Real-world crashes show a wide variability in terms of the people involved, the characteristics of the vehicles and the crash configuration. To protect all road users systems should not be optimised for one specific crash test, instead they should have versatile and robust designs that together provide the optimum protection for the full crash population. The current use of a small range of crash conditions to specify the performance of cars in crashes opens the possibility that vehicles will be optimised for these tests rather than for the full range of real-world conditions. Research is needed to develop methodologies to engineer systems for maximum benefit, particularly for side-impact protection where safety systems are less developed. Additionally, a wider range of crash types needs to be incorporated into the development process of new cars, and methodologies based on physical or virtual testing are needed to support this. These methods should take account of the natural bio-mechanical variations between individuals as well as the range of vehicle types within national fleets.
- Car occupants
- Other road users