In 2001 the European Union set itself the ambitious objective of halving the number of fatalities on European roads by 2010 (from 50 000 to 25 000). While progress is being made, road accidents have still caused 41.600 deaths on EU roads in 2005. This large number of accident related deaths causes high costs to society. Besides action on behaviour and the vehicle, infrastructure is the third pillar of the EU road safety action programme. Therefore, it is now time to integrate the three lines of action and to address in a comprehensive way infrastructure measures on Community level. Present road designs result from many decades of construction and maintenance, in a time when safety issues were not always considered to the same extent. Today, several road features no longer meet the latest safety requirements. Moreover, traffic conditions may have changed since the road was designed and built. While roads are usually designed according to criteria concerning urban or regional planning, travel time, user comfort and convenience, fuel consumption, construction cost and environmental impact, safety is often implicitly assumed to be achieved by simply adhering to prescribed standards of alignment and layout. Experience shows that abiding by those standards is not sufficient to avoid hazardous features. Many lives could be saved and many accidents avoided if the existing road infrastructure was managed according to the best practice of safety engineering. Action needs to be taken on the selection of high risk road sections or black spots on the basis of local accident records. The thematic network EURORAP II1 has shown how affordable and well-designed engineering and enforcement measures applied in the right place can reduce the risk that a particular type of crash might lead to death or severe injury of the occupants of the vehicle. For instance, appropriate new signals at junctions can reduce the risk of fatal side impacts by up to 75%; pedestrian crossings at dangerous junctions can lead to a potential reduction of the risk of collisions with vulnerable users by up to 85%. (Author/publisher)
Abstract