Sonstige
Filtern
Erscheinungsjahr
Dokumenttyp
- Konferenzveröffentlichung (62)
- Buch (Monographie) (43)
- Arbeitspapier (1)
Schlagworte
- Statistik (106) (entfernen)
Institut
Estimation of the benefits for the UK for potential options to modify UNECE Regulation No. 95
(2010)
The side impact problem in Europe remains substantial. UK data shows that between 22% and 26% of car occupant casualties are involved in a side impact, but this rises to between 29% and 38% for those who are fatally injured. This indicates the more injurious nature of side impacts compared with frontal impacts. The European Enhanced Vehicle safety Committee (EEVC) has performed work to address the side impact issue since 1979. As part of its continuing work, it has recently investigated potential options for regulatory changes to improve side impact protection in cars further. To support this work the UK undertook an analysis to estimate the benefit for potential options to modify UNECE Regulation 95. The analysis used the UK national STATS19 and detailed Co-operative Crash Injury Study (CCIS) accident databases. Of the potential options reviewed, it was found that the addition of a pole test offered the greatest benefit.
Since 2005, the motorcycle crash fatalities in the US exceeded 10% of the overall annual traffic fatalities. Consequently, it has become critical to gain in-depth understanding of the factors and characteristics contributing to motorcycle crashes. Unfortunately, there currently exists no database gathering the necessary information for an in-depth analysis of the US motorcycle crashes. So this study utilizes the NASS/CDS database (National Automotive Sampling System, Crashworthiness Data System) in order to gain insights into the patterns and factors leading to a NASS/CDS motorcycle crash, from 1997 to 2007. NASS/CDS samples about 5,000 passenger car tow-away crashes per year. Each case includes photographs and detailed data on crash and pre-crash characteristics, vehicle types, trajectories, types of impact, and other pertinent roadway and crash scene information, allowing an in-depth investigation of the crash mechanisms. However, the NASS/CDS sampling process specifically focuses on passenger car crashes, so the cases extracted only correspond to crashes in which a passenger vehicle was towed, and a motorcycle was somehow involved. Thus, a by-hand in-depth review of about 200 cases allowed retrieving 106 relevant crashes for this study, tending to represent the severe passenger vehicle(s) versus motorcycle(s) crashes on US roads. The findings lead to the conclusion that these crashes mostly result from the low conspicuity of the motorcycle, and from the inability of the car drivers to fully appreciate and anticipate the behavior of a motorcycle. Indeed, it has been shown that, first, the car drivers involved in these cases did not attempt any avoidance maneuver, second, they were largely of ages under 25, and finally, the majority of the crashes were in an intersection scenario. In addition, the two major scenarios unveiled were the car attempting a left turn from the opposite direction and the car attempting a left turn from the right. The paper mentions several solutions to enhance the motorcycle- conspicuity and to allow the car drivers to better anticipate its behavior, which seem to be key factors in the intersection-related crashes (and more generally in the passenger vehicle(s) versus motorcycle(s) crashes).
The share of high-tensile steel in car bodies has increased over the last years. While occupant safety has generally benefited from this measure, there is a potential risk that, as a result, rescue time may increase considerably. In more than 60% of all car occupant fatalities a technical rescue has been necessary. These are in particular those cases where occupants die immediately at the accident scene. Therefore, in these cases "rescue time" is a very sensitive parameter. In addition to the general analysis of the need of technical rescue and the actual rescue time depending on model years, the injury pattern of occupants requiring technical rescue will be analysed to provide advice for rescue teams. Furthermore, a detailed analysis of rescue measures for the most popular car models depending on the safety cell design is given.
Entwicklung eines methodischen Rahmenkonzepts für Verhaltensbeobachtung im fließenden Verkehr
(2012)
Vielfach soll durch Verkehrsbeobachtung die Entwicklung des Verkehrsverhaltens im Zeitverlauf aufgezeigt und/oder die Wirksamkeit von Verkehrssicherheitsmaßnahmen bewertet werden. Charakteristisch dabei ist, dass bislang für jedes Vorhaben ein eigenes Studiendesign entwickelt wird, obwohl sich die jeweiligen Forschungsfragen trotz ihrer inhaltlichen Vielfalt aus methodischer Sicht auf einige wenige Grundmuster von Aufgabenstellungen reduzieren lassen. Im vorliegenden Forschungsvorhaben geht es deshalb darum, für Erhebungen, bei denen das Verhalten von Verkehrsteilnehmern im fließenden Verkehr beobachtet wird, ein möglichst allgemein anwendbares Design zu entwickeln. Ziel des vorliegenden Projektes ist es somit auch, das Spektrum der Methodik von Verhaltensbeobachtungen im fließenden Verkehr darzustellen sowie auf Probleme und Randbedingungen aufmerksam zu machen. Damit soll all denjenigen eine Hilfestellung gegeben werden, die beabsichtigen, solche Studien durchzuführen. In Kapitel 2 des Berichts werden zunächst einige nationale und internationale Projektbeispiele für Verhaltensbeobachtungen im fließenden Verkehr vorgestellt, auf deren Basis dann das hier in Rede stehende methodische Rahmenkonzept entwickelt wird. Kapitel 3 enthält eine Einordnung der Erhebungsmethode Verhaltensbeobachtung in den konzeptuellen Rahmen der Statistik und empirischen Sozialforschung. Im 4. Kapitel werden statistische Kennzahlen des Verkehrsverhaltens, die aus Beobachtungen gewonnen werden können, systematisiert und zusammengestellt. Kapitel 5 beschäftigt sich mit für Verhaltensbeobachtungen adäquaten Stichprobendesigns, in Kapitel 6 sind die zugehörigen Formeln für die Hochrechnung und Genauigkeitsbeurteilung (Konfidenzintervalle) der Ergebnisse aufgeführt. Kapitel 7 enthält einige methodische Hinweise zu Möglichkeiten und Grenzen der Ziehung einer Zufallsstichprobe für Verhaltensbeobachtungen, in Kapitel 8 wird der Prozess einer Verhaltensbeobachtung unter erhebungspraktischen Gesichtspunkten - basierend auf dem entwickelten Rahmenkonzept - beleuchtet. Vor der Zusammenfassung und den Empfehlungen (Kap. 10) wird in Kapitel 9 noch ein Vergleich des hier entwickelten methodischen Rahmenkonzepts mit der Vorgehensweise bei den aktuellen Erhebungen der BASt zum Schutzverhalten angestellt. In den Schlussbericht sind auch die Ergebnisse eines Workshops eingeflossen, der im Rahmen dieses Projekts am 18.01.2010 in Bergisch Gladbach durchgeführt wurde.
Tree impacts are still one of the most important focal points of road deaths in Germany. For the year 2008, the latest figures in the national statistics show a share of 28% of road users killed in crashes with trees alongside a road amongst all crashes on rural roads (except the Autobahn). The official German statistics show the attribute "impact on a tree" since 1995. For this first reported year, the share of road users killed in such crashes was 30%. During the last 14 years, fatal accidents with road users killed on rural roads (except the Autobahn) after impacts on a tree declined by 60% from 1,737 (year 1995) to 696 (year 2008). But this is more or less in line with the general evolution of vehicle and traffic safety in Germany. For Germany as a whole the accident statistics do not show a reduction for "treer crashes" which is clearly more than the average for all accidents. But, as shown with the paper, there are different evolutions in the several German States. In public awareness the topic "tree impacts" is mostly associated with the situation in Germany after the reunification. At that time a lot of road users were killed on the avenues in the so called "new countries". The fact that "tree impacts" are still a big share within the figure of killed road users seems to be little-known. Using updated information coming from the official statistics and in-depth-studies, accident researchers can identify a big potential for further improvements of traffic safety on the associated district roads, state roads and federal highways. There is still a need to analyse more details of the accident occurrence with impacts on trees to generate new and updated findings on the current limits and potentials of measures to improve vehicle and traffic safety. To make further efforts in reducing the figures of victims of "tree impacts" the intensification of well-known conventional solutions " for example implementation of guard rails and reduction of speed - is an option. Measures related to vehicle safety technology especially in the field of primary (active) safety will have additional benefit within the physically imposed limits. With this background it can be seen that the subject "tree impacts" should be analysed with a holistic approach taking into account the entire system of driver, vehicle, road, the environment and a social consensus as well.
The accident research of Hanover and (from 1999 on) Dresden registered 736 leg injuries (AIS ≥ 2) from 1983 to March 2007. 174 of these injuries (23.6 %) were fractures or dislocations of foot and ankle. 149 feet of 141 front seat car occupants in 140 cars were affected. Of these 117 were drivers, 24 were front seat passengers. The mean age of occupants was 38.5 -± 16.8 years. Ankle fractures were the most frequent injury (n = 82; 80 malleolar fractures, 2 pilon fractures). 34 fractures and dislocations affected the hindfoot (5 talus and 26 calcaneal fractures, 2 subtalar dislocations and 1 subtotal amputation) , 16 to midfoot (4 navicular fractures, 5 cuboid fractures, 3 fractures of cuneiformia, 2 dislocations of chopart joint, 1 subtotal amputation, and one severe decollement) and 39 the forefoot (metatarsal fractures). Open fractures were seldom seen (2 malleolar fractures, 1 metatarsal fracture). Both feet were injured in 10 cases. 33 occupants (23.4 %) were polytaumatic had a polytrauma, 17 of them died. 81 percent of the occupants were belted. The cars were divided in pre EuroNCAP (year of manufacture 1997 and older) and post EuroNCAP cars (year of manufacture 1998 and newer). Most of the foot injuries were seen in pre EuroNCAP cars. Most of the occupants sat in compact cars (40 drivers and 9 front seat passengers) and large family cars (27 drivers and 7 co-drivers). 49 of 140 accidents occurred on country roads, 26 on main roads and 13 on motorways. The crash direction was mostly frontal. Generally were found no differences of delta v- and EES-level between the injured foot regions, but divided into pre- and post-EuroNCAP cars there was a tendency to higher delta v- and EES-levels in newer cars. The frequency of foot injuries increased linearly with increasing delta v-level; but above delta v-level of 55 km/h the linear increase only was seen in pre-EuroNCAP cars, post-EuroNCAP cars showed no further increase of injuries. The footwell intrusion showed no difference between the injured foot regions but pre-EuroNCAP cars had a tendency to higher footwell intrusion. There were no differences in footwell intrusion between the car types. Only 29 of 174 fractures or dislocations of foot were seen in post-EuroNCAP cars, the predominate number of these injuries (n = 145) were noticed in pre-EuroNCAP cars. A lower probability of long-term impairment was found in post-EuroNCAP cars for equal delta v levels, using the AIS2008 associated Functional Capacity Index (FCI) for the foot region.
Aim of the study was to evaluate the protective effect of bicycle helmets particularly considering injuries to the head and to the face. Accidents with the participation of bicyclists which occurred from 2000 to 2007 were chosen from GIDAS. We observed that injuries to the head and face were more severe in the group of non-helmeted riders. There seems to be no significant difference in injuries with AIS 3-6. Altogether 26 cyclists were killed. 2 of them wore a helmet (1% of helmeted cyclists), 24 did not (1% of non-helmeted cyclists). Only one killed rider (without helmet) did not suffer from polytrauma (only head injuries recorded). The findings seem to support the thesis of a preventive effect of the bicycle helmet, however the two groups are different in their characteristics related to riding speed. Necessarily we need a multivariate model to evaluate the effect of helmets.
Although the statistics show a decreasing rate of child injuries and fatalities in German road accidents more efforts can be made to protect children in cars e.g. by developing appropriate child restraint systems. An important part in of this work can be achieved with the help of crash tests using child dummies. However these crash tests cannot completely reflect the situation of real world crashes as factors like children moving out of the optimal position or children incorrectly fastened by their parents are difficult to predict. Therefore this study gives an overview over the current accident and injury situation of child occupants in cars in German road accidents.
In order to enable foreseeing or comparing the benefit of safety systems or driver assistance systems in Germany, in the United States and in Japan, the traffic accident databases in those three countries are examined. The variables used are culpable party, collision partner, accident type, and injury level and the method to re-classify the databases for comparison are proposed. The result indicates that single passenger car fatality is the most frequent in Germany and in the United States, while passenger car vs. pedestrian is the most frequent fatality scenario in Japan. When the casualty by fatality ratio is focused, the greatest difference is observed in rear-end collisions. The ratio of slight injuries in Japan yields about eighteen times as many as those in Germany, and about eight times as many as those in the United States.
The purpose of this study was to analyse the actual injury situation of bicyclists regarding accidents involving more than one bicyclist. Bicyclists were included in a medical and technical analysis to create a basis for preventive measures and discovered repeating accident patterns and circumstances such as daytime, environment, helmet use rate. Technical and medical data were collected at the scene, shortly after accident. The population was compared focusing on bicycle versus bicycle accidents. Technical analysis included speed at crash, type of collision, impact angle, environment, used lane and relative velocity. Medical analysis included injury pattern and severity (AIS, ISS). Included were 578 injured bicyclists in 289 accidents from years 1999 to 2008, 61 percent were male (n=350) and 39 percent female (n=228). Sixty-seven percent ranged between 18 to 64 years of age, twelve percent each between 13 to 17 years of age and older than 65 years, eight percent between 6 to 12 years and one percent between 2 to 5 years.. Crashes took place in urban areas in 92 percent, in rural areas in 8 percent. Weather conditions were dry lanes in 97 percent and wet conditions in 3 percent. Eighty-three percent of all accidents happened during daytime, ten percent during night, and seven percent during dawn. The helmet use rate was only 7,5 percent in all involved bicyclists. The mean Maximum Abbreviated injury scale, Injury severity score was 1,31. Bicyclists are still minimally- or unprotected road users. The helmet use rate is unsatisfactorily low. The incidence of bicycle to bicycle crashes is high. Most of these accidents take place in urban areas. The level and pattern of injuries is moderate. Most of the more severe injuries occur to the head and could have been avoided by frequent helmet use.
An eCall device has been mounted on some vehicles in France since 2003. It is an integrated car radio/GSM/GPS system that can be used with a SIM card. When an accident occurs, a call can be sent manually or automatically made to a telephone call centre. Knowing the geographic location, the vehicle identity and the possibility of a direct communication with the people involved enables the nearest emergency services to be called out. In this context, the LAB / CEESAR have set up a study aimed at evaluating the effectiveness of this system. The purpose of this paper is to detail the E-call system evaluation method of effectiveness used and give a global synthesis of the results.
Although ATV accidents account for numerous deaths in the US and Australia, the role in traffic accidents and hospital admissions in Germany is unknown. At a level I trauma centre, hospital and crash charts were analysed for medical and technical parameters of ATV accidents. ATV drivers were 0.1% of emergency trauma patients. The mean total hospital stayrnwas 15 days; there were 1.5 stays per patients with 2.0 surgical procedures needed. One patient died, only two recovered fully. 14 cases of ATV accidents out of 18990 (0.1%) were documented within 10 years. The mean impact velocity was 35 km/h. Car collisions were predominant. The upper extremity was the predominant injured region (AIS 0.7), Mean maximum AIS was 1.4. ATV accidents in Germany are rare but pose high risk for severe injuries. Possible reasons are low active and passive security, limited experience and risky driving behaviour. Preventive measures are discussed.rn
Die vorliegende Untersuchung gibt einen zusammenfassenden Überblick über die Leistungen des öffentlichen Rettungsdienstes in der Bundesrepublik Deutschland im Zeitraum 2008/09. Das Forschungsprojekt 87.012/2008 "Analyse des Leistungsniveaus im Rettungsdienst für die Jahre 2008 und 2009" erfasst und analysiert eine repraesentative Stichprobe von Einsatzdaten zur Beurteilung der Leistungsfähigkeit des öffentlichen Rettungsdienstes in der Bundesrepublik Deutschland. Die wesentlichsten Ergebnisse des Forschungsprojektes sind: " Bundesweit werden im öffentlichen Rettungsdienst im Zeitraum 2008/09 jährlich rund 11,4 Mio. Einsätze mit insgesamt 14,2 Mio. Einsatzfahrten durchgeführt. Die Einsatzrate beträgt rund 143 Einsätze pro 1.000 Einwohner und Jahr. " An einem mittleren Werktag gehen bundesweit rund 35.000 rettungsdienstliche Hilfeersuchen in den Rettungsleitstellen ein. Am Wochenende sinkt die Zahl der eingehenden Hilfeersuchen auf rund 26.000 an einem mittleren Samstag und auf rund 24.000 an einem mittleren Sonntag. " 49 % des Einsatzaufkommens werden vom Leitstellenpersonal als Notfall eingestuft, 51 % entfallen auf die Kategorie Krankentransport. " Praktisch die Hälfte aller Notfalleinsätze werden unter Hinzunahme eines Notarztes durchgeführt (Notarzteinsatz). Ein Drittel der Notfälle zu Verkehrsunfaellen (32 %) wird von einem Notarzt bedient. " Das Rendezvous-System hat sich mit einem Anteil von 99,1 % gegenüber dem Stationssystem bundesweit durchgesetzt. " Rund jeder 17. Notfalleinsatz gilt einem Verkehrsunfall, was bundesweit rund 336.000 Einsätzen entspricht. Die Verteilung der übrigen Einsatzanlässe bei Notfällen mit und ohne Notarztbeteiligung beträgt: Internistischer Notfall 46 %, Sonstiger Notfall (z. B. Verbrechen, Suizid, dringende Blut- und Organtransporte) 37 %, Sonstiger Unfall (z.B. Haus-, Schul- und Sportunfall) 11 % und Arbeitsunfall unter 1 %. " Die Verteilung der Rettungsmitteltypen am bundesweiten Einsatzfahrtaufkommen im Zeitraum 2008/09 betraegt: RTW 52 %, KTW 29 %, NEF 18 %, NAW und RTH/ITH unter 1 %. " Beim Einsatzfahrtaufkommen werden rund die Haelfte der Einsatzfahrten mit Sonderrechten auf Anfahrt durchgeführt. Dies entspricht bundesweit jährlich 7,2 Mio. Einsatzfahrten unter Sonderrechten auf Anfahrt. " Das Einsatzfahrtaufkommen weist im Bundesgebiet 2008/09 einen Fehlfahrtanteil von unter 6 % auf. Bundesweit sind dies jährlich rund 808.000 Fehlfahrten. " Die Dispositions- und Alarmierungszeit bei Einsatzfahrten mit Sonderrechten auf Anfahrt beträgt im Mittel 2,1 Minuten. Bei Einsatzfahrten ohne Sonderrechte auf Anfahrt liegt die Dispositionsund Alarmierungszeit im Mittel bei 14,3 Minuten. " Bei Einsätzen mit Sonderrechten auf Anfahrt errechnet sich nach dem zuerst eingetroffenen Rettungsmittel am Einsatzort eine mittlere Hilfsfrist von 8,7 Minuten, wobei 95 % der Notfälle innerhalb von 16,7 Minuten mit einem Rettungsmittel bedient werden. " Die mittlere Hilfsfrist zu Verkehrsunfällen beträgt an Straßen innerorts am Tag 8,8 Minuten und in der Nacht 9,4 Minuten, an Straßen außerorts am Tag 10,4 Minuten und in der Nacht 10,9 Minuten. " Die Unterscheidung der Einsatzzeit nach Notfällen und Krankentransporten unter zwei Stunden ergibt eine mittlere Einsatzzeit von 51 Minuten für Einsatzfahrten mit Sonderrechten auf Anfahrt und 53 Minuten für Einsatzfahrten ohne Sonderrechte auf Anfahrt. " Die Transportzeit bei Einsatzfahrten mit Sonderrechten auf Anfahrt beträgt im Mittel 12,3 Minuten. Bei Einsatzfahrten ohne Sonderrechte auf Anfahrt liegt die Transportzeit im Mittel bei 16,1 Minuten. " Die Verweilzeit am Transportziel/Wiederherstellungszeit bei Einsatzfahrten mit Sonderrechten auf Anfahrt beträgt im Mittel 20,1 Minuten, während bei Einsatzfahrten ohne Sonderrechte auf Anfahrt der Vergleichswert im Mittel bei 15,9 Minuten liegt. " Die weiteren Ergebnisse der Pilotstudie zur Machbarkeit einer Datenerhebung und -analyse über die Ermittlung der Verletzungsschwere bei Verkehrsunfallopfern zeigen, dass die Analyse mittels Daten sowohl zur Rückmeldezahl in Hessen als auch mit Hilfe von DIVI-Notarztprotokollen möglich ist. Dabei ist eine Klassifikation mit Hilfe von Geodaten EDV-gestuetzt umsetzbar, um eine vergleichende Auswertebasis zu bilden.
Who doesn't wear seat belts?
(2009)
Using real world accident data, seat belts were estimated to be 61% effective at preventing fatalities, and 32% effective at preventing serious injuries. They were most effective for drivers with an airbag. Seat belts were estimated as having prevented 57,000 fatalities and 213,000 seriously injured casualties in the UK since 1983. Seat belt legislation was estimated to have prevented 31,000 fatalities and 118,000 seriously injured casualties. A future increase in effective seat belt wearing rate (which takes into account seating position) in the UK from 92.5% to 93% may prevent casualties valued at a societal cost of over -£18 million per year. To target a seat belt campaign, the question "who doesn"t wear seat belts?" must be answered. Seat belt wearing rates and the number of unbelted casualties were analysed. It was primarily young adult males who didn"t wear seat belts, and they made up the majority of unbelted fatalities and seriously injured casualties.
The bicyclist accidents were analyzed to get better understanding of the occurrences and frequency of the accidents, injury distributions, as well as correlation of injury severity/outcomes with engineering and human factors in two different countries of China and Germany. The accident cases that occurred from 2001 to 2006 were collected from IVAC database in Changsha and GIDAS database in Hannover. Based on specified sampling criteria, 1,570 bicyclist cases were selected from IVAC database in Changsha, and 1806 cases were collected from Hannover, documented in GIDAS database. Statistical analyses were carried out by using these selected data. The results from the statistical analysis are presented and discussed in this study.
One of the major problems of road safety in Europe is the powered two wheelers accidents. One of the European countries with one of the highest rates is Portugal where in 2006, mopeds and motorcycles fatalities represented 27% of all road users deaths. In this work, a deep analysis and overview of the current state of mopeds and motorcycles accidents for the 2004-2006 period is presented. Within this period 830 PTW occupants die, 2958 have been severely injured and 25000 suffer slight injuries. A detailed analysis of the conditions of these accidents has been carried out, using the data of the national accident database. This analysis provides global information, about geographic environmental conditions, driver- characteristics among others. From this data detailed information is obtained allowing to know when, where and who. In order to answer the question why more a widely collection of data has been collect for 70 accidents. The data has been collected using OECD methodology. For these accidents a detailed reconstruction has been carried out, what is especially important for fatal accidents where for instance speed in an important factor. From these collection and analysis of data a wider overview of facts and measures are extracted. Among them, some are emphasized such as that the quality and non-use of helmets plays an important role in severe and fatal accidents especially for accidents involving moped vehicles, or speed is the most important factor in fatal accidents involving motorcycles. Concerning motorcycle accident reconstruction, different tools can be used depending of the accident scenario and complexity. For simple cases, with specific characteristics, analytical formulation based in vehicle crash dynamics can be use in order to determine the impact speed of the vehicles impact, analysing the skid marks, deformations, victims rest position and considering parameters (EES, vehicle deceleration, etc). Aspects such as the energy absorption capability of motorcycles are also discussed. In the general cases the accident reconstruction software Pc-Crash has been used for the reconstruction of the accident. In very complex cases, has for instance the impact between motorcyclist and barriers, Madymo software is used especially to determine speed from injuries. An example of the impact of a motorcyclist and a motorcyclist-friendly barrier is present to illustrate the benefits and limitations of such systems.
In Germany averagely two million traffic accidents happen each year and emergency medical services are called to more than 400 000 patients. Even though this number is decreasing continuously (due to improvements in the fields of vehicle safety, road construction, and accident prevention) every case is yet a challenge for the rescuers and requires improvements in emergency medicine as well. Especially during diagnostics right at the accident scene, there are only limited instruments available to gain the necessary knowledge of the injuries suffered, to come to essential decisions about treatment or transport. To provide an additional diagnostic aid by scouting and estimating the situation, a software-tool calculating the likeliness of the most frequent severe injuries (AIS 3-6) of front occupants in passenger cars has been developed to deliver this necessary information about particular accident scenarios. To achieve this, logistic likelihood functions have been calculated in a multivariate regression analysis analysing all AIS 3+ injuries in the GIDAS database of the years 1999-2006 that happened more than four times
Crash involvement studies using routine accident and exposure data : a case for case-control designs
(2009)
Fortunately, accident involvement is a rare event: the chance of an individual road user trip to end up in a crash is close to zero. Thus, according to general epidemiological principles one can expect the case-control study design to be especially suitable for quantifying the relative risk (odds ratio) of accident involvement of road users with a certain risk factor as compared to road users that do not have this characteristic. Ideally, of course, the database for such a case-control study should be established by drawing two independent random samples of cases (accidental units) and controls (nonaccidental units), respectively. If, however, special data collection is not an option, it is nevertheless possible to analyze routine accident and exposure data under a case-control design in order to fully exploit the information contained in already existing databases. As a prerequisite, accident and exposure data from different sources are to be combined in a single file of micro or grouped data in a way consistent with the case-control study design. Among other things, the proposed methodological approach offers the possibility to use in-depth data of the GIDAS type also in investigations of active vehicle safety by combining this data with appropriate vehicle trip data collected in mobility surveys.
Nowadays, traffic accidents are recorded in historical databases. Regarding the huge quantity of data, the use of data mining tools is essential to help Experts, for automatically extracting relevant information in order to establish and quantify relations between severity and potential factors of accidents. An innovative approach is here proposed for an in depth investigation of real world accidents data base. Mutual information ratio based on conditional entropies is used to quantity the association strength between an accident outcome descriptor (injury severity) and other potential association factors. Information theoretic methods help to select automatically groups of factors mostly responsible of the severity of accident.
A lot of factors are related to a road traffic accident; particularly human factors such as road use characteristic, driving maneuver characteristic and safety attitude are the major ones. As a random factor is also included, so it is necessary to minimize the contribution of a random factor to identify human factors related to a road traffic accident. There are several standpoints for traffic accident analysis, such as vehicle-based, location-based and driver-based. And it is effective to analyze driver-based traffic accident data for discussion on the relation between human factors and accidents. An integrated traffic accident database system was developed for analysis considering driver- accident and violation records by ITARD, and several studies were carried out for the evaluation. Useful data for discussion on the relation between types of collision and traffic violations, and the effect of accident experience to the following accident were obtained.