GAP ACCEPTANCE INVESTIGATIONS IN CANADA

This paper presents the findings of a research project that exami the behaviour of drivers making left turns across the path of opposing vehicles at T-intersections without traffic signals. Th objective was to identify factors affecting the decision to accep reject an available gap. The focus of the paper is on the result rather than the techniques used.The data base involved five intersections in the Edmonton area. Over 3400 left turning and 13,000 main stream vehicles, and 17 driver characteristics and ga attributes, were observed. The probability of accepting or reject a gap was investigated as a binary choice in a logit model. Over different model structures were examined. Gap size, driver gende presence of passengers in the turning vehicle, and queue delay we found to be important factors in decision-making. Probability of accepting a gap was also influenced by the speed and type of the opposing vehicle, queue delay, microscopic flow of opposing traff and the presence of vehicles behind the turning vehicle. Perhaps most interesting results were obtained by the analysis of the imp of front delay. While for delay duration shorter than 30 seconds probability of accepting a gap decreased, it increased after that period of time. (a)