Abstract: Through field investigation data, the critical gap of vehicles to pedestrians was calculated in this paper, finding that the cumulative Weibull distribution could be used to fit the critical gap of right-turning vehicles. The binary logistic regression model was used to establish gap acceptance model. The results show that when the gap is larger, and the pedestrian flow is lower, drivers are more likely to accept the gap, and they tend to accept the gap easily when the crossing pedestrian is alone. Drivers are more cautious about short time intervals and tend to accept shorter time lags. The study on the gap acceptance behavior of right-turn drivers to pedestrians crossing the street can not only lay the foundation for the safety assessment of the intersection, but also can be used to analyze the delay and traffic capacity of the intersections.

Key words: pedestrian safety, gap acceptance behavior, critical gap, binary logistic regression