Development of traffic conflicts technique for different environments: A comparative study of pedestrian conflicts in Sweden and Jordan

Abstract: This study is aimed at improving the current Swedish Traffic conflicts Technique [TCT] in relation to vehicle-pedestrian conflicts. The present definition of conflict severity appears to produce less severe conflicts than they might be, particularly if the relevant road user (RRU) is the pedestrian. Another aim is to apply the Swedish TCT in Jordan as a method for evaluating safety conditions at junctions. The difference in conflict perception in the two countries is explored throughout this thesis. The study was based on collecting data [accident, conflict, and traffic volume] from selected junctions with and without signal control in urban areas (20 junctions in Sweden and 22 junctions in Jordan). Alternative definitions of traffic conflicts and RRU were validated against the present definition of serious conflicts. The results indicated that the present definition of serious conflicts could be improved. The most valid definition of conflicts for non-signalised or signalised junctions or both junction types together was obtained by shifting the present threshold to the more severe direction of the time to accident-speed graph by 0.25 seconds. The present definition of RRU (road-user who acted evasively and was subjected to the least sever situation) produced the most valid definition for non-signalised junctions. The high definition of RRU (the road-user who in the situation produced the highest severity, i.e. the driver) produced the most valid definition for signalised junctions. Sub-groups of conflicts were formed according to the present definition of conflicts by excluding conflicts that involved RRU under the high definition who is driving at a speed less than 20km/h. They produced accident predictions as those produced by the most valid definition of conflicts, particularly for non-signalised junctions. The results indicated that the most valid definitions in Jordan were not the most valid definitions in Sweden. Some conflict definitions, however, have high validity for non-signalised junctions in both countries. The technique must take into account the local conditions in order to be applicable in other countries. According to this study, a minimum requirement is to establish local conversion factors between accidents and conflicts, if the technique is to be applied for accident prediction purposes. Finally, the road-user perception of conflicts is different in the two countries.