Search Results

Driver injury probabilities in real-world frontal crashes were statistically modeled to estimate the relative roles of five variables of topical interest. One variable pertained to behavior (belt-wearing rate), one pertained to crash circumstances (speed change), and three pertained to occupant demographics (sex, age, and body mass index). The attendant analysis was composed of two parts: (1) baseline statistical modeling to help recover the past, and (2) sensitivity analyses to help consider the future. In Part 1, risk functions were generated from statistical analysis of real-world data pertaining to 1998-2014 model-year light passenger cars/trucks in 11-1 o’clock, full-engagement frontal crashes documented in the National Automotive Sampling System (NASS, 1997-2014). The selected data yielded a weighted estimate of 1,269,178 crash-involved drivers.

An eleven-group taxonomy was created to classify real-world side crashes from the Crashworthiness Data System (CDS) component of the National Automotive Sampling System (NASS). Three steps were taken to develop the classification scheme: (1) side-impact towaway crashes were identified by examining 1987-2016 model year light passenger vehicles with Collision Deformation Classification (CDC) data from the 1997-2015 calendar years of NASS; (2) case reviews, engineering judgments, and categorization assessments were conducted on these data to produce the eleven-group taxonomy; and (3) taxonomic groups were reviewed relative to regulated crash test procedures. Two of the taxonomic groups were found to have the most frequent crash types, each contributing approximately 22% to the total, followed closely by a third taxonomic group contributing approximately 19%.

The present study had three objectives: (1) define a reasonable number of categories to bin head injuries, (2) develop an overarching risk function to estimate head-injury probability based on injury probabilities pertaining to those subordinate categories, and (3) assess the fidelity of both the overarching function and approximations to it. To achieve these objectives, we used real-world data from the National Automotive Sampling System (NASS), pertaining to adult drivers in full-engagement frontal crashes. To provide practical value, we factored the proposed US New Car Assessment Program (US NCAP) and the corresponding Request for Comments from the government. Finally, the NASS data stratifications included three levels of injury (AIS1+, AIS2+, AIS3+), two levels of restraint (properly-belted, unbelted), and two eras based on driver-airbag fitment (Older Vehicles, Newer Vehicles).

Transfer or response equations are important as they provide relationships between the responses of different surrogates under matched, or nearly identical loading conditions. In the present study, transfer equations for different body regions were developed via mathematical modeling. Specifically, validated finite element models of the age-dependent Ford human body models (FHBM) and the mid-sized male Hybrid III (HIII50) were used to generate a set of matched cases (i.e., 192 frontal sled impact cases involving different restraints, impact speeds, severities, and FHBM age). For each impact, two restraint systems were evaluated: a standard three-point belt with and without a single-stage inflator airbag. Regression analyses were subsequently performed on the resulting FHBM- and HIII50-based responses. This approach was used to develop transfer equations for seven body regions: the head, neck, chest, pelvis, femur, tibia, and foot.

In the present study, various risk curves for moderate-to-fatal head injury (AIS2+) were theoretically assessed by comparing model-based injury rates with field-based injury rates. This was accomplished by applying the risk curves in corresponding field models. The resulting injury rates were considered from two perspectives: aggregate (0-56 kph events) and point-estimate (higher-speed, barrier-like events). Four risk curves were studied: a HIC15-based curve from Mertz et al. (1997), a BRIC-based curve from Takhounts et al. (2011), a BrIC-based curve from Takhounts et al. (2013) and a Concussion-Correlate-based curve from Rowson et al. (2013). The field modeling pertained to adult drivers in 11-1 o'clock, towaway, full-engagement frontal crashes in the National Automotive Sampling System (NASS, calendar years = 1993-2012), and the model-year range of the passenger vehicles was 1985-2010.

Injury distributions of belted drivers in 1998-2013 model-year light passenger cars/trucks in various types of real-world frontal crashes were studied. The basis of the analysis was field data from the National Automotive Sampling System (NASS). The studied variables were injury severity (n=2), occupant body region (n=8), and crash type (n=8). The two levels of injury were moderate-to-fatal (AIS2+) and serious-to-fatal (AIS3+). The eight body regions ranged from head/face to foot/ankle. The eight crash types were based on a previously-published Frontal Impact Taxonomy (FIT). The results of the study provided insights into the field data. For example, for the AIS2+ upper-body-injured drivers, (a) head and chest injury yield similar contributions, and (b) about 60% of all the upper-body injured drivers were from the combination of the Full-Engagement and Offset crashes.

In the present study, transfer equations relating the responses of post-mortem human subjects (PMHS) to the mid-sized male Hybrid III test dummy (HIII50) under matched, or nearly-identical, loading conditions were developed via math modeling. Specifically, validated finite element (FE) models of the Ford Human Body Model (FHBM) and the HIII50 were used to generate sets of matched cases (i.e., 256 frontal impact cases involving different impact speeds, severities, and PMHS age). Regression analyses were subsequently performed on the resulting age-dependent FHBM- and HIII50-based responses. This approach was conducted for five different body regions: head, neck, chest, femur, and tibia. All of the resulting regression equations, correlation coefficients, and response ratios (PHMS relative to HIII50) were consistent with the limited available test-based results.

A three-step process was developed to estimate fracture criteria for a human body model. The process was illustrated via example wherein skull fracture criteria were estimated for the Ford Human Body Model (FHBM)~a finite element model of a mid-sized human male. The studied loading condition was anterior-to-posterior, blunt (circular/planar) cylinder impact to the frontal bone. In Step 1, a conditional reference risk curve was derived via statistical analysis of the tests involving fractures in a recently reported dataset (Cormier et al., 2011a). Therein, Cormier et al., authors reported results for anterior-to-posterior dynamic loading of the frontal bone of rigidly supported heads of male post mortem human subjects, and fracture forces were measured in 22 cases. In Step 2, the FHBM head was used to conduct some underlying model validations relative to the Cormier tests. The model-based Force-at-Peak Stress was found to approximate the test-based Fracture Force.

The objective of this study was to further validate three previously-developed, age-dependent finite element models representing 35, 55, and 75 year old mid-sized males. The validation was based on comparisons with the following published tests involving post mortem human subjects: oblique thoracic and abdominal pendulum impact (4-10 m/s), oblique and lateral thoracic pendulum impact (2.5 m/s), and lateral thoracic pendulum impact (4.3 and 6.7 m/s). The responses of the models were compared to cadaveric response corridors and responses from specific cadavers similar in size and age. When compared to the cadaveric response corridors, the model responses were generally within those corridors. When compared to the responses of specific cadavers, the results were mixed. In some of the cases the model responses predicted the age-dependency of the cadaveric responses. In other cases, the model responses had the opposite trend of those of the cadavers.

Various frontal impact risk curves were assessed for the next-generation USA New Car Assessment Program (NCAP). Specifically, the “NCAP risk curves” — those chosen by the government for the 2011 model year NCAP — as well as other published risk curves were used to estimate theoretically the injury rates of belted drivers in real-world frontal crashes. Two perspectives were considered: (1) a “point” estimate of NCAP-type events from NCAP fleet tests, and (2) an “aggregate” estimate of 0 ≤ ΔV ≤ 56 km/h crashes from a modeled theoretical vehicle whose NCAP performance approximated the average of the studied fleet. Four body regions were considered: head, neck, chest, and knee-thigh-hip complex (KTH). The curve-based injury rates for each body region were compared with those of real-world frontal crashes involving properly-belted adult drivers in airbag-equipped light passenger vehicles. The assessment yielded mixed results.

An eight-group taxonomy was created to classify real-world frontal crashes from the Crashworthiness Data System (CDS) component of the National Automotive Sampling System (NASS). Three steps were taken to develop the taxonomy: (1) frontal-impact towaway crashes were identified by examining 1985-2005 model year light passenger vehicles with Collision Deformation Classification (CDC) data from the 1995-2005 calendar years of NASS; (2) case reviews, engineering judgments, and categorization assessments were conducted on these data to produce the eight-group taxonomy; and (3) two subsets of the NASS dataset were analyzed to assess the consistency of the resulting taxonomic-group frequencies. “Full-engagement” and “Offset” crashes were the most frequent crash types, each contributing approximately 33% to the total. The group identified as “D, Y, Z No-Rail” was the most over-represented crash type for vehicles with at least one seriously-injured occupant.

Lower-body injury data for adults in real-world frontal impacts in the National Automotive Sampling System (NASS) were collected, analyzed, and modeled via statistical methods. Two levels of lower-body injury were considered: maximum serious-to-fatal (MAIS3+) and moderate-to-fatal (MAIS2+). In the analysis, we observed that a substantial fraction of all lower-body injured occupants had no recorded floor/toe pan intrusion: 47% of all MAIS3+ injured occupants; 69% of all MAIS2+ injured occupants. In the statistical modeling, we developed binary logistic regression models to fit the MAIS3+ and MAIS 2+ injury data. The statistically significant variables (p ≤ 0.05) were the speed change of the crash, postcrash floor/toe pan intrusion, level of restraint, occupant age, and occupant gender.

An age-dependent, serious-to-fatal (AIS3+), thoracic risk curve was derived and evaluated for frontal impacts. The study consisted of four parts. In Part 1, two datasets of post mortem human subjects (PMHS) were generated for statistical and sensitivity analyses. In Part 2, logistic regression analyses were conducted. For each dataset, two statistical methods were applied: (1) a conventional maximum likelihood method, and (2) a modified maximum likelihood method. Therefore, four statistical models were derived — one for each dataset/statistical method combination. For all of the resulting statistical models (risk curves), the linear combination of maximum normalized sternum deflection and age of the PMHS was identified as a feasible predictor of AIS3+ thoracic injury probability. In Part 3, the PMHS-based risk curves were transformed into test-dummy-based risk curves. In Part 4, validation studies were conducted for each risk curve.

A theoretical, mathematical, risk assessment procedure was developed to estimate the fraction of drivers that incurred head and thoracic AIS3+ injuries in full-engagement frontal crashes. The estimates were based on numerical simulations of various real-world events, including variations of crash severity, crash speed, level of restraint, and occupant size. The procedure consisted of four steps: (1) conduct the simulations of the numerous events, (2) use biomechanical equations to transform the occupant responses into AIS3+ risks for each event, (3) weight the maximum risk for each event by its real-world event frequency, and (4) sum the weighted risks. To validate the risk assessment procedure, numerous steps were taken. First, a passenger car was identified to represent average field performance.

A new, AIS3+ thoracic risk equation based on chest deflection was derived and assessed for drivers subjected to concentrated (belt-like) loading. The new risk equation was derived from analysis of an existing database of post mortem human subjects in controlled, laboratory sled tests. Binary logistic regression analysis was performed on a subset of the data, namely, 25th-75th percentile men (by weight) from 36-65 years old whose thoracic deformation patterns were due to concentrated (belt-like) loading. Other subsets of data had insufficient size to conduct the analysis. The resulting thoracic risk equation was adjusted to predict the AIS3+ thoracic risks for average-aged occupants in frontal crashes (i.e., 30 years old). Biomechanical scaling was used to derive the corresponding relationships for the small female and large male dummies. The new thoracic risk equations and three other sets of existing equations were evaluated as predictors of real-world crash outcomes.

In the 1999 Supplemental Notice for Proposed Rulemaking (SNPRM) for Advanced Airbags, the National Highway Traffic Safety Administration (NHTSA) sought comments on the maximum speed at which the high-speed, unbelted occupant test suite will be conducted, i.e., 48 kph vs. 40 kph. To help address this question, an analysis of constraints was performed via extensive mathematical modeling of a theoretical restraint system. First, math models (correlated with several existing physical tests) were used to predict the occupant responses associated with 336 different theoretical dual-stage driver airbag designs subjected to six specific Regulated and non-Regulated tests.

A new inflator specification, the “inflator thrust variable,” was developed to better explain measured mid-sized male, instrumented test dummy responses in the chest-on-module test condition. Specifically, controlled laboratory experiments were conducted with non-production, driver airbag modules with inflators of various outputs and gas constituents in an effort to assess their effects on a pertinent occupant response. Regression analyses showed that the inflator thrust variable is a better predictor of the observed variation in peak viscous criterion responses than either peak tank pressure or the related pressure rise rate when inflators of differing gas composition were compared.

Locations of key body segments of Hybrid III dummies used in FMVSS 208 compliance tests and NCAP tests were measured and subjected to statistical analysis. Mean clearance dimensions and their standard deviations for selected body segments of driver and passenger occupants with respect to selected vehicle surfaces were determined for several classes of vehicles. These occupant locations were then investigated for correlation with impact responses measured in crash tests and by using a three dimensional human-dummy mathematical model in comparable settings. Based on these data, the importance of some of the clearance dimensions between the dummy and the vehicle surfaces was determined. The study also compares observed Hybrid III dummy positions within selected vehicles with real world occupant positions reported in published literature.