Accident frequencies have inherent limitations as indicators of traffic safety. These limitations are especially paramount in small-scale research. To overcome these difficulties, substitution of accident frequencies by conflict counts is often suggested. Some methodological aspects of the use of conflict counts as indicators of traffic safety are discussed. This provides a framework for the evaluation of available research results and/or the planning of new research. More detailed information is given in the appendices. Most methods to collect traffic conflict data use individual observers to decide on conflict type etc. Since this introduces subjective elements in the method, also objective quantitative measures of traffic conflicts have been proposed. Some of these objective measures are compared, resulting in some preference of the authors for the time-to-collision concept (TTC). A description of the observation method of the Institute for Perception TNO is given, which allows for computing TTC values. (Author/publisher).
Samenvatting