ANALYSIS OF TRAFFIC ACCIDENTS-THE GEOMETRIC DESIGN OF HIGHWAYS AND TRAFFIC CONDITIONS
Norihiko Kurimoto
Abstract
Norihiko Kurimoto
Abstract
To grasp the relationship between geometric design and traffic accidents on the national highways of Japan surveys were conducted on approximatley 31,800 traffic accidents involving personal injuries or deaths which occured during the 3-year period of 1971 to 1973 along a section of approximately 1,800 km of the arterial national highways (mostly sections designed for speeds of 50-80 km/h). The surveys covered primarily relations between accidents and geometric design of non-intersection portions and studies were carried out on the single and multiple correlations between these factors. Principal results of the single correlative analysis on non-intersection roads were as follows. Conditions of occurence by type of accident: Accidents were similar for each year. For example in 1974, vehicle-vs-vehicle accidents accounted for 80% of the total and consisted of approximately 60% rear-end collisions and 15% head-on collisions. Intersection density: The number of intersections per unit length also greatly influences the accident rate (number of accidents/100,000,000 vehicle-kilometers) on non-intersection portions. Horizontal curves: Although the accident rate (excluding pedestrian-vs-vehicle accidents) is extremely high where the horizontal radius is less than 100 meters, substantial differences on the accident rate could not be observed within the range of 250 to 1,000 meters. However, in head-on collisions, the accident rate increases as the horizontal radius decreases under the range of 800 meters. Also, the accident rate increases with increases in the variation of the horizontal radius. Vertical gradient: The accident rate (excluding pedestrian-vs-vehicle accidents) in ares where the vertical gradient exceeds 6% is considerably higher than that where the gradient is less than 6%. On downgrades in particular, the accident rate is high when it exceed 6% in rural areas and 4% in urban areas. Alignment combinations: In connection with relations between horizontal curves and vertical gradient combinations and accidents, accident rates were high particularly in sections where the horizontal curve radius was under 250 meters and the vertical gradient over 4%. Roadway width: Accident rates are higher on narrow roads, even sections where the width remains even, and a similar trend in increased accidents is observed on roads where the width varies. Traffic volume: Accident rates decrease with increases in vehicle traffic density (traffic volume/design capacity).
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
To grasp the relationship between geometric design and traffic accidents on the national highways of Japan surveys were conducted on approximatley 31,800 traffic accidents involving personal injuries or deaths which occured during the 3-year period of 1971 to 1973 along a section of approximately 1,800 km of the arterial national highways (mostly sections designed for speeds of 50-80 km/h). The surveys covered primarily relations between accidents and geometric design of non-intersection portions and studies were carried out on the single and multiple correlations between these factors. Principal results of the single correlative analysis on non-intersection roads were as follows. Conditions of occurence by type of accident: Accidents were similar for each year. For example in 1974, vehicle-vs-vehicle accidents accounted for 80% of the total and consisted of approximately 60% rear-end collisions and 15% head-on collisions. Intersection density: The number of intersections per unit length also greatly influences the accident rate (number of accidents/100,000,000 vehicle-kilometers) on non-intersection portions. Horizontal curves: Although the accident rate (excluding pedestrian-vs-vehicle accidents) is extremely high where the horizontal radius is less than 100 meters, substantial differences on the accident rate could not be observed within the range of 250 to 1,000 meters. However, in head-on collisions, the accident rate increases as the horizontal radius decreases under the range of 800 meters. Also, the accident rate increases with increases in the variation of the horizontal radius. Vertical gradient: The accident rate (excluding pedestrian-vs-vehicle accidents) in ares where the vertical gradient exceeds 6% is considerably higher than that where the gradient is less than 6%. On downgrades in particular, the accident rate is high when it exceed 6% in rural areas and 4% in urban areas. Alignment combinations: In connection with relations between horizontal curves and vertical gradient combinations and accidents, accident rates were high particularly in sections where the horizontal curve radius was under 250 meters and the vertical gradient over 4%. Roadway width: Accident rates are higher on narrow roads, even sections where the width remains even, and a similar trend in increased accidents is observed on roads where the width varies. Traffic volume: Accident rates decrease with increases in vehicle traffic density (traffic volume/design capacity).
Key concepts: Intersection (aeronautics), Geometric design, Range (aeronautics), RADIUS, Pedestrian, Transport engineering, Kilometer, Traffic volume