Empirical vehicle crashworthiness studies are usually based on national or in-depth traffic accident surveys: Data on accident-involved cars/drivers are analysed in order to quantify the chance of driver injury and to assess certain risk factors like car make and model. As the cars/drivers involved in the same accident form a "cluster", where the size of the cluster equals the number of accident-involved parties, traffic accident survey data are typical multi-level data with accidents as first-level or primary and cars/drivers as secondlevel or secondary units (car occupants in general are to be considered as third level units). Consequently, appropriate statistical multi-level models are to be used for driver injury risk estimation purposes as these models properly account for the cluster structure of traffic accident survey data. In recent years various types of regression models for clustered data have been developed in the statistical sciences. This paper presents multi-level statistical models, which are generally applicable for vehicle crashworthiness assessment in the sense that data on single and multiple car crashes can be analysed simultaneously. As a special case of multi-level modelling driver injury risk estimation based on paired-by-collision car/driver data is considered. It is demonstrated that assessment results may be seriously biased, if the cluster structure inherent in traffic accident survey data is erroneously ignored in the data analysis stage.
Internationally, the need is expressed for harmonized traffic accident data collection (PSN, PENDANT, etc.). Together with this effort of harmonization, traffic accident investigation moves more and more in the direction of accident causation. As current methods only partly address these needs, a new method was set up. The main characteristics of this method are: • Accident/injury causation (associated) factors can objectively be identified and quantified, by comparison with exposure information from a normal population. • All relevant accident and exposure data can be included: human-, vehicle-, and environmental related data for the pre-crash, crash and postcrash situation (the so-called Haddon matrix). The level of detail can be chosen depending on interest and/or budget, which makes the method very flexible. In this paper the accident collection and control group method are presented, including some of the achieved results from a pilot study on 30 truck accidents and 30 control locations. The data were analyzed by using cross-tabulations and classification-tree analysis. The method proved useful for the identification of statistically significant causational aspects.
During the last 5 years, the number of cars fitted with side airbags has dramatically increased. They are now standard equipment, even on many smaller cars or less luxurious vehicles. While some side airbags offer thoracic protection alone, there are those that combine thoracic and head protection (of which most deploy from the seat). Other systems employ separate airbags for head and thorax protection, which are designed to be effective noticeably in a crash against a pole. This paper proposes an evaluation of the effectiveness of side airbags in preventing thoracic injuries to passenger car occupants involved in side crashes. First, the target population (who can take benefit of side airbag deployment and in what circumstances) is defined. Side airbags can be especially effective in cases of impacts on the door with intrusion at a certain impact speed. Then, an example case of a side impact with side airbag deployment is given were side airbag deployment is thought to have had a positive effect on injury outcome. A further case is presented where the impact configuration is likely to have reduced the effect of side airbag deployment on injury outcome. Finally, the estimation of side airbag effectiveness (in terms of additional occupant protection brought exclusively by the airbag) is proposed by comparing injury risk sustained by occupants in (more or less) similar cars (fitted or non fitted with airbags) because, during these years, car structure, and side airbag conception have considerably evolved. In-depth accident data from France, the UK and Germany has been collected. Out of 2,035 side impact accident cases available in the databases, we selected 435 occupants of passenger cars (built from 1998 onwards) involved in an injury accident between year 1998 and year 2004 for EES (Energy Equivalent Speed) values between 20km/h and 50km/h. The occupants, belted or not, were sat on the struck side, whatever the obstacle and type of accidents (intersection, loss of control, etc.). For multiple impact crashes, the side impact is assumed to be the more severe one. Passenger cars were fitted with (96) or without (339) side airbags. Most of the potential risk explanatory variables were correctly and reliably reported in the databases (velocity " impact zone " impact angle " occupant characteristics, etc.). The analysis compared injury risks for different levels of EES and different types of side airbags. A logistic regression model was also computed with injury variables (such as thoracic AIS 2+ or AIS 3+) as the dependant variable and other variables (including airbag type and EES) as explanatory injury risk factors. Results revealed statistically non-significant reductions in thoracic AIS 2+ and AIS 3+ injury risk in side airbag equipped cars in the impact violence range selected (odds ratio between 0.84 and 0.98 depending on types of airbags). The results are discussed. The non-significance is assumed to be due to a low number of cases. Statistical analysis for head injuries was not possible due to the low number of accident cases with passenger cars fitted with head airbags in the databases. Moreover, the discrepancies between the data coming from different countries (especially calculation of EES) might have introduced instability in the analysis.
In Germany, in-depth accident investigations are carried out in the Hannover area since 1973. In 1999 a second region was added with surveys in Dresden and the surrounding area. Internationally, the acronym GIDAS (German In-Depth Accident Study) is commonly used for these surveys. Compared to many other countries, the sample sizes of the GIDAS surveys are much larger. The goal is to collect 1.000 accidents involving personal injuries per year and region. Data collection takes place by using a sampling procedure, which can be interpreted as a two-stage process with time intervals as primary units and accidents as secondary units. An important question is, to what extend these samples are representative for the target population from which they are drawn. Analyses show, for example, that accidents with persons killed or seriously injured are overrepresented in the samples compared to accidents with slightly injured persons. This means, that these data are subject to biases due to uncontrolled variation of sample inclusion probability. Therefore, appropriate weighting and expansion methods have to be applied in order to adjust or correct for these biases. The contribution describes the statistical and methodological principles underlying the GIDAS surveys with respect to sampling procedure, data collection and expansion. In addition, some suggestions regarding potential improvements of study design are made from a methodological point of view.
Annually within the European Union, there are over 50,000 road accident fatalities and 2 million other casualties, of which the majority are either the occupants of cars or other road users in collision with a car. The European Commission now has competency for vehicle-based injury countermeasures through the Whole Vehicle Type Approval system. As a result, the Commission has recognised that casualty reduction strategies must be based on a full understanding of the real-world need under European conditions and that the effectiveness of vehicle countermeasures must be properly evaluated. The PENDANT study commenced in January 2003 in order to explore the possibility of developing a co-ordinated set of targeted, in-depth crash data resources to support European Union vehicle and road safety policy. Three main work activity areas (Work Packages) commenced to provide these resources. This paper describes some of the outcomes of Work Package 2 (WP2, In-depth Crash Investigations and Data Analysis). In WP2, some 1,100 investigations of crashes involving injured car occupants were conducted in eight EU countries to a common protocol based on that developed in the STAIRS programme. This paper describes the purposes, methodology and results of WP2. It is expected that the results will be used as a co-ordinated system to inform European vehicle safety policy in a systematic, integrated manner. Furthermore, the results of the data analyses will be exploited further to provide new directions to develop injury countermeasures and regulations.
Validation of human pedestrian models using laboratory data as well as accident reconstruction
(2007)
Human pedestrian models have been developed and improved continually. This paper shows the latest stage in development and validation of the multibody pedestrian model released with MADYMO. The biofidelity of the multibody pedestrian model has been verified using a range of full pedestrian-vehicle impact tests with a large range in body sizes (16 male, 2 female, standing height 160-192cm, weight 53.5-90kg). The simulation results were objectively correlated to experimental data. Overall, the model predicted the measured response well. In particular the head impact locations were accurately predicted, indicated by global correlation scores over 90%. The correlation score for the bumper forces and accelerations of various body parts was lower (47-64%), which was largely attributed to the limited information available on the vehicle contact characteristics (stiffness, damping, deformation). Also, the effects of the large range in published leg fracture tolerances on the predicted risk to leg fracture by the pedestrian model were evaluated and compared with experimental results. The validated mid-size male model was scaled to a range of body sizes, including children and a female. Typical applications for the pedestrian models are trend studies to evaluate vehicle front ends and accident reconstructions. Results obtained in several studies show that the pedestrian models match pedestrian throw distances and impact locations observed in real accidents. Larger sets of well documented cases can be used to further validate the models especially for specific populations as for instance children. In addition, these cases will be needed to evaluate the injury predictive capability of human models. Ongoing developments include a so-called facet pedestrian model with a more accurate geometry description and a more humanlike spine and neck and a full FE model allowing more detailed injury analysis.
While the number of fatal accidents is diminishing every year, there is still a need of improvement and action to prevent these deaths. Basis for this purpose has to be an analysis about the factors influencing the car crash mortality. There are various studies describing the univariate influence of several factors, but crash scenarios are too complex to be described by a single variable. The multivariate analysis respects the interference of the variables and gets so to more detailed and representative results. This multivariate analysis is based on about 2,600 cases (the data have been collected by the accident research units Hannover and Dresden (during the years 1999-2003). This paper presents a multivariate model (containing ten different variables) which detects 93% of these cases properly. This means it detects the cases as truly survived and truly death.
The need for improved EU level accident information and data was identified in the EU White Paper on Transport Policy (2001)1 and detailed in the Road Safety Action Plan (2003)2. The plan specifies that the EC will develop a road safety observatory to coordinate data collection within an integrated framework.
This study is aimed to investigate the correlations of impact conditions and dynamic responses with the injuries and injury severity of child pedestrians by accident reconstruction. For this purpose, the pedestrian accident cases were selected from Sweden and Germany with detailed information about injuries, accident cars, and accident environment. The selected accident cases were reconstructed using mathematical models of pedestrian and passenger car. The pedestrian models were generated based on the height, weight, and age of the pedestrian involved in accidents. The car models were built up based on the corresponding accident car. The impact speeds in simulations were defined based on the reported data. The calculated physical quantities were analyzed to find the correlation with injury outcomes registered in the accident database. The reconstruction approaches are discussed in terms of data collection, estimating vehicle impact speeds, pedestrian moving speeds and initial posture, secondary ground impact, validity of the mathematical models, as well as impact biomechanics.
The data situation for quantifying the proportion of accidents avoided by the introduction of active safety systems is incomplete, since there is generally no data available on the accidents avoided by the technology in question. In this paper, a split-register approach is suggested and compared with the classical case-control approach known from epidemiologic applications. Provided a set of assumptions hold, which can reasonably be made in such data situations, the split register approach allows inferences on the population accident risk. For both approaches the benefits of basing the analysis on the results of a logistic regression to adjust for confounding factors are outlined. The biasing effects of violating key assumptions are discussed and the split-register approach is demonstrated using the example of the active safety system ESP with data from the German in-depth accident study GIDAS.
Active safety systems are aimed at accident prevention, hence the knowledge required for their development is different from that required for passive safety systems aimed at injury prevention. Particularly, knowledge about accident causation is required. When looking at existing accident causation data, it is argued it fails to explain in sufficient detail how and why the accidents occur. Therefore, there is a need for detailed micro-level descriptions of accident causation mechanisms, and also of methodologies suitable for creating such descriptions. One study addressing these needs is the Swedish project FICA (Factors Influencing the Causation of Accidents and Incidents), where an accident investigation methodology suitable for active safety is developed, and in-depth accident investigations following this methodology are carried out on-scene in the area of Gothenburg by a multidisciplinary team. A preliminary aggregated analysis of different cases shows that the methodology developed is adequate for pointing out common contributing factors and devising principal countermeasures.
Electronic Stability Program (ESP) aims to prevent the lateral instability of a vehicle. Linked to the braking and powertrain systems, it prevents the car from running wide on a corner or the rear from sliding out. It also helps the driver control his trajectory, without replacing him, in the case of loss of control where the driver is performing an emergency manoeuvrer (confused and exaggerated steering wheel actions). A new ESP function optimizes ESP action in curves with hard under steering (situations in which the front wheels lose grip and the vehicle slides towards the outside of the curve). A complementary feature prevents the wheels from spinning when pulling away and accelerating. The name given to the ESP system varies according to the vehicle manufacturer, but other terms include: active stability control (ASC), automotive stability management system (ASMS), dynamic stability control (DSC), vehicle dynamic control (VDC), vehicle stability control (VSC) or electronic stability Control (ESC). This paper proposes an evaluation of the effectiveness of ESP in terms of reduction of injur accidents in France. The method consists of 3 steps: - The identification, in the French National injury accident census (Gendarmerie Nationale only), of accident-involved cars for which the determination of whether or not the car was fitted with ESP is possible. A sample of 1 356 cars involved in injury accidents occurred in 2000, 2001, 2002 and 2003 was then selected. But we had to restrict the analysis to only 588 Renault Lagunas. - The identification of accident situations for which we can determine whether or not ESP is pertinent (for example ESP is pertinent for loss of control accidents whilst it is not for cars pulling out of a junction). - The calculation, via a logistic regression, of the relative risk of being involved in an ESPpertinent accident for ESP equipped cars versus unequipped cars, divided by the relative risk of being involved in a non ESP-pertinent accident for ESP equipped cars versus unequipped cars. This relative risk is assumed to be the best estimator of ESP effectiveness. The arguments for such a method, effectiveness indicator and implicit hypothesis are presented and discussed in the paper. Based on a few assumptions, ESP is proved to be highly effective. Currently, the relative risk of being involved in an ESP pertinent accident for ESP-equipped cars is lower (-44%, although not statistically significant)rnthan for other cars.rn
The so-called "seat-belt injuries" or "seat-belt syndromes", described as 2-point seat-belt injuries, contain heavy inflection injuries of the lumbal spinal column, combined with heavy abdominal injuries as rupture of the upper intestinal bold or heavy injuries of the upper entrails. With "playing" children in the font of the car, with inappropriate plant of 3-point belts, identical injuries can occur.
Rollover scenarios in Europe
(2005)
Rollover accidents seem to be a rising problem in Europe and therefore the systematic of this accident scenario should be investigated. Based on statistical investigations on major European accident databases for different countries a series of 73 real world rollover accidents was analysed. These cases were reconstructed using PC-Crash and preliminary categorised using a modified USbased rollover classification. In a first step, the rollover events were reconstructed from the point of conflict to the vehicle- rest position. The vehicles kinematics as well as its linear and rotational velocities were derived. In a second step typical velocity characteristics as well as kinematics were identified and the events categorised according to these criteria. Based on these results four main categories were defined, covering all reconstructed accidents. This categorisation was based on mechanical parameters (rotatory and translator kinematical data of the vehicle). Significant differences can be seen for different scenarios for the "first phase of rollover".
This paper describes the methodology of In-Depth Investigation in Germany on the example of GIDAS (German In-Depth Accident Study). Since 1999 in Germany a joint project between FAT (Forschungsvereinigung Automobiltechnik or Automotive Industry Research Association) and BASt (Bundesanstalt für Straßenwesen or the Federal Road Research Institute) is being carried out in Hannover and Dresden. The methodology of this project is based on a statistically orientated procedure of data sampling (sampling plan, weighting factors). The paper describes the possibilities of such in-depth investigation on the results of the offered title. The accident cases were collected randomly within GIDAS at Hannover. There are more cases existing from previous investigation started in 1985 under the same methodology. The portion of rollovers can be established at 3.7% of all accidents with casualties in the year 2000. For the study 434 cases of car accidents with rollovers are used for a detail comprehensive analysis. The accidents happened in the years 1994 to 2000 in the Hannover area. The injury distribution will report about 741 occupants with rollover accident event. The presented paper will give an overview of the accident situations following in rollover movements of cars. The distributions of injury frequencies, injury severity AIS for the whole body and for the body regions of occupants will be presented and compared to technical details like the impact speed and the deformation pattern. The speed of the car was determined at the point of rollover and on the point of accident initiency. The characteristics of the kinematics followed in a rollover movement are analyzed and the major defined types of rollover will be shown in the paper. The paper will describe the possibilities of In-Depth Investigation methods for the approach of finding countermeasures on the example of car accidents with rollover and explaining the biomechanics of injuries in rollover movements.
This paper set out to examine the possibilities for injury avoidance implications for older drivers in crashes, based on crash and injury patterns among older drivers and current trends in ageing in most western societies. A number of safety technologies were identified and discussed which have potential for improving vehicle older driver crash avoidance and crashworthiness. While there were some promising estimates available of the likely benefits of this technology for improving safety, it is evident that they need to be confirmed for older drivers, given their age-related disabilities and sensory limitations. Further research is urgently required to ensure that these technologies yield safety benefits without any disbenefits for older drivers.rn
Nowadays airbags are part of the standard equipment in almost all new cars. While airbags are saving an increasing number of people from severe injuries and death in moderate and high speed crashes, they do not completely prevent dashboard injuries. The most common mechanism in dashboard injuries is a posteriorly directed force to the proximal tibia with the knee flexed. This may occur during a motor vehicle frontal impact accident when a knee of the driver or the front-seat passenger strikes the dashboard. The posterior force can be combined with a abducting or rotational force leading to concomitant lateral or posterolateral injury. Car and airbag manufacturers therefore develop special inflatable systems to reduce the impact force in dashboard injuries. Every new inflatable system, however, has to be evaluated in out of position situations in which the system might cause injuries to certain body areas. Therefore, we investigated a new kneebag system in different critical seating positions of post mortem test subjects (PMTS). The tested knee airbag module is a folded airbag (18 litre volume) which is installed below the lower section of the instrument panel of a passenger car. Using four PMTS (2 male, 2 female, age 36"67) the following positions were tested: normal seating position, knee flexed >90 degrees and knee flexed <60 degrees in static deployment tests with direct contact. In addition a dynamic test (48.8kph, AAMA-pulse) was carried out with the PMTS belted in a normal seating position. The inflation phase and the impact of the system on the knee/lower leg were analysed by high speed videos. After the test the lower legs of the PMTS were examined by Xray and autopsy. All soft tissue injuries and bone fractures were recorded. All the tests could be evaluated. Except some superficial skin lesions in the impact area no fracture of the bones around the knee and no knee ligament and tendon injuries were observed. Neither video analysis nor autopsy of the PMTS showed any critical contact injuries caused by the inflation process of the bag. Therefore, it can be concluded that in the tested seating positions which are the most critical for the knee area the knee bag system is safe.
This paper reviews briefly the evolution of the investigation of transport accidents from the early beginnings when individual events were studied but systematic data was not collected. In the transport modes other than on the roads, accident investigation early on, even of single events, was important in introducing safety improvements. Road accidents, however, evolved enormously with the growth of car ownership without any comparable political response to the consequent deaths and injuries, equivalent to what happened with the other modes. From the 1950s data bases started to contribute to our knowledge of the epidemiology of road traffic injuries, and in-depth sample studies have contributed much to the body of knowledge in the last 30 years. However, even the basic input and output variables of a crash, its severity and the seriousness of the outcomes in terms of injuries and their consequences are not complete or agreed upon. Issues of experimental design and sampling are discussed. It is proposed that the most important area for current research to address is the effect of population variations on injury outcomes. The need for the establishment of good data bases for active safety issues is emphasised with the consequent need for better links between the research community and the police.
Traffic accidents were ranked the third among the major causes of death in Thailand. About 13,438 deaths and the death rate from traffic accident was 21.5 per 100,000 of population in 2002. The deaths and death rate varied upon the economic situation. After the economic crisis, traffic accidents were increased as well as the period of the bubble economy. In the Central region of Thailand numbers of road traffic crashes were lower than Bangkok Metropolis, but the highest in the number of deaths, death rate and serious injuries in 2002. Men aged 15"29 years old had higher numbers of deaths than men in other age groups and higher than women. Deaths and injuries from road traffic crashes were the highest in April and January, because there was a long weekend in those months. About 80 percent of road traffic crashes were caused by private car and motorcycle. In 2000 about 51 percent of traffic accidents took place on the straight way, followed by the junction and curves. In 2002, about 97 percent of road traffic crashes were caused by human factors including improper passing, speeding and disregarding to traffic signal, however, the identification of causes of traffic accident needed to improve. Drunk driving, disregarding on safety equipment usage, inefficiency of law enforcement and discontinuing of road safety programs were the deepest causes of traffic accidents. Research based information, a broad coalition of stakeholder and urban planning policy were needed to incorporate for a comprehensive road safety policy formulation and actions.
Pelvic fracture, cracking or breaking of a portion of the pelvis are extremely common injuries in the side impact collisions of motor vehicles. Due to both its shape and structural architecture, mechanics of the pelvic bone is complicated. There is a lack of knowledge regarding the dynamic behavior of the pelvis and its biomechanical tolerance under impact environment. Hence this study is aimed at the understanding of the mechanical response of the human pelvis with three-dimensional finite element (FE) models, under side impact load, applied through a structure, equivalent to a car door. The door structure was modeled, considering few layers, consisting of foam (Styrodur®, 3035 CS), plastic (UHMWPE), steel, glass and steel, putting them in series. A soft tissue layer (equivalent to fat) was also considered on the greater trochanter location. These FE models (with and without the car door structure) were analyzed with ANSYS-LS-DYNA-® dynamic finite element software to compare the effect of the car door padding system for shock absorption. It was observed that with proper combination of shock absorbing material (foam, etc.) and its thickness, the transmission of impact load to the body part (pelvis, etc.) from the outer surface of the car door could be reduced.