Impact of vehicular traffic stream on pedestrian crossing behavior at an uncontrolled mid-block section

Abstract

A term Total Crossing Time (TCT) is coined in this study. TCT of a pedestrian includes the time spent by the pedestrian in waiting and his/her crossing of a road. This paper argues that considering a distribution of TCT is required for analysis instead of analyzing a crisp value of TCT of pedestrians. This study quantifies, perhaps for the first time, that by what extent vehicular traffic streams affect TCT distribution of pedestrians crossing an uncontrolled mid-block section of an urban road. For this, a suitable parameter of traffic stream that influences TCT distribution is identified. The suitable parameter of traffic is found to be the traffic density. Further, it is hypothesized that pedestrians can only perceive a range of traffic density and not discrete values of density. In order to incorporate this, traffic density observed on the road is categorized into nine groups; and, for each group of densities corresponding TCT distribution is computed. Curve fitting, using two kinds of density functions, for each of the nine TCT distributions is performed and Pearson Type-III density function is found to be a more appropriate distribution. Additionally, the parameters of the fitted Pearson Type-III distribution are found to be dependent on the corresponding mean of the traffic density group. Hence, a simple regression model is also suggested using which one can predict TCT distribution if traffic density of the stream is known. The findings of this study is going to be useful for researchers/practitioner those who are interested in simulating pedestrians, crossing an uncontrolled mid-block section of a road.