Publication

Frontal crashes account for around 50% of all fatalities in passenger cars and thorax injuries are common in this type of crash. To evaluate car crash performance, mechanical representations of human occupants are used, which are called Anthropomorphic Test Devices. The most advanced for frontal crashes is the Test device for Human Occupant Restraint (THOR), which has multiple deflection measurement locations in the chest. This study used a generic Finite Element (FE) car interior model, that was based on several different cars, and was meant to represent the cars present in a selection of NASS/CDS accident cases. Several different injury criteria were used to evaluate the risk of sustaining a three or more severe injury on the Abbreviated Injury Scale (AIS3+), four or more Number of Fractured Ribs (NFR4+) or NFR3+ level of thorax injury. A population based stochastic simulation study was set up and 1000 simulations with parameterized and morphed car interior models were run in the explicit FE solver LS-DYNA, and parameters were sampled using the optimization software LS-OPT. Using logistic regression and a generalized linear model, risk curves were generated for several rib fracture risk criteria, based on the simulations results. These risk curves were then compared to the AIS3+ rib fracture risk from the NASS/CDS field data. From the analysis of the results, it was apparent that the rib fracture risk criterion TIC NSFR fit the NASS/CDS risk curve the best. The rib fracture risk criteria based on Rmax and Dmax overpredicted the injury risk considerably, with a Winsmash ∆v of 56 km/h and 49 km/h for 50% risk of AIS3+ rib fracture for Rmax and Dmax, respectively, while the NASS/CDS risk curve predicted a risk of 50% at 99 km/h. The rib fracture risk criterion T ICNSF R had the 50% risk at a Winsmash ∆v of 95 km/h, which was indeed closer to the NASS/CDS risk curve. However, since TIC NSFR seemed to be too insensitive, with few risk predictions above 20%, it is not recommended as the best available rib fracture risk function to be used with the THOR ATD based on comparison to the risk of rib fracture in the selection of NASS/CDS data. The injury criteria TIC NFR and PCscore showed better sensitivity, while performing well compared to the selected NASS/CDS data. PCscore is slightly preferred, for it is defined for the same injury classification as used in NASS/CDS, AIS3+. PCscore is therefore recommended as the best risk function, among the ones considered, for evaluating the risk of rib fracture with the THOR-50M ATD