Trends in aggressivity and driver risk for cars,SUVs, and pickups: Vehicle incompatibility from 1989 to 2016

Abstract

Objective: When 2 vehicles of different sizes collide, the occupants of the smaller vehicle are more likely to be injured than the occupants of the larger vehicle. The larger vehicle is both more protective of its own occupants and more aggressive toward occupants of the other vehicle. However, larger, heavier vehicles tend to be designed in ways that amplify their incompatibility with smaller, lighter vehicles (by having a higher ride height, for example). A 2012 study by the Insurance Institute for Highway Safety (IIHS) concluded that fatalities caused by design incompatibility have decreased in recent years. The current study was conducted to update the 2012 IIHS analysis and to explore trends in vehicle incompatibility over time.

Methods: Analyses examined deaths in crashes involving 1- to 4-year-old passenger vehicles from 1989 to 2016 collected from the Fatality Analysis Reporting System (FARS). Trends in driver risk were examined by comparing driver death rates per million registered vehicle years across vehicle type and size. Trends in aggressivity were examined by comparing partner driver death rates across vehicle type and size.

Results: Cars and SUVs have continued their trend toward reduced incompatibility. In 1989–1992, SUVs were 132% more likely to kill the driver in a partner car compared with when a car crashed with another car. By 2013–2016, this value had dropped to 28%. Pickups and cars remain just as incompatible in 2013–2016 as they were in 1989–1992, however (159% vs. 158%). Remaining pickup incompatibility may be largely due to excess curb weight rather than to shape or design features, because light pickups were just 23% more likely to kill the driver in a partner car compared with when a car crashed with another car.

Conclusions: The trend toward reduced fleet incompatibility has continued in the latest crash data, particularly for cars and SUVs. Although pickup–car incompatibility has also decreased over time, pickups remain disproportionately aggressive toward other vehicles, possibly due to their greater average curb weight. Reducing the weight of some of the heaviest vehicles and making crash avoidance technology fitment more widespread may be promising means to reduce remaining fleet incompatibility. Identifying the source of remaining incompatibility will be important for safety improvements going forward.     

Efficacy of side airbags in reducing driver deaths in driver-side car and SUV collisions

OBJECTIVE: To estimate the efficacy of side airbags in preventing driver deaths in passenger vehicles struck on the driver side. METHODS: Risk ratios for driver deaths per driver-side collision were computed for side airbag-equipped cars and SUVs, relative to vehicles without side airbags. Driver fatality ratios also were calculated for the same vehicles in front and rear impacts, and these were used to adjust the side crash risk ratios for differences in fatality risk unrelated to side airbags. Risk ratios were calculated separately for side airbags providing torso-only protection and side airbags with head protection; almost all head protecting airbags also had airbags protecting the torso. RESULTS: Car driver death risk in driver-side crashes was reduced by 37 percent for head protecting airbags and 26 percent for torso-only side airbags. Car driver death risk was reduced for older and younger drivers, males and females, and drivers of small and midsize cars, and when the striking vehicle was an SUV/pickup or a car/minivan. Death risk for drivers of SUVs was reduced by 52 percent with head protecting side airbags and by 30 percent with torso-only airbags. The effectiveness of side airbags could not be assessed for pickups and minivans due to the small number of these vehicles with airbags involved in crashes. CONCLUSION: Side airbags substantially reduce the risk of car and SUV driver death in driver-side collisions. Making side airbags with head protection available to drivers and right front passengers in all passenger vehicles could reduce the number of fatalities in motor vehicle crashes in the United States by about 2,000 each year.     

The Crash Compatibility of Cars and Light Trucks

This paper investigates the compatibility of cars, light trucks, and vans (LTVs) involved in traffic crashes. An analysis of U.S. crash statistics shows that, although LTVs currently account for approximately one–third of registered U.S. passenger vehicles, collisions between cars and LTVs account for over one–half of all fatalities in light vehicle–to–vehicle crashes. In these crashes, 81 percent of the fatally injured are found to be occupants of the car. These statistics suggest that LTVs and passenger cars are incompatible in traffic crashes, and that LTVs are the more aggressive of the two vehicle classes. The fundamental incompatibility between cars and LTVs is observed even when the analysis is restricted to collisions between vehicles of model year 1990 or later - indicating that, despite the availability of newer safety countermeasures, e.g., airbags, the incompatibility between cars and LTVs will persist in future fleets. Through examination of crash test results, field crash statistics, and vehicle measurements, the paper explores the design imbalances between cars and LTVs, e.g., mass, stiffness, and geometry, which lead to these severe crash incompatibilities.     

Safety belt effectiveness in preventing driver fatalities versus a number of vehicular, accident, roadway, and environmental factors

Safety belt effectiveness in preventing fatalities to drivers is examined versus a number factors (vehicle, accident, and environmental) by applying the double-pair comparison method to appropriate subsets of the Fatal Accident Reporting System (FARS) data. For each of 13 factors studied, safety belt effectiveness (the percent of fatally injured unbelted drivers who would not have been killed if they had been wearing safety belts) is estimated, as is an associated standard error of the estimate. The results, which are presented graphically, provide no evidence that safety belt effectiveness is systematically influenced by most of the factors investigated, including car mass and model year. The absence of any systematic relationship with car mass is in agreement with an earlier finding based on the pedestrian fatality exposure method; this agreement adds plausibility to the assumptions used for both the earlier and the present methods. Safety belt effectiveness is greater for single-car crashes (62 ± 5) % than for crashes involving two cars (30 ± 8) % , this difference being statistically significant at p<.02. The results suggest weakly that safety belt effectiveness is greater for two-door (48 ± 6) % than for four-door (38 ± 10) % cars, and is greater for striking (44 ± 6) % cars than for struck (27 ± 12)% cars. The above differences probably reflect higher effectiveness in frontal (or rollover) crashes than in side impacts.     

Smeed's law and expected road fatality reduction: An assessment of the Italian case

IntroductionSmeed's law defines the functional relationship existing between the fatality rate and the motorization rate. While focusing on the Italian case and based on the Smeed's law, the study assesses the possibility for Italy of reaching the target of halving the number of road fatalities by 2020, in light of the evolving socio-economic situation.MethodA Smeed's model has been calibrated based on the recorded Italian data. The evolution of the two indicators, fatality and motorization rates, has been estimated using the predictions of the main parameters (population, fleet size and fatalities). Those trends have been compared with the natural decreasing trend derived from the Smeed's law.ResultsNine scenarios have been developed showing the relationship between the fatality rate and the motorization rate. In case of a limited increase (logistic regression) of the vehicle fleet and according to the estimated evolution of the population, the path defined by motorization and fatality rate is very steep, diverging from the estimated confidence interval of the Smeed's model. In these scenarios the motorization rate is almost constant during the decade.ConclusionsIn the actual economic context, a limited development of the vehicle fleet is more plausible. In these conditions the target achievement of halving the number of fatalities in Italy may occur only in case of a structural break (i.e., the introduction of highly effective road safety policies).Practical applicationThe proposed tools can be used both to evaluate retrospectively the effectiveness of road safety improvements and to assess if a relevant effort is needed to reach the established road safety targets.     

Estimates of fatal chest and abdominal injury prevention in side-impact crashes

The analysis reported in this study shows that the current NHTSA estimates of potential fatality reductions from improved side-impact protection in passenger cars are unrealistically high. This study is based on recent statistical estimates of fatality prevention by eliminating ejection and mitigating interior impact, and on an analysis of the limits of crash protection in fatal side-impact accidents. Because many fatalities involve high impact speed and significant deformation of vehicle side structures, about 70% of the near-side occupant fatalities from chest and abdominal injury are unpreventable by practical design changes. This implies that 30% of current fatalities may be prevented. Estimates of fatality prevention were then made as a fraction of the effectiveness of airbags in frontal crashes. Assuming sideinterior changes that may produce 20% of frontal airbag effectiveness, which is 6% effectiveness and is roughly the same level of safety benefit achieved with the energy-absorbing steering system, 140 fatalities per year would be prevented in all types of side-impact crashes. This estimate is a realistic but formidable goal. It is in striking contrast with NHTSA projection of 1,185 fatality reductions with relatively straightforward design changes. Their projection exaggerates potential safety improvements.     

Projecting effects of improvements in passive safety of the New Zealand light vehicle fleet

OBJECTIVES: In the year 2000, as part of the process for setting New Zealand road safety targets, a projection was made for a reduction in social cost of 15.5 percent associated with improvements in crashworthiness, which is a measure of the occupant protection of the light passenger vehicle fleet. Since that document was produced, new estimates of crashworthiness have become available, allowing for a more accurate projection. The objective of this paper is to describe a methodology for projecting changes in casualty rates associated with passive safety features and to apply this methodology to produce a new prediction. METHOD: The shape of the age distribution of the New Zealand light passenger vehicle fleet was projected to 2010. Projected improvements in crashworthiness and associated reductions in social cost were also modeled based on historical trends. These projections of changes in the vehicle fleet age distribution and of improvements in crashworthiness together provided a basis for estimating the future performance of the fleet in terms of secondary safety. RESULTS AND CONCLUSIONS: A large social cost reduction of about 22 percent for 2010 compared to the year 2000 was predicted due to the expected huge impact of improvements in passive vehicle features on road trauma in New Zealand. Countries experiencing improvements in their vehicle fleets can also expect significant reductions in road injury compared to a less crashworthy passenger fleet. Such road safety gains can be analyzed using some of the methodology described here.     

Characterization of fatal occupational versus nonoccupational motor vehicle collisions in Kentucky (1998-2000)

Motor vehicle collisions (MVCs) are the leading cause of occupational fatalities in Kentucky as well as in the nation. The characteristics of and contributing factors for occupational versus nonoccupational MVC fatalities in the Commonwealth of Kentucky were examined from 1998 to 2000. Semi trucks were most frequently involved in fatal occupational MVCs, and passenger cars were most frequently involved in nonoccupational MVCs. More than half of the decedent drivers resided outside of Kentucky. The percentage of occupational fatalities occurring on a four-lane highway was double the percentage observed for nonoccupational MVC fatalities. In addition, an increased proportion of occupational MVC deaths occurred on limited access highways compared to nonoccupational fatalities. When human factors contributing to these fatal incidents were examined, the two primary human factors involved in occupational motor vehicle fatalities were driver distraction/inattention and falling asleep, whereas unsafe speed and alcohol were the primary human factors contributing to a nonoccupational fatality. These results suggest that semi drivers traveling on four-lane highways are more at risk for a fatal occupational injury in Kentucky. Therefore, additional epidemiological studies are needed to further examine human factors, the nature of the Kentucky highway system, and trucking controls (e.g., weigh station hours of operation) within the Kentucky transportation industry.     

An application of a conditional logistic regression to study the effects of safety belts, principal impact points, and car weights on drivers' fatalities

In addition to experimental trials in automotive factories, there is a fundamental need to monitor real people involved in real motor vehicle collisions to determine the impact of automotive design characteristics on injury mortality. To aid in designing future safety features for drivers' seats, data from the Fatal Accident Reporting System (FARS) were used to assess the effects of safety belts, directions of crash impacts, age, sex, and car weights on motor vehicle injury fatalities. Furthermore, because the FARS includes only accidents in which there was at least one fatality, this paper introduces a multivariate approach — a logistic regression conditioned on each accident —to avoid the sampling biases inherent in the FARS. The resulting model is used to quantify the relations among the safety belts, the directions of crash impacts, and vehicle weights and their effects on fatalities. Recause the proposed approach allows researchers to study many important variables simultaneously and eliminates the biases resulting from many possible confounders, the estimates presented in this paper are considered to be more precise and more firmly established than earlier estimates. The new approach to analyzing the FARS data will also be useful for investigating the effects of other risk factors or automotive characteristics on crash fatalities.     

Relative Risk of Fatal Injury in “High-Performance-Small-Motorcycle” Crashes in Malaysia

This study presents an analysis of injuries (severe and fatal) sustained by “high-performance-small-motorcycle” users in Malaysia and establishes the relationship between fatality risk and human factors, exposure and vehicle factors resulting from motorcycle crashes. From the data gathered a statistical model based on logistic regression modeling technique has been developed. Five variables found to have a significant influence on fatality risk (p < 0.05) were age, location sites, engine capacity, object/vehicle struck and collision types.     

Situational characteristics of fatal pedestrian accidents involving vehicles traveling at low speeds in Japan

Abstract

Objective: This study aimed to investigate the situational characteristics of fatal pedestrian accidents involving vehicles traveling at low speeds in Japan. We focused on vehicles with 4 or more wheels. Such characteristics included daytime or nighttime conditions, road type, vehicle behaviors preceding the accident, and vehicle impact locations.

Methods: Pedestrian fatality data on vehicle–pedestrian accidents were obtained from the Institute for Traffic Accident Research and Data Analysis of Japan (ITARDA) from 2005 to 2014. Nine vehicle classifications were considered: Trucks with gross vehicle weight (GVW) ≥7.5 tons and <7.5 tons, buses, box vans, minivans, sport utility vehicles (SUVs), sedans, light passenger cars (LPCs), and light cargo vans (LCVs). We compared the situational daytime or nighttime conditions, road type, vehicle behaviors preceding the accident, and vehicle impact locations for accident-involved vehicles traveling at low and higher speeds across all vehicle types.

Results: The results indicate that pedestrian fatalities involving vehicles traveling at low speeds occurred more often under daytime conditions across all vehicle types. At signalized intersections, the relative proportions of pedestrian fatalities were significantly higher when vehicles were traveling at low speed, except when the accidents involved box vans or SUVs. Similarly, when vehicles turned right, the relative proportions of pedestrian fatalities were significantly higher when vehicles traveling at low speed were involved across all vehicle types. In terms of the frontal right vehicle impact location, the relative proportions of pedestrian fatalities were significantly higher when trucks with GVW ≥7.5 tons or <7.5 tons, sedans, or LCVs traveling at low speed were involved.

Conclusions: The situational characteristics of fatal pedestrian accidents involving vehicles traveling at low speeds identified in this study can guide targeted development of new traffic safety regulations or technologies specific to vehicle–pedestrian interactions at low vehicle travel speeds (i.e., driver alert devices or automated emergency braking systems). Ultimately, these developments can improve pedestrian safety by reducing the frequency or severity of vehicle–pedestrian accidents for vehicles turning right at intersections and/or reducing the number of resultant pedestrian fatalities.     

Age,period, and cohort effects in motor vehicle mortality in the United States, 1980–2010: The role of sex,alcohol involvement,and position in vehicle

IntroductionAlthough substantive declines in motor vehicle fatalities in 1980–2010 have been observed, declines by position in the vehicle and alcohol involvement have not been well elucidated.MethodAnalyses of FARS data use the Intrinsic Estimator (IE) to produce estimates of all age, period, and cohort effects simultaneously by position in the car and by alcohol involvement.ResultsDeclines in MVC deaths by position in the car vary for men and women by age and cohort over time. Cohorts born before 1970 had higher risks than those born later. Analyses using proxy indicators of alcohol involvement found the highest risks for those aged 16–24. By period, these risks declined more rapidly than non- alcohol related traffic fatalities.ConclusionChanges in risk patterns are consistent with evidence regarding the contributions of new technologies and public policy efforts to reduce fatalities, but gains have not been shared evenly by sex or position in the car.Practical applicationsGreater attention is needed in reducing deaths among older drivers and pedestrians. Gender differences should be addressed in prevention efforts aimed at reducing MVCs due to alcohol involvement.     

Land-traffic crash leading to passenger vehicle submersion,drowning and other fatal injuries: A 44-year study based on records from the Finnish Crash Data Institute

Background: Land motor traffic crash (LMTC) -related drownings are an overlooked and preventable cause of injury death. The aim of this study was to analyze the profile of water-related LMTCs involving passenger cars and leading to drowning and fatal injuries in Finland, 1972 through 2015. Materials and methods: The database of the Finnish Crash Data Institute (FCDI) that gathers detailed information on fatal traffic accidents provided records on all LMTCs leading to drowning during the study period and, from 2002 to 2015, on all water-related LMTCs, regardless of the cause of death. For each crash, we considered variables on circumstances, vehicle, and fatality profiles. Results: During the study period, the FCDI investigated 225 water-related LMTCs resulting in 285 fatalities. The majority of crashes involved passenger cars (124), and the cause of death was mostly drowning (167). Only 61 (36.5%) fatalities suffered some–generally mild–injuries. The crashes frequently occurred during fall or summer (63.7%), in a river or ditch (60.5%), and resulted in complete vehicle’s submersion (53.7 %). Half of the crashes occurred in adverse weather conditions and in over 40% of the cases, the driver had exceeded the speed limit. Among drivers, 77 (68.8%) tested positive for alcohol (mean BAC 1.8%). Conclusion: Multidisciplinary investigations of LMTCs have a much higher potential than do exclusive police and medico-legal investigations. The risk factors of water-related LMTCs are similar to those of other traffic crashes. However, generally the fatal event in water-related LMTC is not the crash itself, but drowning. The paucity of severe physical injuries suggests that victims’ functional capacity is usually preserved during vehicle submersion. Practical Applications: In water-related LMTCs, expansion of safety measures is warranted from general traffic-injury prevention to prevention of drowning, including development of safety features for submerged vehicles and simple self-rescue protocols to escape from a sinking vehicle.     

The effectiveness of electronic stability control (ESC) in reducing real life crashes and injuries

Electronic Stability Control (ESC) was introduced on the mass market in 1998. Since then, several studies showing the positive effects of ESC have been presented. OBJECTIVE: In this study, data from crashes occurring in Sweden during 1998 to 2004 were used to evaluate the effectiveness of ESC on real life crashes. The effectiveness was analyzed for different road conditions, and some accident types and injury levels. METHODS: The study used statistical analysis. To control for exposure, induced exposure methods were used, where ESC-sensitive to ESC-insensitive crashes and road conditions were matched in relation to cars equipped with and without ESC. Cars of similar or, in some cases, identical make and model were used to isolate the role of ESC. RESULTS: The study shows a positive and consistent effect of ESC overall and in circumstances where the road has low friction. The overall effectiveness on all injury crash types, except rear end crashes, was 16.7 +/- 9.3%, while for serious and fatal crashes; the effectiveness was 21.6 +/- 12.8%. The corresponding estimates for crashes with injured car occupants were 23.0 +/- 9.2% and 26.9 +/- 13.9%.For serious and fatal loss-of-control type crashes on wet roads the effectiveness was 56.2 +/- 23.5% and for roads covered with ice or snow the effectiveness was 49.2 +/- 30.2%. It was estimated that for Sweden, with a total of 500 vehicle related deaths annually, that 80-100 fatalities could be saved annually if all cars had ESC.CONCLUSIONS: ESC was found to reduce crashes with personal injuries, especially serious and fatal injuries. The effectiveness ranged from at least 13% for car occupants in all types of crashes with serious or fatal outcome to a minimum of 35% effectiveness for single/oncoming/overtaking serious and fatal crashes on wet or icy road surface. No difference in deformation pattern was found for cars with or without ESC.     

The fatality consequences of the 65 mph speed limits, 1989

This study examined whether the number of fatalities on rural interstates in 1989 was higher than would be expected based on experience during 1982–1986 and experience on all other roads. Among the 40 states that increased the speed limit to 65 mph on rural interstates, the number of fatalities was 29% higher than expected. Among the eight states retaining a 55 mph maximum speed limit on rural interstates, the observed number of fatalities was 12% lower than expected, although this reduction was not statistically significant. After adjusting the fatality risk on rural interstates for differences in vehicle miles traveled on those roads and for higher passenger vehicle occupancy rates attributable to possible increases in vacation travel, the increased fatality risk was 19%. These data suggest that the majority of the estimated increase in fatalities on rural interstates in 1989 (almost 400 of the approximately 600 extra deaths) can be attributed to the higher speeds resulting from the higher speed limits. Changes in mileage account for the remaining 200 extra deaths.     

基于蒙特卡洛方法的着陆擦机尾风险预测

为解决对航班和机队着陆时擦机尾风险客观预测问题，基于飞行QAR数据和蒙特卡洛模拟方法，建立擦机尾风险预测模型，将某航空公司Boeing737-800机队的380套QAR数据作为样本数据，运用MATLAB编程进行5 000次模拟抽样试验，得到不同机队着陆时俯仰角的分布和擦机尾风险预测曲线。研究结果表明:当着陆俯仰角大于4.5°时，机队2擦机尾风险较大；运用蒙特卡洛模拟预测的飞机着陆俯仰角更为稳定和准确。该模型可进一步软件化，为航空公司的擦机尾超限事件管理提供可靠性指标参考，实现对飞行员操作风险的动态管理。     

Society's aggression level as a predictor of traffic fatality rate

This research investigated the relationship of violence/aggression and other societal variables to traffic accidents. In the first of two studies, multiple regression was applied to 1977 data from each of the 50 states and the District of Columbia. Traffic fatalities per registered vehicle was the dependent variable. The independent variables were homicide rate, suicide rate, fatality rate from other causes, unemployment rate, personal income, density of physicians, alcohol consumption, motor vehicles per capita, road mileage per vehicle, sex and age distribution of drivers, and attained education. The main finding was that the homicide rate (but not the suicide rate) predicted the traffic fatality rate; additional significant predictors were the proportion of young drivers and the fatality rate from non-motor-vehicle accidents. The two primary predictors (homicides and young drivers) accounted for 64 % of the variance of traffic fatalities. In the second study, validation was performed by using the 1977 regression coefficients to estimate 1978 traffic fatalities. The results indicate that when the 1977 regression coefficients were applied to the 1978 values for homicides and young drivers, they accounted for 49 % of the variance of the 1978 traffic fatalities. The findings are discussed in terms of how society's violence/aggression may contribute to traffic accidents.     

Shipyard fatalities in Turkey

The objective of this paper is to investigate causes and incidence rates of 115 fatalities between 2000 and 2010 in Turkish shipyards and to propose mitigation measures. The fatality rate in Turkish shipbuilding industry is unacceptably high, being 3.5 times the average of all industry groups. Classification of fatal occupational accidents revealed five major fatality reasons for the shipbuilding industry; falling from higher elevation to a lower level, exposed to electric shock, fire and/or explosion, being struck by or struck against objects, and caught in between. The highest number of cases of fatal accidents was found among welders, blasters, painters and substructure workers. Statistically analyzed data showed that working on Mondays and Saturdays have special impact on the fatality probability. Also, the number of fatalities is highest in working temperatures above average 25 °C, which covers from June through September. Strategies recommended to minimize the fatalities at shipyards are discussed. Furthermore a risk framework of fatal occupational accidents were drawn so that when working conditions are within the high risk region, extra precautions can be taken in order to minimize the occurrence of accidents and fatalities.     

Estimated benefit of automated emergency braking systems for vehicle–pedestrian crashes in the United States

Abstract

Objective: The objective of this research study is to estimate the benefit to pedestrians if all U.S. cars, light trucks, and vans were equipped with an automated braking system that had pedestrian detection capabilities.

Methods: A theoretical automatic emergency braking (AEB) model was applied to real-world vehicle–pedestrian collisions from the Pedestrian Crash Data Study (PCDS). A series of potential AEB systems were modeled across the spectrum of expected system designs. Both road surface conditions and pedestrian visibility were accounted for in the model. The impact speeds of a vehicle without AEB were compared to the estimated impact speeds of vehicles with a modeled pedestrian detecting AEB system. These impacts speeds were used in conjunction with an injury and fatality model to determine risk of Maximum Abbreviated Injury Scale of 3 or higher (MAIS 3+) injury and fatality.

Results: AEB systems with pedestrian detection capability, across the spectrum of expected design parameters, reduced fatality risk when compared to human drivers. The most beneficial system (time-to-collision [TTC]?=?1.5?s, latency = 0?s) decreased fatality risk in the target population between 84 and 87% and injury risk (MAIS score 3+) between 83 and 87%.

Conclusions: Though not all crashes could be avoided, AEB significantly mitigated risk to pedestrians. The longer the TTC of braking and the shorter the latency value, the higher benefits showed by the AEB system. All AEB models used in this study were estimated to reduce fatalities and injuries and were more effective when combined with driver braking.     

Rage against the machine? Google's self-driving cars versus human drivers

IntroductionAutomated driving represents both challenges and opportunities in highway safety. Google has been developing self-driving cars and testing them under employee supervision on public roads since 2009. These vehicles have been involved in several crashes, and it is of interest how this testing program compares to human drivers in terms of safety.MethodsGoogle car crashes were coded by type and severity based on narratives released by Google. Crash rates per million vehicle miles traveled (VMT) were computed for crashes deemed severe enough to be reportable to police. These were compared with police-reported crash rates for human drivers. Crash types also were compared.ResultsGoogle cars had a much lower rate of police-reportable crashes per million VMT than human drivers in Mountain View, Calif., during 2009–2015 (2.19 vs 6.06), but the difference was not statistically significant. The most common type of collision involving Google cars was when they got rear-ended by another (human-driven) vehicle. Google cars shared responsibility for only one crash.ConclusionsThese results suggest Google self-driving cars, while a test program, are safer than conventional human-driven passenger vehicles; however, currently there is insufficient information to fully examine the extent to which disengagements affected these results.Practical applicationResults suggest that highly-automated vehicles can perform more safely than human drivers in certain conditions, but will continue to be involved in crashes with conventionally-driven vehicles.