Search

The National Highway Traffic Safety Administration (NHTSA) recently published a Request for Comments regarding a potential upgrade to the US New Car Assessment Program (US NCAP) - a star-rating program pertaining to vehicle crashworthiness. Therein, NHTSA (a) cited two metrics for assessing head risk: Head Injury Criterion (HIC15) and Brain Injury Criterion (BrIC), and (b) proposed to conduct risk assessment via its risk curves for those metrics, but did not prescribe a specific method for applying them. Recent studies, however, have indicated that the NHTSA risk curves for BrIC significantly overstate field-based head injury rates. Therefore, in the present three-part study, a new set of BrIC-based risk curves was derived, an overarching head risk equation involving risk curves for both BrIC and HIC15 was assessed, and some additional candidatepredictor- variable assessments were conducted. Part 1 pertained to the derivation. Specifically, data were pooled from various sources: Navy volunteers, amateur boxers, professional football players, simple-fall subjects, and racecar drivers. In total, there were 4,501 cases, with brain injury reported in 63. Injury outcomes were approximated on the Abbreviated Injury Scale (AIS). The statistical analysis was conducted subject to ordinal logistic regression analysis (OLR), such that the various levels of brain injury were cast as a function of BrIC. The resulting risk curves, with Goodman Kruksal Gamma=0.83, were significantly different than those from NHTSA. Part 2 pertained to the assessment relative to field data. Two perspectives were considered: "aggregate" (ΔV=0-56 km/h) and "point" (high-speed, regulatory focus). For the aggregate perspective, the new risk curves for BrIC were applied in field models pertaining to belted, mid-size, adult drivers in 11-1 o'clock, full-engagement frontal crashes in the National Automotive Sampling System (NASS, 1993-2014 calendar years). For the point perspective, BrIC data from tests were used. The assessments were conducted for minor, moderate, and serious injury levels for both Newer Vehicles (airbag-fitted) and Older Vehicles (not airbag-fitted). Curve-based injury rates and NASS-based injury rates were compared via average percent difference (AvgPctDiff). The new risk curves demonstrated significantly better fidelity than those from NHTSA. For example, for the aggregate perspective (n=12 assessments), the results were as follows: AvgPctDiff (present risk curves) = +67 versus AvgPctDiff (NHTSA risk curves) = +9378. Part 2 also contained a more comprehensive assessment. Specifically, BrIC-based risk curves were used to estimate brain-related injury probabilities, HIC15-based risk curves from NHTSA were used to estimate bone/other injury probabilities, and the maximum of the two resulting probabilities was used to represent the attendant headinjury probabilities. (Those HIC15-based risk curves yielded AvgPctDiff=+85 for that application.) Subject to the resulting 21 assessments, similar results were observed: AvgPctDiff (present risk curves) = +42 versus AvgPctDiff (NHTSA risk curves) = +5783. Therefore, based on the results from Part 2, if the existing BrIC metric is to be applied by NHTSA in vehicle assessment, we recommend that the corresponding risk curves derived in the present study be considered. Part 3 pertained to the assessment of various other candidate brain-injury metrics. Specifically, Parts 1 and 2 were revisited for HIC15, translation acceleration (TA), rotational acceleration (RA), rotational velocity (RV), and a different rotational brain injury criterion from NHTSA (BRIC). The rank-ordered results for the 21 assessments for each metric were as follows: RA, HIC15, BRIC, TA, BrIC, and RV. Therefore, of the six studied sets of OLR-based risk curves, the set for rotational acceleration demonstrated the best performance relative to NASS.

A provisional, age-dependent thoracic risk equation (or, "risk curve") was derived to estimate moderate-to-fatal injury potential (AIS2+), pertaining to men with responses gaged by the advanced mid-sized male test dummy (THOR50). The derivation involved two distinct data sources: cases from real-world crashes (e.g., the National Automotive Sampling System, NASS) and cases involving post-mortem human subjects (PMHS). The derivation was therefore more comprehensive, as NASS datasets generally skew towards younger occupants, and PMHS datasets generally skew towards older occupants. However, known deficiencies had to be addressed (e.g., the NASS cases had unknown stimuli, and the PMHS tests required transformation of known stimuli into THOR50 stimuli). For the NASS portion of the analysis, chest-injury outcomes for adult male drivers about the size of the THOR50 were collected from real-world, 11-1 o'clock, full-engagement frontal crashes (NASS, 1995-2012 calendar years, 1985-2012 model-year light passenger vehicles). The screening for THOR50-sized men involved application of a set of newly-derived "correction" equations for self-reported height and weight data in NASS. Finally, THOR50 stimuli were estimated via field simulations involving attendant representative restraint systems, and those stimuli were then assigned to corresponding NASS cases (n=508). For the PMHS portion of the analysis, simulation-based closure equations were developed to convert PMHS stimuli into THOR50 stimuli. Specifically, closure equations were derived for the four measurement locations on the THOR50 chest by cross-correlating the results of matched-loading simulations between the test dummy and the age-dependent, Ford Human Body Model. The resulting closure equations demonstrated acceptable fidelity (n=75 matched simulations, R2≥0.99). These equations were applied to the THOR50-sized men in the PMHS dataset (n=20). The NASS and PMHS datasets were combined and subjected to survival analysis with event-frequency weighting and arbitrary censoring. The resulting risk curve--a function of peak THOR50 chest compression and age--demonstrated acceptable fidelity for recovering the AIS2+ chest injury rate of the combined dataset (i.e., IR_dataset=1.97% vs. curve-based IR_dataset=1.98%). Additional sensitivity analyses showed that (a) binary logistic regression yielded a risk curve with nearly-identical fidelity, (b) there was only a slight advantage of combining the small-sample PMHS dataset with the large-sample NASS dataset,, ((c) use of the PMHS-based risk curve for risk estimation of the combined dataset yielded relatively poor performance (194% difference), and (d) when controlling for the type of contact (lab-consistent or not), the resulting risk curves were similar.)

While seat belts are the most effective safety technology in vehicles today, there are continual efforts in the industry to improve their ability to reduce the risk of injury. In this paper, seat belt pretensioners and current trends towards more powerful systems were reviewed and analyzed. These more powerful systems may be, among other things, systems that develop higher belt forces, systems that remove slack from belt webbing at higher retraction speeds, or both. The analysis started with validation of the Ford Human Body Finite Element Model for use in evaluation of abdominal belt loading by pretensioners. The model was then used to show that those studies, done with lap-only belts, can be used to establish injury metrics for tests done with lap-shoulder belts. Then, previously-performed PMHS studies were used to develop AIS 2+ and AIS 3+ injury risk curves for abdominal interaction with seat belts via logistic regression and reliability analysis with interval censoring. Finally, some considerations were developed for a possible laboratory test to evaluate higher-powered pretensioners.

The increasing number of people over 65 years old (YO) is an important research topic in the area of impact biomechanics, and finite element (FE) modeling can provide valuable support for related research. There were three objectives of this study: (1) Estimation of the representative age of the previously-documented Ford Human Body Model (FHBM) -- an FE model which approximates the geometry and mass of a mid-sized male, (2) Development of FE models representing two additional ages, and (3) Validation of the resulting three models to the extent possible with respect to available physical tests. Specifically, the geometry of the model was compared to published data relating rib angles to age, and the mechanical properties of different simulated tissues were compared to a number of published aging functions. The FHBM was determined to represent a 53-59 YO mid-sized male. The aforementioned aging functions were used to develop FE models representing two additional ages: 35 and 75 YO. The rib model was validated against human rib specimens and whole rib tests, under different loading conditions, with and without modeled fracture. In addition, the resulting three age-dependent models were validated by simulating cadaveric tests of blunt and sled impacts. The responses of the models, in general, were within the cadaveric response corridors. When compared to peak responses from individual cadavers similar in size and age to the age-dependent models, some responses were within one standard deviation of the test data. All the other responses, but one, were within two standard deviations.

A set of risk equations was derived to estimate the probability of sustaining a moderate-to-serious injury to the knee-thigh-hip complex (KTH) in a frontal crash. The study consisted of four parts. First, data pertaining to knee-loaded, whole-body, post-mortem human subjects (PMHS) were collected from the literature, and the attendant response data (e.g., axial compressive load applied to the knee) were normalized to those of a mid-sized male. Second, numerous statistical analyses and mathematical constructs were used to derive the set of risk equations for adults of various ages and genders. Third, field data from the National Automotive Sampling System (NASS) were analyzed for subsequent comparison purposes. Fourth, the fidelity of the resulting set of risk equations was assessed by using the risk equations to transform the axial compressive femur loads from simulated, full-engagement, frontal crashes into event risks, and the resulting model-based injury rates were compared with the field-based injury rates. The results were promising: For unbelted drivers in towaway frontal crashes involving 1985-1997 model year passenger cars whose speed changes were less that 58 km/h, the model-based average injury rate was 1.10%; the field-based rate was 1.30%. Moreover, some of the trends in the field were confirmed with the model (e.g., there were more KTH-injured males than KTH-injured females). The risk equations demonstrated better fidelity for lower-speed crashes than high-speed crashes.

A theoretical math model was created to assess the net effect of aging populations versus evolving system designs from the standpoint of thoracic injury potential. The model was used to project the next twenty-five years of thoracic injuries in Canada. The choice of Canada was topical because rulemaking for CMVSS 208 has been proposed recently. The study was limited to properly-belted, front-outboard, adult occupants in 11-1 o'clock frontal crashes. Moreover, only AIS3+ thoracic injury potential was considered. The research consisted of four steps. First, sub-models were developed and integrated. The sub-models were made for numerous real-world effects including population growth, crash involvement, fleet penetration of various systems (via system introduction, vehicle production, and vehicle attrition), and attendant injury risk estimation. Second, existing NASS data were used to estimate the number of AIS3+ chest-injured drivers in Canada in 2001. This served as data for model validation. Third, the projection model was correlated favorably with the 2001 field estimate. Finally, for the scenario that 2004-2030 model-year systems would perform like 2000-2003 model-year systems, a projection was made to estimate the long-term effect of eliminating designs that would not comply with the proposed CMVSS 208. The 2006-2030-projection result for this scenario: 764 occupants would benefit from the proposed regulation. This projection was considered to be conservative because future innovation was not considered, and, to date, the fleet's average chest deflections have been decreasing. The model also predicted that, through 2016, the effect of improving system performance would be more influential than the population-aging effect; thereafter, the population-aging effect would somewhat counteract the effect of improving system performance. This theoretical math model can provide insights for both designers and rule makers.