appears to be the only production vehicle with less aggressive air bags for which NHTSA has received any significant information about real-world crash experience. In 1993, the Holden VR Commodore vehicle, manufactured in Australia, was fitted with a driver-side air bag. In 1995, the VS Commodore was fitted with both driver side and passenger-side air bags. These vehicles have experienced crashes and Monash University in Australia has investigated those crashes, as well as a baseline set of Holden vehicles without driver air bags involved in crashes. This field study provides limited data to assess the effectiveness of both restraint systems in preventing injury. Because belt use is over 95 percent in Australia, the comparison of injury rates for the two vehicle restraint populations allows an estimate of the incremental effectiveness of air bags relative to lap-shoulder belts.

The air bag systems in the Commodore are designed to deploy as unaggressively as possible while still providing the necessary protection to occupants of different size, weight and sex who will be potentially involved in a variety of collisions. Great efforts have been taken in the development of the inflators and cushions to ensure they present as little risk as possible to occupants during inflation. Since the air bags have been designed to operate in conjunction with the safety belts, they are only required to decelerate the occupant's head and upper torso, as the primary load path is through the belts. This is fundamentally different from many other air bag designs, especially those used to protect unrestrained occupants. Systems optimized to protect unrestrained occupants typically utilize high-output inflators in conjunction with cushions with low venting rates. This combination ensures that the air bags are sufficiently stiff to decelerate unbelted occupants. The field test of the Holden air bag resulted in a total of 148 drivers in frontal crashes, 63 with air bags and 85 without air bags. Data from this study (see Table V-1) indicate that air bags with lap/shoulder belts reduced AIS 2+ injuries by 39 percent compared to lap/shoulder belts alone. Since the Holden air bag has a higher deployment threshold than air bags in the U.S., it is not a perfect comparison with U.S. crash data, but this 39 percent effectiveness level is higher than the level calculated using NASS data of 22 percent for U.S. air bag systems by a factor of 1.77 (39/22).

The data from the Holden bag indicates that there could be an increase in effectiveness of the depowered air bag for a lap-shoulder belted occupant, compared to the effectiveness of U.S. air bags in combination with lap-shoulder belts. Test data provided in Chapter III indicate that on the passenger side there was a small increase in benefits for belted occupants (e.g. lower HIC and chest g's) with depowered air bags. The same finding was not made on the driver side, where modelling indicates there were minimal increases in HIC and chest g's measured on the belted dummies, but it could be possible; the theory being that softer air bags optimized for a belted population would provide a higher level of protection than current U.S. systems. If this were true, the findings from this evaluation of the increased fatalities to the unrestrained occupants from depowered bags could be offset by safety benefits to the restrained population.

Comparing the Holden air bag experience to the U.S. air bag experience finds that the Holden bag provided a higher overall benefit than the U.S. air bag (See Tables V-1 and V-2 -- The "Air Bag Effect" estimates are comparable between the two tables). In addition, the Holden bag showed a higher effectiveness for serious (MAIS 3+) chest injuries (73 percent) than the U.S. air bags for chest injuries (26 percent). In these more serious crashes, the different deployment thresholds between the U.S. and Holden bags would come less into play, but the number of cases of serious injury in the Holden bag analysis is limited. Based on the testing and analysis provided earlier on U.S. bags, one would have expected that a less aggressive air bag would show about the same or less benefit for serious chest injuries as a more aggressive air bag for belted occupants. However, if a less aggressive air bag is optimized for belted occupants, the Holden bag crash experience indicates that more benefit can be achieved for chest injuries. In this case, a less aggressive bag could result in a reduction of fatalities for belted occupants compared to current air bags.

Current analyses indicate that air bags are 8.5 percent effective in reducing belted fatalities. Based on calculations discussed on Page I-13, the target population is 20,924 drivers and 6,491 right front passengers for a total of 27,415 fatalities. Of those total 27,415 fatalities, an estimated 9,887 (7,248 drivers and 2,639 right front passengers) are restrained occupant fatalities. If the relationship in overall effectiveness of the Holden bag to the U.S. bags for AIS 2+ injuries was the same for fatalities, air bag effectiveness for belted occupants would be 15 percent (1.77 x 8.5 percent effectiveness) Thus, if less aggressive air bags increased the overall effectiveness of air bags for restrained occupants from 8.5 percent to 15 percent, the estimated benefit would be an incremental life saving of 471 for drivers [(7,248 x .15) = 1,087 - (7,248 x .085 = 616) = 471]. The incremental life saving for right front seat passengers would be [(2,639 x .15) = 396 - (2,639 x .085 = 224) = 172]. Thus, the total potential savings based on the Holden bag analysis is 643 lives saved.

Table V-1 Holden Bag Injury Rates and Effectiveness Estimates
Restraint System	MAIS 3+ Overall	MAIS 3+ Chest	MAIS 3+ Head
Air Bag + L/S Belt	0.030	0.016	0.016
L/S Belt Alone	0.070	0.059	0.024
Air Bag Effect	57%	73%	33%
	MAIS 2+ Overall	MAIS 2+ Chest	MAIS 2+ Head
Air Bag + L/S Belt	.019	0.063	0.048
L/S Belt Alone	.031	0.141	0.129
Air Bag Effect	39%	55%	63%
L/S = Manual Lap/Shoulder Belt

Table V-2 U.S. NASS/CDS Effectiveness Estimates
Restraint System	MAIS 3+ Overall	MAIS 3+ Chest	MAIS 3+ Head
Air Bag + L/S Belt	59%	66%	75%
L/S Belt Alone	60%	54%	38%
Air Bag Effect	neg.	26%	60%
	MAIS 2+ Overall	MAIS 2+ Chest	MAIS 2+ Head
Air Bag + L/S Belt	60%	59%	83%
L/S Belt Alone	49%	14%	59%
Air Bag Effect	22%	52%	59%
L/S = Manual Lap/Shoulder Belt

VI. COST

Costs are not a significant issue in this rulemaking. Depowering could be accomplished by a simple substitution of a lower powered inflator, which might reduce variable costs by a small amount. However, there will be engineering, development, and testing costs to design, develop, and certify the depowered air bags.

The generic sled test includes a designated time of air bag deployment initiation in addition to the universal crash pulse for all vehicles. Since the actual crash pulse and deployment initiation time associated with each different vehicle platform will not be addressed by the generic sled test, it will be theoretically possible for greater commonality of air bag module components among all vehicles. If commonality is focused on and accomplished by the industry, air bag module cost reductions could result by the economies of scale of producing millions of identical components rather than the current 100 thousands.

Lead Time

Safety groups and air bag manufacturers (e.g., AORC, NAII, Public Citizen, NTSB, CAS, ATC, NAGSAR, GDDS, AVS Technologies, TRW Automotive, Morton Automotive) strongly favor a temporary amendment to FMVSS No. 208, or an interim measure (also referred to as "sunset" provisions), which would permit depowering. They are concerned about; (1) the accumulation of disbenefits if the amendment is in place too long and (2) the vehicle manufacturers not having any incentive to develop smart air bag systems, if the amendment is made permanent. They believe the termination of the depowering allowance should be linked with the introduction of smart air bags. CAS suggested a 2 year "moratorium" on FMVSS No. 208 unbelted testing.

On the other hand, the manufacturers (AAMA, AIAM, Volvo, Isuzu, VW, Nissan, Honda, Mitsubishi) and IIHS strongly oppose making the amendment temporary claiming that depowering is beneficial to most occupants, has no significant disbenefits, and the availability of smart air bags is uncertain. AIAM and Isuzu specifically called for de-linking any provision from the implementation of smart air bags. The timing of smart air bag technology is unknown at best and, as far as the industry is concerned, there would be little, if any, disbenefits to depowering with which to justify a smart air bag rule. The manufacturers also expressed concern that a temporary amendment would affect planning decisions.

The agency is adopting a temporary amendment with a termination date of September 1, 2001. A temporary amendment promotes the development of smart air bags systems, ensures that depowering disbenefits will be short-term and accommodates solutions such as higher deployment thresholds and dual level inflators. However, if a temporary amendment were made too short, say two years, it might discourage depowering efforts and could discourage optimizing depowered air bag systems. Given the lag in time between model year and crash data collection/recordation, a temporary amendment would probably not be able to be accurately assessed by the agency until after the allowance is terminated. The agency will have to rely on crash investigation case studies to do this on a real-time basis during the temporary allowance period.

AAMA supports the sled test provision as this would allow depowering of 100 percent of new cars in 2 years with 6-9 months lead time. Isuzu indicated they would need about a 6 months of lead time. They stated that if the final rule was published in March, 1997, they could have depowered air bags in some of their MY98 designs. Subaru stated that if the final rule on depowering was published in February, 1997, they could introduce depowered air bags in the 1998 models. Toyota agreed with Subaru. Ford suggested that NHTSA make depowering applicable to all MY98 vehicles. In addition, Ford indicated that some manufacturers already have implemented depowered air bags or are well along in plans to do so. The manufacturer's consensus, although limited, appears to be September 1, 1997, or MY98 before they could begin to introduce depowered air bags. Advocates and Morton suggested that the amendment should be effective immediately, meaning depowered products can be marketed immediately as long as they comply with the new unbelted sled test requirements. Further, Advocates suggested that NHTSA publish advanced restraint system performance requirements (revised FMVSS 208) by September, 1997, with 2 years lead time and 2 years phase-in.

GDDS stated that they do not believe that dual stage or 2 stage inflator products would be available on the U.S. market until MY99 and estimate MY99 for their own product a variable control inflator. Bosch indicated that some of the sub-system building blocks for smart air bags may not be available until MY99. Ford indicated that they have been working with Bosch to develop a rear-facing infant carrier detection system, based on infrared/ultrasound technology, for light trucks. This system cannot detect a child that slides forward with pre-impact braking.

NHTSA received many comments to the docket (74-14- Notice 100, September 20, 1996) concerning the components and sub-systems of technology needed for smart air bags which have been assembled into prototype systems. The agency recognizes that it will be several years before highly reliable and fully perfected advanced or smart air bag components, sub-systems and systems can be implemented across the entire fleet.

The RVIA and Atwood Mobile Products, which serve or support the conversion vehicle industry requested another year be added to the effective date of the depowering amendment to accommodate incomplete vehicle manufacturers. The RVIA stated that final stage manufacturers do not have access to either vehicle or vehicle components until after the single stage manufacturer's deadline for compliance. For Atwood, a supplier of restraint hardware (e.g., seat frames, pedestal seats, seat adjusters, seat belt attachment hardware and seat belts), to incomplete manufacturers, depowering of air bags would require the re-testing or re-design of belt attachment hardware or belt systems, as depowering would increase loads and forces.

The agency notes that many of these final stage manufacturers are small entities.

The agency has decided to specify a termination date of September 1, 2001. The agency reserves the right to re-visit these dates in the smart air bag rulemaking as new information about the reliability and production availability of smart air bag components and sub-systems becomes available. Based on information the agency has received, it believes that some kind of smart air bags or other solutions would be available by MY 2002 that provide at least the benefits associated with depowering without the potential tradeoffs.

The agency has decided not to differentiate the effective date based on whether the vehicle is manufactured in multiple stages. This final rule allows depowering and does not require depowering at any given time. Thus, the agency does not know when depowering would occur, nor can it determine a date by which vehicles manufactured in multiple stages would be likely to be able to certify depowered air bags. This change is like any other voluntary change made by the manufacturers. Typically, the first stage manufacturer provides to a second stage manufacturer a compliance envelope of weight and center of gravity specifications. As long as the second stage manufacturer stays within this compliance envelope, there is a pass-through certification to the second stage manufacturer. The second stage manufacturer has problems when it wants to make a vehicle outside the compliance envelope. These same problems exist today for second stage manufacturers. However, if a depowered air bag is used by a first stage manufacturer, certified by using the new generic sled test, then the second stage manufacturers need to determine how they can certify compliance for vehicles made outside the compliance envelope. This may take some time. However, the agency does not know when the clock begins, because it is a voluntary standard.

Second stage manufacturers use vehicles from different first stage manufacturers. It might work out that they will introduce depowered air bags at different times, in which case a second stage manufacturer can use models that have not been changed for several months, while certifying those vehicles that have been changed. First stage manufacturers have a strong sales incentive to provide second stage manufacturers with information as soon as possible about product changes.

VII. SMALL BUSINESS IMPACTS

The Regulatory Flexibility Act of 1980 (Public Law 96-354) requires agencies to evaluate the potential effects of their proposed and final rules on small business, small organizations and small governmental jurisdictions.

This final rule affects motor vehicle manufacturers, and air bag manufacturers, almost all of which would not qualify as small businesses. There are some small volume manufacturers that operate primarily in the United States that would qualify as small businesses. The Coalition of Small Volume Automobile Manufacturers, Inc. supports having alternatives available that give small business more flexibility. This final rule allows an additional alternative method of compliance, which small businesses could utilize. No vehicle manufacturer, regardless of its size, will be required to take any action as a result of the final rule. Manufacturers can continue using the air bags already designed for their vehicles.

There is also an issue of leadtime for second stage manufacturers that are small businesses. This issue was discussed in the previous section.