Symposiums & Conferences

Publications / Symposiums & Conferences

Potential Effects of Automatic Braking on Accident Fatalities and Serious Injuries, SAE 2017, Detroit, Michigan, April 4-6, 2017.

Abstract: Automatic Emergency Braking will become a standard feature in light duty vehicles beginning in 2023 due to a voluntary agreement between vehicle manufacturers, NHTSA, and IIHS. The agreed performance criteria will result in a system that reduces the incidence of low-speed crashes and will likely have little effect on severe injuries and fatalities. Opportunities for fatality reduction associated with automatic braking are significant and are based on implementation approaches. Potential fatality reductions resulting from automatic braking activation thresholds in various crash modes and closing speed ranges were considered. The effects of alternative performance requirements on potential fatality reductions were then examined.

Methodology Developed for Dynamic Rollover Regulation and Ratings, ICRASH 2014, Sarawak, Malaysia, August 25-28, 2014.

Abstract: At present, most governments require vehicle compliance to a quasi-static Roof Crush Resistance test similar to the Federal Motor Vehicle Safety Standard (FMVSS) 216. In this paper Center for Injury Research (CfIR) proposes a global full-scale dynamic rollover compliance test and rollover rating system. Compliance in these tests is a function of both vehicle structural and dummy responses. The methods include: 1) design and build a rollover test fixture, 2) apply structural injury risk statistics, 3) evaluate the relative structural injury risk performance of production vehicles 4) modify the Hybrid III dummy to be more humanlike, 5) develop momentum exchange dummy head, neck and spine injury measures and criteria, 6) demonstrate the match between structural injury risk relative to criteria and dummy injury measures relative to criteria, 7) identify a real-world test protocol, 8) demonstrate the real-world injury risk rating system and 9) summarize the proposed rollover compliance test regulation.


Integrating OEM Vehicle ROPS to Improve Rollover Injury Probability, ICRASH 2014, Sarawak, Malaysia, August 25-28, 2014.

Abstract: Geometry is a subset of styling and almost all cars have the same generic roof geometry. Studies have shown, however, that geometry plays a big role in the depths of roof crush for a given rollover impact scenario. However, consumers are misled by a rollover propensity rating on safety charts with dynamic frontal, front offset, side and rear crashworthiness ratings. Rollover crashworthiness has and continues to have the lowest priority in sales potential with consumers although 3% of accidents result in 30% of fatalities. Off-road fleet customers of work vehicles with high rollover fatality rates see the situation differently and equip their vehicles with aftermarket rollover occupant protection systems (ROPS). This paper compares the styling and rollover performance of the original designs, the same vehicles with a patented external ROPS (HALOTM) and as modified with an OEM version of the same patent.


Reliability and Relevance of Rollover Occupant Injury Potential Tests, ICRASH 2014, Sarawak, Malaysia, August 25-28, 2014.

Abstract: From as early as 1965, US auto manufacturers have been researching means to reduce rollover injuries and fatalities. Dynamic rulemaking in 1970 was rejected by industry and in 1973 the Department of Transportation issued a static roof crush test rule proposed by the industry. That rule stood for 35 years until 2009 when the strength requirement was doubled and research was initiated into dynamic rollover means of testing. This paper reviews the literature, examines and compares the frequency and injury potential of rollover accident modes. Two modes are represented in test methodology by the lateral roll over (corresponding to trip-overs) and the ramp (or corkscrew) roll over (corresponding to flip-overs). The energy and impact orientation of the typical ramp rollover tests are dramatically more violent and not comparable to the results of lateral tests conducted with a real-world protocol representing 95% of serious injury rollovers.


Vehicle Roof Structure Design Can Significantly Reduce Occupant Injury, ICRASH 2014, Sarawak, Malaysia, August 25-28, 2014.

Abstract: Vehicle design has been driven by sales and marketing factors for many years, with a few exceptions like the brutish looking Hummer Sport Utility Vehicle (SUV) which was marketed for its structural safety. While it’s partially true for public customers, consumers of work trucks commercial operations like in the oil, gas and mining industry are a large fleet consumer for dual cab transportations and an even larger segment for simple single cab work trucks. These vehicles are designed to meet the minimum specifications for safety equipment and structural tolerances and therefore can be inadequate in various crash modes. Most manufacturers focus on sales in the U.S. where regulations are design determinants, but outside the U.S., there are some times little or no minimum safety design regulations. The Mitsubishi Triton and the Toyota Hilux 4 and 2 door pickup vehicles, all of which are not sold in the U.S., require structural reinforcement for safe operation in these non-ordinary environments. This paper focuses on the design and performance of these work trucks and the means by which rollover safety can be measured and significantly improved.


Electronic Crash and Injury Causation Analyses, AAFS 2014, Seattle, Washington, February 17-22, 2014.

Hypotheses: Any crash is related to a combination of roadway factors, vehicle defects, and/or driver errors. The ever-increasing electronic control of vehicles has its benefits and limitations. Modern vehicle drive-by-wire (DBW) systems can have lifesaving outcomes because they can identify and take action (e.g., deploy airbag) based on potentially-dangerous roadway factors and vehicle and/or driver errors. Most of the time, the DBW systems save lives. However, some DBW commands can be fatal (e.g., misinterpreted data causing a crash or preventing airbag deployment causing injury). A system analysis that utilizes the stored vehicle control module data, in conjunction with physical evidence and CDR download, enables the forensic engineer to quantitatively apportion causation and provide proof of roadway factors, vehicle defect, and/or driver error.


Correlating Human and Flexible Dummy Head-Neck Injury Performance, ESV Conference, Seoul, S. Korea, May 27-30, 2013.

Abstract: The Center for Injury Research (CfIR) has developed methods to derive and correlate rollover dummy head-neck injury with NASS/CIREN data. In this paper, these methods are applied to other accident modes. Specifically, we investigated the application of the dummy rollover head-neck modifications, as well as structural injury risk, IARV, momentum exchange injury measures and criteria to frontal, offset and small overlap frontal and side impact testing.

Recently, NHTSA has implemented a comprehensive series of component regulations (FMVSS 126, FMVSS 216, FMVSS 226) [1-3] which, in combination, are intended to drastically reduce the number of crashes and occupant injury and fatalities in rollovers and other modes. However, the stiffness of the dummy neck and the disparity between IARV and momentum exchange injury measures were not addressed. We opine that injury and fatality rates are high because of poor dummy-to-human stiffness and substantially underestimated IARV injury criteria compared to consensus momentum exchange injury measures.

IIHS 40% offset and small overlap frontal and side impact tests were studied to observe the trajectory of the Hybrid III dummy head with production neck and evaluate injury measures. Then, the effect of substituting the production neck with the more flexible rollover neck was investigated. Estimates were made of the dummy head excursion, proximity of the head to vehicle structures at maximum excursion, the likelihood and severity of vehicle structure contact, and injury measures.

Results indicate that, while the flexible neck in a rollover increases head excursion by 3 inches when contacted at 7 mph, the frontal and side impact tests described here result in head contact with vehicle structures and exceed the rollover-developed AIS ≥3 momentum exchange injury criteria of the integrated bending moment (IBM) and single and double integration product of head resultant acceleration (HRA).


Predicting a Vehicle's Dynamic Rollover Injury Potential from Static Measurements, ESV Conference, Seoul, S. Korea, May 27-30, 2013.

Abstract: The purpose of this research was to demonstrate a methodology for deriving a real world dynamic rollover injury potential rating system from static measurements. The methodology consists of an evaluation of vehicle strength to weight ratio (SWR), roof structure elasticity from static testing, major radius, minor radius, and major radius extension to predict residual roof crush. In addition to providing a hypothesis for evaluating the vehicles the major radius extension (MRE) will be looked at to provide insight for correcting existing anomalous static SWR measurements. These parameters are important because a 43 nation Global NCAP has been established to rate vehicles in all crash modes. Rollover performance is to be rated by SWR. Global NCAP will be responsible for reducing the 1,200,000 vehicle fatalities per year of which 25% can be rollovers when comparing rollover fatality proportionality to U.S vehicle fatality statistics.

Based on our rollover research of the past 12 years structural and occupant protection countermeasures can be used to significantly counter those fatalities. Disseminating the dynamic injury performance provides a world-wide opportunity to save many tens of thousands of lives annually. Jordan Rollover System (JRS) vehicle rollover dynamic testing apparatus has identified a significant number of vehicles which meet the most rigorous static roof strength criteria, but fail to provide occupant protection from injury risk.

Manufacturers can reduce the injury risk within size class by minimizing geometry effects and the likelihood of a high pitch rollover. While large, tall, heavy vehicles are protective in frontal and side impact accidents they are very high injury risk vehicles in rollovers for the very same reasons. This paper provides a prediction method for assessing dynamic injury probability from static test data and measurements.


Electronic Crash, Defect and Causation Analyses, ESV Conference, Seoul, S. Korea, May 27-30, 2013.

Abstract: Since about 1990, all motor vehicles are equipped with Collision Data Recorders (CDR). These devices initially provided impact and status data, as well as deployment commands for occupant protection systems. More recently, vehicles are equipped with drive-by-wire systems with electronically-aided driver controls derived from more than 40 control modules interconnected by communication networks. A vast amount of additional data is collected and stored by these control modules. Diagnostic Trouble Codes (DTCs) identify, describe and store events, faults, limitations exceeded and corrective actions made by each control module. The functioning of a control module, access, and storage location codes are defined in its Product Definition Description (PDD) manual.

Several case studies are presented to demonstrate the effects of control module algorithms, events, faults and actions. A complete case study identified a defect and proved that defect was the proximate cause of the injury and death. Surprisingly, these modules can seize control of a drive-by-wire vehicle and actually cause loss of control, resulting in a crash and injury ranging in severity from minor to fatal.


The Dynamic Rollover Protection (DROP) Research Program, ICRASH International Crashworthiness Conference, Milan, Italy, July 18-20, 2012

Abstract: The Dynamic Rollover Protection (DROP) research program is attempting to establish which combination of crash severity, roll kinematics, biomechanical injury criteria, crash test dummy, and restraint systems, address the major proportion of fatalities and serious injuries occurring to seat belted and restrained occupants involved in rollover crashes. The outcomes of this three to four year research program will be an understanding of those factors most important for regulators, industry and consumer groups to consider when developing a dynamic rollover crashworthiness compliance or consumer rating crash test protocol. The research program and progress on the sub tasks from the DROP program are presented. In particular, investigations of how head, chest or thorax fatal injuries are to be replicated for a typical rollover crash, are outlined. The advanced UNSW version of the Jordan Rollover System, recently installed at Sydney’s Roads and Maritime Services Crashlab test facility, is also described in the paper.


Jordan Rollover System Test Results, ICRASH International Crashworthiness Conference, Milan, Italy, July 18-20, 2012.

Abstract: Initial static measurements identified vehicle parameters affecting roof crush performance such as pitch and lateral loading. The JRS fixture itself was designed to replicate an on-the-road rollover, one roll at a time in a laboratory setting with a small footprint. Since 2004, more than 50 dummy-occupied vehicles have been tested dynamically with the JRS. Up to 50 data channels were collected and examined as possible metrics. These included vehicle structural, geometric, dummy injury measures and kinematic data. The selection of parameters as possible test criteria, independently or in combination, was based upon results of dynamic tests by CfIR and other laboratories, case studies, and real-world crash databases.

This study examined 1.) Vehicle structural and geometric measures and 2.) Dummy injury measures. The degree of residual roof crush was selected as the vehicle structural injury risk measure. The Integrated Bending Moment (IBM) and criteria were selected as the dummy neck injury measure. Research on other body part injury measures is in process. A major observation was that the severity and pattern of injury is dependent on the variable and unknown musculature tension and flexibility of the human spinal column during the rollover and the dummy’s ability to replicate it in dynamic tests.


Is BFD a Hyperflexion Injury or Compression with Localized Bending Injury or Both?, ICRASH International Crashworthiness Conference, Milan, Italy, July 18-20, 2012.

Abstract: Few biomechanical engineers have had the opportunity to study post-mortem human subjects or anthropomorphic test dummies in an instrumented, controlled rollover test environment. It is no surprise that there is a lack of consensus with respect to human cervical spine injury mechanisms in rollovers and head-neck alignment with respect to the roof intrusion vector. In the 1980’s and 1990’s, this lead author conducted experiments on fresh cadaveric head-neck specimens that produced bilateral facet dislocation (BFD) injuries with rotation constraint. Pintar et al. produced BFD with hyperflexion. Nightingale et al. produced BFD with compression, causing column buckling and localized lower neck flexion. The authors of this paper opine that mechanical determinants dictate the injury patterns when the neck is overloaded and fails. BFD failure occurs by 3 known mechanisms: hyperflexion; compression with rotation constraints; or compression with higherorder buckling (i.e., localized bending). The validity of the hyperflexion mechanism does not preclude the validity of compression mechanisms, or vice versa. It is possible for the roof intrusion force vector to be aligned with the head, neck, and spine. However, because the varying vehicle yaw and pitch upon ground-roof contact, it is more likely that the preponderance of the catastrophic rollover neck injuries are bending injuries.


Design, Development and Validation of Rollover Dummy Injury Measures, ICRASH International Crashworthiness Conference, Milan, Italy, July 18-20, 2012.

Abstract: Existing and proposed U.S. rollover standards test only vehicle structural roof strength; there is no U.S. regulatory requirement based on dummy neck injury criteria for rollover occupant protection. The global New Car Assessment Program (NCAP) has adopted or is adopting a static roof strength test protocol similar to the U.S. Federal Motor Vehicle Safety Standard (FMVSS) 216 to certify and rate vehicles to strength-to-weight ratios (SWRs) varying from country to country between SWR= 2.5 to SWR=4.0. These vehicle tests and measurements were seemingly developed to predict the likelihood of inadequate dynamic neck injury protection performance. There now exists an extensive body of literature that lend credence to the claim that static roof strength tests are unreliable ways of ensuring safe vehicle designs. The vehicle and human structures in dynamic circumstances are nonlinear. Dummy neck injury measures best evaluate vehicle, geometry and occupant protection device effectiveness.


The Development of a Dynamic Rollover Rating Test, ESV Conference, Washington, DC, June 13-16, 2011.

Abstract: The goal of this research is to develop a dynamic rollover test rating system similar to the star-rating system of frontal Federal Motor Vehicle Safety Standard (FMVSS) 208 and side FMVSS 214 compliance, New Car Assessment Program (NCAP) and Insurance Institute for Highway Safety (IIHS) tests. Until now, the requirement for vehicle and occupant crashworthiness in rollovers has been a structural measure only, the vehicle’s strength-to-weight ratio (SWR), in a static roof crush test.

The short-term objective of this paper is to develop a quasi-dynamic rating system based on predictions derived from the Jordan Rollover System (JRS) dynamic rollover tests, IIHS static tests and finite element parameter sensitivity studies, verified by dynamic test sampling. The rating for the protocol is based on the National Accident Sampling System (NASS) and Crash Injury Research Engineering Network (CIREN) injury risk probability functions.

One method of predicting performance is to adjust the results of a dynamically-tested vehicle, similar to the vehicle whose performance is to be predicted, by the parameter sensitivity relationships correlated to a larger number of dynamically-tested vehicles. Another method is to formulate and then apply a multivariate equation based on the correlated parameters of a larger number of dynamically-tested vehicles.

This paper presents the prediction procedure based on a limited number of vehicles with a wide range of SWRs. The intent is to apply the procedure to vehicles compliant with 2009 FMVSS 216 and, as such, the illustrations herein are examples. In this paper, the procedure is illustrated by a calculation of two parameters, SWR and major radius (MR). Normalization procedures have also been developed to estimate real-world dynamic test protocol performance, as well as the injury measures for 5th, 50th and 95th percentile dummies. This prediction procedure is an interim solution, not a substitute, for compliance or NCAP dynamic rollover testing.

A more detailed summary of the research basis for this effort is in a companion paper 11-0090 “Predicting and Verifying Dynamic Rollover Occupant Protection.”


Predicting and Verifying Dynamic Occupant Protection, ESV Conference, Washington, DC, June 13-16, 2011.

Abstract: The objective of this paper is to describe the developments that provide the basis for predicting new car occupant protection in real-world rollovers.

An analytical technique has been developed for predicting a vehicle’s dynamic occupant protection performance at any severity from a Jordan Rollover System (JRS) 50-vehicle rollover test database; static test roof strength, stiffness and elasticity data; inertial-influenced impact pitch orientation; size, roll moment and geometry dimensions; and occupant protection features. Only sampling, updating and verification of the JRS database will be necessary to reflect innovative construction and protection techniques until dynamic testing is implemented.

A noteworthy finding of this study was that reducing a vehicle’s major radius (i.e., its shape at the windshield) was more effective in reducing rollover deaths and injuries than increasing roof strength-to-weight ratio (SWR) above 3.0.


Matched Pair Testing of Injury Potential in Repeatable Rollover Tests with the CRIS and JRS, IMECE Conference, Vancouver, BC, November 12-18, 2010.

Abstract: The availability of repeatable dynamic rollover fixtures, like the Controlled Rollover Impact System (CRIS) and Jordan Rollover System (JRS), has changed the face of rollover structural and occupant protection development and evaluation. Tests performed with these devices have demonstrated scientific principles of occupant protection and injury potential which were previously resolvable only by expert rhetoric. Matched-pair experiments with instrumentation measuring dynamic roof crush and dummy injury metrics are now possible. The effectiveness of occupant protection features such as padding, window curtain airbags, belt pretensioners and headrests are qualitatively and quantitatively measureable. The sensitivity of rollover parameters themselves and their effect on injury potential can be determined by tests with different roll rates, pitch angles, impact angles and drop heights. Simulating injury potential to humans with ultimately biofidelic dummy musculature can also be demonstrated. This paper presents two matched pair test sets performed on the CRIS and two matched pair test sets performed on the JRS. The matched pair test sets performed on the CRIS compare the dummy injury measures in reinforced and production versions of the 1998 Ford Crown Victoria and the 1996 Chevrolet Blazer. The CRIS test of the matched pair Crown Victoria vehicles has been presented previously in a paper by Moffatt et al [1]. The matched pair tests that were performed on the JRS were conducted to study the effect of a reinforced roof on dummy injury measures. These tests, performed on production and reinforced versions of the 1998 Ford Explorer and the 1999 Hyundai Sonata, included the measurements of road loads, roof crush and crush speed, dummy upper and lower neck loads, belt loads, as well as the movement of the vehicle during the test.


The Effect of Static Roof Crush Tests Relative to Real World Rollover Injury Potential, IMECE Conference, Vancouver, BC, November 12-18, 2010.

Abstract: Rollover crashworthiness for passenger vehicles is currently evaluated by the Federal Motor Vehicle Safety Standard (FMVSS) 216 static roof strength compliance test. However, research clearly shows that the static test is inadequate in evaluating a vehicle's injury potential performance in a real-world rollover event. Studies previously conducted by the Insurance Institute for Highway Safety (IIHS) show a general relationship between a vehicle's Strength-to-Weight-Ratio (SWR) and its real world injury potential. Although this general relationship is fairly accurate for most vehicles, there are many individual vehicle anomalies. The real world injury performance of the vehicles which make up these anomalies depends much less on the static roof strength (as measured in a FMVSS 216 test) and more on the dynamic performance of the roof and occupant protection systems during a real world rollover (as simulated on the Jordan Rollover System [JRS]). Repeatable dynamic crash tests are used by IIHS, National Highway Traffic Safety Administration (NHTSA), and the New Car Assessment Program (NCAP) to evaluate the performance of a vehicle in every major crash mode except rollovers. Dynamic tests represent the real world effect of vehicle dynamics, orientation, geometry, roof strength, occupant position and kinematics, restraint and other safety system effectiveness while directly measuring comparative dummy injury criteria. Because National Accident Sampling System (NASS) investigations can only measure the cumulative effect of post crash roof crush, NHTSA has established an empirical relationship that a vehicle with post crash negative headroom (PCNH) is five times more likely to injure the occupant. However, data indicates that the anomalies in head, neck, and spinal cord injury are related to the momentum exchange of dynamic head impact speed and the duration of neck loading in each roll, not the cumulative amount of residual roof crush. This paper suggests a means of comparatively evaluating a vehicle's dynamic rollover occupant injury potential performance.


Commercial, Police, and Military Vehicle Rollover Protection and Evaluating the Effectiveness of Geometry and Retrofit Rollover Testing, ICRASH International Crashworthiness Conference, Washington, DC, September 22-24, 2010.

Abstract: Rollover crashes cause more than 10,000 fatalities and nearly 30,000 serious injuries per year in the U.S. alone. This is due to the fact that the vast majority of vehicles, including commercial, police, and military, lack the roof strength to preserve occupant survival space and protect their occupants in a rollover. Recent statistical and epidemiological studies have shown a significant relationship between roof crush and injury. This rollover occupant protection problem is well known to industries with large vehicle fleets; until now, this problem has eluded solution. Within these various industries a wide variety of rollover occupant protection systems (ROPS), both internal and external, have been designed, purchased, manufactured, installed, and maintained locally with little expert consultation. A wide variety of designs have emerged with an alarming variance in "assumed" crashworthiness. Couple this alarming trend with the risk of rendering the existing occupant protection features (e.g., airbags) ineffective, which has resulted in vehicles with inadequate crashworthiness. This paper describes how rollover damage to a vehicle with a weak roof and the resulting reduction of occupant headroom can beminimized to an inconsequential amount using an innovative externally retrofitted rollover load distribution device. This system was based on an understanding of road crash data, empirical evidence, and innovative state of the art testing and analysis to provide effective external ROPS structures for the commercial, police and military fleets.


The Development of IARV’s for the Hybrid III Neck Modified for Dynamic Rollover Crash Testing, ICRASH International Crashworthiness Conference, Washington, DC, September 22-24, 2010.

Abstract:In the U.S., more than 27,000 catastrophic and fatal injuries occur annually in rollover crashes. This study is part of an ongoing research program aimed at mitigating these injuries. Recent papers introduced a prototype “soft” low-durometer Hybrid III neck design and presented results of matched-pair tests, comparing production and prototype Hybrid III neck responses. This paper
• discusses neck injury criteria, and
• proposes preliminary rollover injury criteria for the prototype “soft” low-durometer neck.
Then, peak neck injury measures are used to calibrate the dynamic relationship for flexion bending between the tensed production Hybrid III neck IARV and:
• the prototype “soft” low-durometer neck, which is representative of 1/3 tensed, and
• the untensed cadaveric logistic regression curves for major flexion injury risk by Pintar et al.


Status of Comparative Dynamic Rollover Compliance Research and Testing, ICRASH International Crashworthiness Conference, Washington, DC, September 22-24, 2010.

Abstract:In the U.S., about 40,000 catastrophic and fatal injuries occur annually in rollover crashes. A strategy for injury mitigation is dynamic compliance testing with dummy-occupied vehicles and occupant protection requirements, similar to that required for frontal and side impacts. Presently, the CRIS and JRS dynamic vehicle rollover test devices realistically simulate the ballistic phase of real-world rollover crashes. A search for a typical serious injury test protocol is in progress.

Over 300 rolls and more than 50 two-roll JRS tests of mostly low severity ballistic trajectory protocols have been performed with the belted production Hybrid III dummy. These dynamic tests (as compared to static tests) identified significant roof strength, construction and crush effects of vehicle geometry, buckling structure, yaw and pitch impact angle, window breakage and their relationship to occupant injury and protection. A companion paper at this conference “Characterizing the Injury Potential of a Real World Rollover” details these effects.

It was also found that with roof crush, serious neck bending injuries predominated while head injuries and partial ejections did not occur, except with the very weakest roofs. Since the bending stiffness of the joint muscles of the human body during a rollover are unknowable, a Hybrid III dummy with a modified lumbar joint and reduced musculature neck has been developed as the best available surrogate for dynamic rollover tests. The Hybrid III neck is about 3 times stiffer in bending than a normal relaxed human neck and about one third as stiff as the tensed neck of a young soldier on which the production Hybrid III neck was based. Recent results with a yaw and trip derived initial out-of-position of this dummy indicated a 8 to 11 kph (5 to 7 mph) centrifugal erection rate. In combination with roof intrusion speeds of 11 to 21 kph (7 to 13 mph) more head injuries and partial ejections, consistent with crash statistics are predicted.

The JRS roof crush results of 40 production vehicles roughly normalized to a proposed real world severity test protocol has been matched to NHTSA’s post crash negative headroom criteria and to a CDC serious injury risk to various body part analysis. A dynamic rollover crashworthiness compliance test based on a roof crush injury risk criteria with reported injury measures from an instrumented, belted, initially out-of-position dummy is available now.

In 2010, with resolving epidemiology and protocol parameter sensitivity data, our goal is a representative injury risk compliance pilot test series with occupant protection injury measure data demonstrated in 2011 and confirmed for an NPRM by VRTC in 2012. We firmly believe that a dynamic test will be ready for implementation long before the NHTSA plan.


Characterizing the Injury Potential of a Real World Rollover, ICRASH International Crashworthiness Conference, Washington, DC, September 22-24, 2010.

Abstract:There are approximately 270,000 rollover crashes annually in the U.S., causing about 10,000 deaths and 30,000 serious injuries. The objective of a 5-year multivariate NHTSA project is to define the global issue: to characterize a real-world rollover. CfIR seeks, more specifically, to identify the rollover segment with the greatest serious injury potential for FMVSS 216 compliant vehicles that would be consistent with a compliance or comparative evaluation dynamic rollover test. This process requires evaluating the injury potential sensitivity of each segment and its influence on the following segment.

Ten segments of a 2-roll event were considered, because it has been shown that 95% of single vehicle rollovers and serious-to-fatal injuries occur within 8 quarter turns. A description of the preliminary segment-by-segment evaluation and the sensitivity to injury potential is characterized by analysis, experiments and illustrations. Parameters were derived and validated with JRS dynamic rollover tests. The test parameters were then applied and normalized to approximately 40 other JRS tests for a comparative pilot injury risk evaluation to the NHTSA post-crash negative headroom criteria.

Since many of the JRS tests included dummies, injury performance was also evaluated based on dummy injury measures from head and neck data collected during the tests. Tests were conducted with production and prototype Hybrid III necks and lumbar spines representing tensed and partially-relaxed human musculature. The results were also compared and correlated with Injury Assessment Reference Values (IARV), consensus impact speed injury criteria and dummy positioning. The results indicate that while the injury risk evaluation generally supports static compliance test criteria, dynamic tests identify vehicle geometry, structural design deficiencies and dummy injury measure results that roughly account for the substantial variation in injury rate identified by IIHS from the SWR static test norm. Examples of the data for some of these “anomalies” and failures are provided.


Testing of the Prototype Low-Durometer Hybrid III Neck for Improved Biofidelity", ASME Summer Bioengineering Conference, Naples, FL, June 15-19, 2010.

Abstract:This study is part of an ongoing research project aimed at mitigating catastrophic human neck injuries, predominantly due to neck bending, in rollover crashes. Presently, the Hybrid III dummy is considered to be the best available human surrogate for dynamic rollover tests. However, there are known biofidelity and instrumentation limitations associated with its use to predict catastrophic neck injuries in real-world rollover crashes.

A previous study investigated the use of the non-biofidelic Hybrid III dummy in a dynamic rollover test to accurately predict the predominant human neck bending injury sustained in real-world rollover crashes. An empirical relationship between upper and lower Hybrid III neck loading was derived. The effects of neck preflexion angle, roof impact speed, roof crush, onset-to-peak neck axial forces and moments, and impact duration on neck bending injury were identified. Peak neck injury measures were rejected.

For this study, the 67-durometer Hybrid III production neck was fabricated with more compliant 35-durometer butyl rubber in order improve the dummy biofidelity in rollover tests. The tests in the previous study were repeated. Correlations were established between the prototype and production necks. Parametric studies of the prototype neck revealed similar trends as observed with the Hybrid III production neck.


An Improved Dummy Neck Assembly for Dynamic Rollover Testing , ASME Summer Bioengineering Conference, Naples, FL, June 15-19, 2010.

Abstract:In the U.S., more than 27,000 catastrophic and fatal injuries occur annually in rollover crashes. Due to the incidence and severity of injuries in rollover crashes, a strategy for injury mitigation is dynamic compliance testing with dummy-occupied vehicles and occupant protection requirements, similar to that required for frontal and side impacts. Presently, there are dynamic vehicle rollover test devices like the Controlled Rollover Impact System and the Jordan Rollover System that realistically recreate real-world rollover crash scenarios. However, the Hybrid III dummy, which is considered to be the best available human surrogate for dynamic rollover tests, has a very stiff neck with limited biofidelity in rollover crashes; the Hybrid III neck is much stiffer than the human neck. Catastrophic human head or neck injuries resulting from roof interaction and partial ejection in real-world rollover crashes are poorly replicated by dynamic rollover tests with the non-biofidelic Hybrid III dummy neck. Only with a more biofidelic dummy can effective testing result in injury mitigation in rollover crashes.

This study is part of an ongoing research project aimed at mitigating catastrophic human neck injuries in real-world rollover crashes. The goal was to develop a biofidelic neck assembly for the Hybrid III dummy in rollover crash environments. The design goals of this prototype neck included decreased stiffness and a mechanism that represents the unknowable human muscle tension in rollover crash environments.

This paper and its companion paper in this conference introduce the new neck design, present results of matched-pair tests that compare the responses of the new neck with the production Hybrid III neck, and propose preliminary rollover injury criteria for this neck. The neck demonstrates repeatability, improved biofidelity, which results in more realistic occupant kinematics, dynamics, injury prediction, and evaluation of various countermeasures.


A Proposed Rollover and Comprehensive Rating System, Technical Conference on the Enhanced Safety Vehicle (ESV), Stuttgart, Germany, June 15-18, 2009.

Abstract:The US, European and Australian New Car Assessment Program (NCAP) and the Insurance Institute for Highway

Safety (IIHS) produce ratings of new vehicle performance based on dynamic crash tests in frontal, side and rear crashes; and vehicle handling tests. No dynamic based crashworthiness ratings exist to date in relation to rollover crashes. This study fills that gap and proposes a rating system for new vehicle performance in rollover crashes. Combined with existing rating systems, consumers will then have a complete and balanced picture of occupant protection performance.

A database of more than 40 Jordan Rollover System (JRS) dynamic rollover tests, assessing injury potential by roof crush and crush speed has generically validated NHTSA and IIHS statistical data as a function of FMVSS 216 quasi-static, strength to weight ratio (SWR). There is however a wide disparity between the performance of individual vehicles at the same or similar SWR between the IIHS statistical and JRS dynamic test data. That disparity has been partially investigated in a companion paper in this conference (Vehicle Roof Geometry and its Effect on Rollover Roof Performance.

IIHS data indicated a 50% reduction in incapacitating and fatal injury risk with a fleet average SWR = 4. However, the use of a SWR-based rollover criterion does not provide sufficient crashworthiness fidelity essential for consumers, nor does such a criterion provide industry the opportunity to design cost-efficient rollover crashworthy vehicles based on occupant injury performance. Only a dynamic rollover testing protocol based on injury criteria would provide this information.


The Minicars RSV – Still a Car for the Future, Technical Conference on the Enhanced Safety Vehicle (ESV), Stuttgart, Germany, June 15-18, 2009.

Abstract:Nearly a half century ago, the General Motors Research Laboratories, developed the high performance Electrovair, with an induction motor drive and solid state controller; the Lunar Rover, GM’s Mark on the Moon; passive occupant protection; separation cruise control; optical lane following; and an electrochemical rechargeable Lithium Iodine engine.

In 1968, a new company called Minicars grew out of this earlier work. This group developed prototype electric, gas and hybrid electric powered versions of a small car for the U.S. government. In 1970, Minicars was a subcontractor to AMF for the development of its Experimental Safety Vehicle.

The Minicars’ Research Safety Vehicle (RSV) was conceived in 1975 as a 1985 prototype. It was to be an S3E vehicle: Safe, Environmental, Efficient and Economical. It was built with foam filled, thin wall sheet metal sections and a polyurethane skin. This car passively protected occupants in 80 kph (50 mph) full frontal, 129 kph (80 mph) half car offset frontal, 64 kph (40 mph) angled side, rear and 48 kph (30 mph) rollover dynamic tests. An electronic version incorporated antilock brakes, radar separation cruise control, and emergency braking when a crash was unavoidable. A production version was to weigh 2,200 pounds, carry four people, and get 32 mpg. It also had 16 kph (10 mph) frontal and rear no damage bumpers and 80 km (50 mile) run flat tires.

Only years later have advanced air bags – as featured in the RSV – become standard in all light vehicles. In the decades since the ESV program and dynamic regulatory testing began, National Highway Traffic Safety Administration (NHTSA) now estimates that airbags save 2,500 lives annually, but we still lose about 12,000 people in frontal, 9,000 in side and over 10,000 in rollover crashes. We can do better by simply looking back to what the RSV program achieved.


Vehicle Roof Geometry and its Effect on Rollover Roof Performance, Technical Conference on the Enhanced Safety Vehicle (ESV), Stuttgart, Germany, June 15-18, 2009.

Abstract:The Jordan Rollover System (JRS) provides a realistic, highly controlled, repeatable dynamic test of vehicle roof crush performance under typical rollover conditions. The principal use thus far has been in comparing vehicles’ roof crush and injury potential performance in one and two roll events. Because the JRS directly measures the force between the roof and the ground during touchdown, it can be used to measure, assess and optimize occupant protection by adjusting roof geometry, roof structural design and material strength and elasticity, for the least cost and weight.

This study demonstrates that the peak force (load) between the initial leading side roof rail (near side) and the road is roughly four times the vehicle weight (the load-to-weight ratio or LWR) when a vehicle first touches down at around 150º of roll. The force then drops substantially as the vehicle continues to roll over the flat of the roof, in most instances dropping to zero because the vehicle is momentarily airborne. When the vehicle rolls beyond 180º and comes into contact with the side rail opposite to the leading side of roll (far side), the force rapidly rises again. The roof then either collapses or lifts the vehicle center of gravity (COG). The far side rail of a weak roof vehicle that cannot lift the COG may then halt the vehicle’s downward fall, imposing even larger forces on the road segment when the vehicle’s door and main body structure interact with the roadway. To deal with such forces, a long standing and natural presumption has been to substantially increase the roof strength to weight ratio (SWR), which can result in weight efficiency cost penalties. However, one production vehicle that was tested minimized roof crush without substantially increasing its SWR.

Analysis of the results has found that far side roof crush is strongly related to the difference between the major radius (the maximum distance from the principal axis of rotation to the roof rail) and minor radius (distance from that axis to the center of the roof). Three to four inches, as between cars and LTV’s has a significant effect on injury potential. The typical difference in a light truck vehicle LTV is around 15 cm to 25 cm (6” to 10”) while in an passenger car it is around 8 cm to 15 cm (3” to 6”).

These observations were confirmed by physical tests of strong and weak roofed vehicles. These tests led to the conclusion that a geometry change in the roof to minimize the difference in radius across the roof would reduce the degree to which the far side of a less strong roof had to lift the vehicle as it rolled beyond 180º. A finite element analysis confirmed that for a vehicle of modest roof strength, a structurally strong, rounded roof panel will reduce the far side deformation and intrusion speed by about two-thirds without increasing underlying roof strength. These results were confirmed in JRS testing of current production passenger cars and SUV vehicles and with a “HALO” TM – High Attenuation Load Offset (U.S. and International Patent Pending Rollover Damage Minimization Device) retrofit kit for SUVs.


Hybrid III Correlation with Human Injury Potential in Rollovers, ASME Summer Bioengineering Conference, Lake Tahoe, California, June 17-21, 2009.

Abstract:In the U.S., more than 27,000 catastrophic and fatal injuries occur annually in rollovers. This study is part of an ongoing research project aimed at mitigating the likelihood and severity of such injuries.

Last year, the authors developed a dynamic rollover test methodology for replicating, predicting, and differentiating between types of real-world neck injuries using the non-biofidelic Hybrid III dummy as the human surrogate. Based on platen drop and pendulum test results, dummy positioning for flexion injury was determined and peak neck injury measures were rejected. A new neck injury criteria, the integrated bending moment (IBM), was proposed that related human neck flexion injury to Hybrid III lower neck moment-time histories. The IBM was validated by dynamic rollover tests performed with the Jordan Rollover System (JRS) with roofs of different strength-to-weight ratios (SWR's). The measured lateral and flexion neck moments were then roughly correlated with human neck flexion injury measures proposed by Pintar, et al., in 1998.

To date, real-world head injuries resulting from roof interaction and partial ejection could not be replicated in dynamic rollover tests with the non-biofidelic Hybrid III dummy because of stiffness differences between the Hybrid III and human neck.

Findings thus far suggest that improved correlations with human injury measures could be achieved with the development of a dummy neck that is biofidelic in the rollover crash mode. In this paper, the production Hybrid III neck was modified with lower durometer butyl rubber discs and nodding blocks for improved biofidelity. Pendulum tests were repeated to correlate the production and modified Hybrid III neck responses. JRS tests were performed with an increased far-side impact angle to evaluate the capability of the modified preflexed neck to replicate and predict head, neck, thoracic spine, and ejection injury potential in real-world rollovers.

Results of this study indicate the following:

Matched-pair platen pendulum tests with the low-durometer neck were found to be repeatable.

In JRS tests performed with an increased far-side impact angle, the low-durometer preflexed Hybrid III neck reasonably replicated head, neck, thoracic spine, and ejection injuries and kinematics of weak-roofed vehicles.

The low-durometer neck allowed a more direct correlation with human neck flexion injury measures.


Repeatability of a Dynamic Rollover Test System, ICRASH International Crashworthiness Conference, Kyoto, Japan 2008

Abstract:Rollover accidents have the highest serious to fatal injury rates of any accident mode. Research and development on rollover occupant protection has been frustrated by the lack of a low cost, controlled, repeatable, dynamic test. The most widely used tests, dolly and CRIS system rollovers, do not meet all of these conditions, but the Jordan Rollover System (JRS) does. This study demonstrates JRS repeatability using three identical production vehicles with anthropomorphic test dummies. The first test of each vehicle used string potentiometers to measure roof performance. The second used both string potentiometers and an instrumented test dummy. The JRS test parameters, roof structural performance, and Hybrid III dummy injury measures were all shown to be highly repeatable with variation generally not more than 10 percent. The dummy and vehicle repeatability was on par with the repeatability shown in similar crash test studies conducted by IIHS and NHTSA.


People Safe in Rollovers Foundation, Emergency World Summit, Washington DC July 18-20, 2007

Abstract: World renowned expert engineers will meet to debate the cause of injury in a rollover (Diving vs. Roof Intrusion) and to expose the weak U.S. government standard that has led to more than three decades of unnecessary fatalities and catastrophic injuries in rollover accidents. Although the experts supporting the “Diving Theory” have declined to participate, the debate will go on with their testimony under oath on this issue. Other topics of discussion will be Ejection, Severity, Testing, Regulation, Injury Measures, Disabled Living & Spinal Cord Injuries, Societal Costs, Defense and Plaintiff Strategies, Public Information, NCAP, and International Cooperation. Engineers who have studied the issue of Roof Crush in depth will present scientific papers and videos of roof drop test comparisons and vehicle rollover comparisons of strong roofs vs. weak roofs. Survivors of Roof Crush will tell their stories.


Results from Two Sided Quasi-Static (m216) and Repeatable Dynamic Rollover Tests (JRS) Relative to FMVSS 216 Tests, Lyon, France, June 18-21, 2007

Abstract:In an attempt to find a test protocol that characterizes the rollover occupant protection capability of a passenger vehicle better than the test used in Federal Motor Vehicle Safety Standard 216, we developed equipment and protocols for a modified, quasi-static roof crush test (M216, a test conducted sequentially on both sides of the roof over the A pillars at a pitch angle of 10º and roll angles of 25° and 40° respectively) and for a repeatable, dynamic rollover test called the Jordan Rollover System (JRS).

We have conducted M216 and JRS tests on 17 production vehicles to determine roof crush and crush velocities at a number of points in the interior. These tests included complete production vehicles, body bucks at reduced weight to increase the effective roof strength-to-weight ratio, and pairs of identical vehicles where one has had the roof reinforced in a manner that is entirely hidden by the vehicle’s sheet metal and upholstery. Data from the JRS tests and the M216 tests are compared with the results of FMVSS 216 tests.

Analyses of the data highlight the relative value and validity of each test methodology, its ability to predict roof performance in actual rollovers, its use in vehicle roof structure design, and its potential contribution to regulation or consumer information. Based on the roof crush and crush speed in the vicinity of front seat occupants’ heads, we propose a rollover crashworthiness ranking system. While static tests measure the force and deformation of the roof on the outside, the dynamic tests measure the crush on the inside during the sequence of rollover roof impacts, where it is directly related to the occupant’s survival space and injury potential.


Observations from Repeatable Dynamic Rollover Tests, International Crashworthiness Conference (ICRASH), Athens, Greece, July 4-7, 2006

Abstract:In an attempt to understand the relationship between quasi-static and dynamic test results, repeatable, dynamic rollover tests were conducted on production vehicles to determine intrusion and intrusion velocities using the Jordan Rollover System (JRS). These tests included complete production vehicles and body bucks at reduced weight, to vary the roof strength-to-weight ratio. Data from these tests are compared with the results of quasi-static roof strength tests measured at greater roll and pitch angles than are used in FMVSS 216. Biomechanical data indicates that serious head, face, neck or thoracic spine injury are a consequence of rapid impacts with significant amplitude. The test data suggests a correlation between quasi-static roof strength and dynamic roof intrusion velocity. Localized failures (buckling and collapse of structural elements that often translate into the roof panel) are a more critical aspect of roof performance than its strength as measured in FMVSS 216.


Reducing Rollover Occupant Injuries: How and How Soon, Technical Conference on the Enhanced Safety of Vehicles (ESV), Washington D.C. June 6-9, 2005

Abstract: Public release of previously confidential Malibu test data and film [1] provides the basis for this review. These are sixteen well-instrumented, definitive 32 mph dolly rollover tests of production Chevrolet Malibu sedans with unbelted Hybrid III dummies and eight with belted dummies (half of the cars in each group had roll cages to simulate strong roofs). This paper analyzes and reinterprets this material to resolve the principal motivating research question: does a strong roof reduce the potential for rollover head and neck injuries? Our findings are: (1) a rolling vehicle’s center of gravity rises and falls only about 10 cm during a rollover so that its vertical velocity at roof impact is never more than 2.5 m/sec; (2) the six dummies showing the highest head and neck forces were all seated on the far side of Malibus without roll cages; (3) these high head and neck loads occurred after onset of roof intrusion from rapid roof collapse and buckling, not from occupant diving; (4) average roof impact neck forces measured by near side dummies and by far side dummies seated under roofs that did not contact the ground all averaged 3,300 to 3,600 N, and none was sufficient to cause serious injury; (5) the unrestrained Hybrid III dummy drop tests showed that neck loads of 7,000 N correspond to a 2.4 m/sec roof intrusion velocity while 3,500 N neck loads corresponds to a 1.1 m/sec intrusion velocity; (6) the windshields of the production vehicles broke early leaving weakened roof structures that deformed back and forth with subsequent roof impacts; and (7) the tempered side glazing of production Malibus broke far more frequently than in rollcaged vehicles facilitating partial or complete ejection. The Malibu tests provide considerable insight into the potential countermeasures that could reduce rollover injuries.