School of Architecture and Civil Engineering

A contribution to the improvement of road safety investigations.


The road construction and traffic engineer, as one of those significantly responsible for the construction and operation of roads, requires statements on the safety/dangerousness of a traffic facility or the traffic that is handled on a traffic facility. For the one who has to give a safety analysis, the problem arises to find out from existing and emerging methods the one which is appropriate for his problem definition. For many questions, the obvious approach is to analyze accident data in order to arrive at conclusions about the accidents themselves and, finally, about hazard conditions. Investigations that are carried out for the purpose of immediate practical implementation in the area of responsibility of the road operator fall under the term of engineering accident investigations.These investigations are subject to narrow limits, since the underlying ideas force the application of stochastic methods with the corresponding statement uncertainties and fuzziness, and the need for pragmatic procedures in practical investigations (e.g. survey time) generates considerable inaccuracies. The replacement of the consideration of accident collectives by the analysis of observation results of the traffic behavior on site is therefore a necessary part of the discussion about methods of safety analyses.traffic events, however, are not easily observable/measurable in a way that allows to obtain results with reasonable effort for practical questions. Therefore, a proposal for a procedure is elaborated, which allows the survey of traffic events in the context of safety analyses to lead to a form of results adapted to the needs of road construction and traffic engineers.Based on the basic idea that events/situations with the property "potentially dangerous" occur in pre-accident phases and that danger potential can be identified in externally observable traffic situations, it is proposed to classify observed situations as normal (safe), dangerous (with conflict) or as potentially dangerous. The frequencies with which these assessment classes are occupied could be used as input variables for safety analyses by road and traffic engineers, to solve their specific problems. Since the survey of hazard potentials does not depend on the long observation durations of the traffic conflict technique, the practical applicability would be significantly less limited than that of the traffic conflict technique and thus a universal usability as an investigation tool would be given.

This Abstract was translated from German with deepL and could be faulty.

Weitere Infos über #UniWuppertal: