The papers reviews the development of traffic safety research, mainly using examples from the authors own work in the field. The idea is to exemplify the shift from traffic accident research to driving task analysis, drifting from one extreme to the other on a safety continuum. It starts with research in the validity of the Traffic Conflicts Technique with respect to accident data. In international calibration studies national techniques were compared and refined. After an extension on pedestrian and cyclist conflicts the German version of TCT was laid down in a handbook. The idea of observing critical traffic events was soon enlarged to contain errors of traffic participants. Error counting methods were developed and applied to many different contexts. A variety of studies using error data for traffic safety evaluation is mentioned. The main definitions, distinctions and achievements as well as problems of the approach are described. The relation between errors, conflicts and accidents is reflected to evaluate the utility of errors as a measure of safety. Some thought is given to the causes of errors in traffic systems. Our error counting methodology has always been related to the “situational approach”, meaning that a taxonomy of traffic situations and lists of driving tasks had structured our observation data. Recently we have used this approach as an access to a driving task analysis. Its basic concept is presented shortly. Fastenmeier & Gstalter (see the other abstract from our group) describe it in more detail and give examples of different applications in the field of traffic safety The use of a multitude of methods in safety research is finally shown using the topic of older car drivers and some conclusions on the utility of these methods are drawn. (Author/publisher) This publication may be accessed by Internet users at: http://www.ictct.org/workshop.php?workshop_nr=25
Abstract