Improvement of the Child Injury Risk Curves for Road Accidents

Abstract

Research objective: the EC CASPER project aimed at decreasing injuries and fatalities of child travelling in vehicles. One of the ways to achieve this is to obtain knowledge on how injuries occur—identifying the severity and how they can be correlated with the data collected from reconstruction tests. This information leads to improve the injury risk curves by adding new points that were introduced and developed in previous investigation projects. The main idea of these curves is to correlate dummy readings from accident reconstructions with the injury levels recorded in the real cases. These injury risk curves will improve the way children are restraint in cars. Methodology: the first step to obtain new data for the Injury Risk Curves required collecting information from real accidents. This accident data was collected by road accident investigation teams. Then, some accidents were selected following a certain criteria for reconstructions in crash-test laboratories. The accident reconstructions were carefully prepared, taking into account the real life travelling situation. Additionally, virtual tests were carried out to ensure the similarity between the EES, kinetics and kinematics of the accident and the proposed test configuration. Once the test results were validated by analysing the deformation of the vehicles and, a correlation is made between the injury/non-injury of the child and the dummy readings, focussed on five body segments. Finally, using the Weibull cumulative distribution of density for the injuries and the confidence interval associated with this, constitute the injury risk curves for each body segment. Results: the accident database that was developed in previous projects has now information from 806 cases (1300 restrained children) as a result of the efforts of the CASPER accident investigation teams. It is very important to highlight that the database is not representative of the overall European child car passenger population as a selection bias was used to collect the data (a bias for severe injuries and high crash severity was used). CDC descriptions and deformation points of the real accidents and the reconstructions were compared in order to validate the accident reconstructions. The real injuries collected from the accidents were analysed before the reconstructions and compared with the dummies’ readings after them. The data collected by performing accident reconstructions was introduced in a reconstruction database. Then, the proposed new points were added to the injury risk curves for their improvement. Limitations of this study: one of the limitations of this study is the high number of testing variables that can affect the results. For example, unexpected post-crash kinematics of the vehicles and the sensitivity of the impact point. Also, even though the accidents are investigated in-depth, information on the real travelling conditions of the occupants is not always present (e.g. CRS misuse, out of position). These conditions can affect the correlation between real injuries and dummy readings. What does the paper offer that is new in the field in comparison to other works of the author: unfortunately, even though the non-orthogonally type of collision carried out in a linear dragging crash ground laboratory was a new experience, this procedure was recently presented in another conference. Conclusion: the CASPER project took results from previous projects and continued the research to improve the Injury Risk Curves for children in cars. This was done by providing additional points, obtained by accident reconstructions, especially in the ranges that were considered weak in terms of representation. Accidents were collected by investigation teams. Then, some of them were reconstructed in crash test laboratories. Once the reconstruction was validated, the dummies’ readings were compared with the injuries from the real accident. These correlations were analysed and new points for the Injury Risk Curves were established. For more information public results have been posted on www.casper-project.eu.

F2012-F02-023