Driver behaviour research in road traffic encounters the problem that it is often not clear how behavioural effects can be translated into real risk indicators. Most driver behaviour studies that study the effects of a specific driver support system end with conclusions in terms of changes in driving behaviour as a result of driving with the system. But what do these changes really mean for traffic safety? This paper investigates a couple of existing models that can make this step from results in terms of driving speed and headway to accident risk. It is concluded that for average driving speed the model developed by G. Nilsson (1984) is the best one to use. This model does not take into account the standard deviation of driving speed. For computing the consequences of this measure a model developed by Salusjarvi (1990) can be used. Janssen (1999) developed a model describing the relationship between headway and accident risk. These models are applied to results of a driving simulator experiment with Adaptive Cruise Control. For the covering abstract see ITRD E113725 (C 22328 CD-ROM).
Samenvatting