Understanding Injury Mechanisms from Real World Accidents

Abstract

Significant developments have been made in the understanding of passenger car collisions over the recent decades through research with dummies and from real life accidents. However, there is limited understanding of how crashes occur in real life and further knowledge is required to understand injury risk in real world scenarios. Analysis of real-world crashes increases the ability to obtain such knowledge. This study aims to understand the influence of the intrusion and the OLC on the resulting impact severity independently of each other by quantifying the possible crash severity reduction for the car’s occupant in frontal car crashes. In this project, real world data, containing frontal collisions derived from the National Automotive Sampling System (NASS), was used to estimate how the ΔV, the vehicle body intrusion and the Occupant Load Criterion (OLC) affected the occupant´s injury. Multiple logistic regression analysis was used to evaluate the relationship between injury severity, ΔV and intrusion. A Matlab script that calculates the Occupant Load Criterion, induced by a given crash pulse under the protection of the ideal restraint system was made. The calculation was compared against the ΔV to estimate the relationship between them. A regression analysis of the data provided a statistically significant relationship (95% Confidence interval level) between injury, OLC and delta-V but the intrusion was not significant. The study concluded that the OLC is a stronger predictor for severe injuries where ΔV is a more reliable predictor for minor injuries. The risk of receiving MAIS2+ injuries and MAIS3+ injuries increases with higher ΔV, with a higher pulse and a greater intrusion.