| Recently, Kansas City Star published an article stating that about 1,400 occupants had died in the last six years in frontal crashes in which airbags did not deploy. The FARS data investigation shows an increasing trend of fatal crashes involving airbag non-deployment with higher fatality risk in recent model year vehicles. The number of fatalities in such crashes has increased by 50 percent (from 500 per year to 750 per year) in the last five years. Multiple impacts contribute to about 90 percent of such crashes. Crashes with a curb hit or guardrail impact as the first harmful event and a narrow impact crash with a tree or pole as a subsequent harmful event amount to about 35 to 40 percent of non-deployment related fatal crashes.;The NASS-CDS case analysis revealed that for the un-weighted sample of crashes, 30 percent of the non-deployment may be related to the design of the airbag firing threshold including the sensing of long duration impacts, and about 27 percent may be related to the criteria for resetting of the airbag algorithm in multiple impacts - for example a curb hit followed by pole impact. The asymmetric deployment was also observed quite frequently in NASS cases, and the most probable reason seems to be an electronic or mechanical failure of the airbag system.;Finite element simulations suggested that a slight decrease in velocity of the impact and the amount of vehicle structure overlap with the impacted object significantly changes the pulse which can result in the non-deployment of airbags. Also, the deceleration levels, generated in curb hits and guardrail impacts at 25mph, are sufficient to initiate the airbag algorithm. If the airbag deployment algorithm does not reset in time after a curb or guardrail impact, it may lead to non-deployment in a subsequent crash.;Based on these findings, it is proposed to develop testing techniques to check the airbag algorithms resetting and deployment requirements in multiple impact crashes. Introducing a breakaway pole before the barrier may simulate the effects of pre-impacts like curb hits or guardrail impacts prior to main impact. This test may encourage manufacturers to improve the airbag deployment algorithm which could possibly reduce the problem of non-deployment by about 30 to 35 percent.;Further research should be done to evaluate the need for deployment of airbags in crashes at speeds less than 25mph, reduced overlap impacts, and long duration impacts like front-to-rear collision. Also, the malfunctioning of the electronic systems is the suspected source of asymmetric non-deployments. Further research should investigate the causes of asymmetric non-deployments and determine the need for increased reliability of the system. |
