Diss. Lund : Lunds universitet. Tekniska högskolan. Teknik och samhälle. Trafikplanering, 2004

Traffic safety work needs different methods and tools in order to choose and evaluate traffic safety measures. The thesis contributes to this problem by presenting and visualizing a method which describes the traffic safety situation in several dimensions. The method used to describe the traffic safety problem shows the potential of a simultaneous presentation and evaluation of these dimensions and demonstrates that the method can be expanded to several dimensions or ratios estimating the exposure, the risk and the consequence. This is illustrated in describing the traffic safety situation for different road user groups and age groups. The power model, which estimates the relationship between speed and safety, is not a new tool as the model has been used in both theory and practise in several countries for many years. In the thesis the theoretical and practical background are presented. The power model is here also tested and validated in a cross-sectional study. These analyses show that the power model is valid with regard to injury accidents, fatal accidents and the number of injured but not for the number of fatalities. The power model underestimates the effect on fatalities.