The main object of the paper is to illustrate the importance of studying road accidents, and in particular, vehicle accidents, from the viewpoint of probability. Casualty information in these accidents is usually presented as percentages of the various levels of injury, sometimes including the percentages of uninjured. In Great Britain the data are at present available for four injury levels only, death, serious injury, slight injury and no injury. A method is described for calculating these percentages in head-on collisions. The starting points are (1) a diagram giving the probabilities of the four injury levels versus velocity change of the vehicle, and (2) a distribution diagram of the percentages of head-on collisions within successive equal intervals of relative velocity at impact. The injury percentages were determined for ratios of the masses of the two vehicles ranging from 1.0 to 9.0, and satisfactory agreement was found with results of an analysis of national accident statistics. The method is suitable for other types of vehicle collisions, when the basic data becone available. Reasons are given for expecting the probability curves in all types to be similar in general form. The theory and calculations are shown to lead to a better understanding of how the injury percentages are built up. Attention is drawn to the comparatively low probability of serious injury even at high velocity changes, and it is suggested that studies of the uninjured or slightly injured in severe frontal impacts may be as rewarding as studies of those injured more seriously.
Abstract