Integration of car and road infrastructure design : crash severity and occupant injury risk evaluations in frontal real-world crashes

Sammanfattning: During the last decade, traffic safety has not only been a matter of vehicle crashworthiness but an increasing interest can also be observed in the integration between vehicle and road infrastructure design. Research projects with a focus on merging these areas are however in their infancy. In 1997, a new safety concept was introduced in Sweden, Vision Zero , whose goal was, and still is, to reach zero fatalities and serious injuries in road traffic. This vision is not only a new road safety strategy amongst others, it also urges a new way of thinking about the vehicle, road and driver as a system. Since the basis of the vision is human tolerance to trauma, the analysis of injury risk is a fundamental step in understanding occupant injury tolerance in the road transport system. To reduce occupant injury risk in the road transport system, the different components included in the system need to be compatible. Roads have to be designed to avoid crashes or to mitigate the severity of crashes with all kinds of vehicles or objects. The general objective of this project was to investigate frontal crash severity for different types of road designs, road-related parameters and different collision partners in two-vehicle crashes and single-vehicle crashes. Furthermore, an objective was to study the influence of crash pulse characteristics on injury risk in frontal crashes. One common solution adopted by road constructors is to use guardrails to avoid hazard areas and to reduce crash severity by for example extending duration of the crash. Further objectives were to study the dynamic characteristics of different guardrails at various impact angles and find if there is any influence of crash pulse duration on injury risk. Studies I III are retrospective studies of car crashes investigated in-depth in Sweden. Study IV is based on crash test results. The materials used in studies I-III were data from vehicle crash recordings of real-world impacts and vehicle inspections and study III also includes data from insurance files including hospital records and police reports. Data extracted from the measured crash pulse were change of velocity, mean and peak acceleration and pulse duration. In the main, moderate and more severe injuries sustained in frontal crashes were analyzed. There was a 45% higher mean acceleration in crashes into rigid objects compared to deformable objects. A long duration of crash pulse does not lead to a high injury risk as long as the mean acceleration is low. Deformable guardrails show a relatively long duration and a large potential for injury reduction. It was found that the largest average crash severity was found on undivided roads with the most severe impact type, frontal collisions with oncoming vehicles. Intersections were shown to have the second largest crash severity with oncoming vehicles as the most severe impact type. The most hazardous collision partner was shown to be HGVs and rigid objects. The results show increasing crash severity for increasing posted speed limit, except for roads with a posted speed limit of 110 km/h. Roads with a speed limit of 110 km/h most often have separated lanes, indicating that separating lanes is important in reducing crash severity. The results also showed a lower crash severity on roads with lower road surface friction. Most often change of velocity is used in studies of injury risk in vehicle crashes. Change of velocity is often well correlated to mean acceleration. However, this thesis shows that mean acceleration should be the primary parameter to be used as design guideline in the design of a crashworthy road transport system. High change of velocity could be handled as long as the mean acceleration is kept low. The knowledge of crash severity and injury risk from various real-world conditions presented in this thesis could be used in the development and safety effect prediction of new safety technologies.