December 7, 2010
The National Highway Traffic Safety Administration wants to know what you think about its latest technical report on the non-effectiveness of rear underride guards. The request for comments is part of a long, slow evaluation process of FMVSS 223 and 224, which require the underride guards meet a strength test on trailers with a GVWR of 10,000 pounds.
The standard has been in effect since 1998. In 2004, the agency announced that it would be evaluating the efficacy of these standards. The report, a statistical analysis of crash data from two states involving trucks with compliant underride guards found no statistically significant preventative effect. The agency looked at data from Florida and North Carolina and found decreases in fatalities and serious injuries to passenger vehicle occupants in a rear-end crash with a tractor-trailer.
“However, the observed decreases are not statistically significant at the 0.05 level, possibly due to the small sample sizes of the data,” the agency said. Supplemental data from North Carolina showed that “passenger vehicle passenger compartment intrusion is more apt to occur when the corner of the trailer is impacted, rather than the center of the trailer.” Again, however, the result was “statistically significant at the 0.01 level.”
In sum, the agency concluded that it couldn’t get enough data to show much of anything:
“It is not possible to establish a nationwide downward trend in fatalities when a passenger vehicle rear-ends a tractor-trailer – neither in terms of total number of fatalities, percentage of fatalities in rear impacts relative to other passenger vehicle fatalities involved in tractor-trailer accidents, nor number of fatal crashes per 1,000 total crashes. The Fatality Analysis Reporting System does not list the model year of the trailer.”
Is this another case of research malfunction? The Insurance Institute for Highway Safety recently blasted an agency study on the effectiveness of ABS brakes in motorcycles, calling it “junk science.” The NHTSA study compared motorcycles with and without ABS, using two sets of data– fatal crashes and all police-reported crashes. The case comparison “did not find statistically-significant results to suggest that ABS affects motorcycle crash risk.” The IIHS, which had been urging NHTSA to propose a rule requiring ABS brakes in motorcycles, argued that the NHTSA results flew in the face of many other studies.
Or, maybe it’s a case of standard failure. In 1996, the agency went the don’t-ruffle-industry’s-feathers route, passing a final rule that offered only a slight improvement over the Truck Trailer Manufacturers Association’s voluntary recommended practice. The 1994 guideline lacked an energy absorption requirement, but was otherwise identical to the rule. Prior to that, trailers and semi-trailers were regulated by Federal Motor Carrier Safety Regulations (FMSCR), which allowed substantially smaller guards.
Independent researchers worldwide had been testing better prototype guards since the 1970s. Many of these research efforts have concluded that that the current U.S. and European standards are inadequate to the task of protecting smaller cars from rear underride, because they allow too much ground clearance.
In 1971, a study by researchers at the Cornell Aeronautical Laboratory conducted 12 full-scale crash tests and applied mathematical modeling to investigate rear underride and specific guard designs, and found that the maximum allowable underride guard ground clearance was 24 inches, and that even at that clearance, guards will be ineffective for small cars. In 1977, the IIHS launched a program to develop a prototype lightweight and effective guard. Researchers designed and tested two guards and concluded that it was feasible to achieve substantial improvements in underride protection without significant increases in the weight – a huge industry concern.
Several suppliers, including the Budd Company and Quinton Hazell have developed and tested underride guards using rigid and energy absorbing designs that appeared to work effectively without substantially increasing vehicle weight. In 1993, Australian researchers also looked at the possibility of developing an effective lightweight and inexpensive rear barrier and concluded that this could be done.
Brazilian researchers, whose work formed the basis of the Brazilian standard – currently the world’s most rigorous – began publishing the results of their rear underride about a decade ago. A cooperative effort between UNICAMP, General Motors do Brazil and Mercedes-Benz do Brazil, resulted in an articulated guard design, capable of springing back into position after the truck passes over a tall obstacle. This feature allowed the use of a low guard that did not impair truck maneuverability.
In 1998, a former NHTSA safety standards engineer underscored the inadequacy of the U.S. rules in a series of crash tests involving rear underride guards built to reflect the then-newly minted rear impact protection standard. The work by John E. Tomassoni showed that underride guard that minimally complied with the new rule were effective at impact speeds of 30 mph. But in some of the tests, the underride magnitude was such that passenger compartment intrusion occurred. Tomassoni also noted that the underride guard designed to meet the minimum static load requirements “will not provide adequate protection in offset impacts.”
In 2002, Transport Canada conducted a series of tests using a 1998 Ford Windstar, a 1998 Chevrolet Cavalier, and a 1998 Honda Civic, to verify the performance of rear impact guards built under FMVSS 223/224.39 It found that “none of the minimally compliant guards were effective for all three vehicle types tested.”
Quite possibly, the new study reflects a combination of bad research and bad policy.