| description abstract | This study aims to evaluate and compare vehicle delay at two selected midblock rectangular rapid flashing beacon (RRFB) sites and five selected midblock pedestrian hybrid beacon (PHB) sites in Florida using video data collected during the study period of July to November 2021. Computer vision and data processing were used to calculate the average delay for each vehicle that interacted with a crossing pedestrian. The vehicle delays were then averaged separately for each lane and direction and for the combined directions, creating three data sets having total numbers of observations of 937, 618, and 441, respectively. Initial Mann-Whitney-Wilcoxon tests revealed significant differences in average delay values across different treatment types and periods for each of the three data sets. In addition, a generalized linear regression model considering various traffic and pedestrian factors revealed significant differences in vehicle delay across RRFB and PHB sites. PHB presence, weekend, signal activation, pedestrian crossing speed, vehicle speed, pedestrian starting position, traffic flow, and the afternoon/evening period have been found to be significant determinants of vehicle delay. The model results remained relatively stable when average delay was considered separately for each lane and direction; however, performance decreased when the delay was averaged for the combined directions of the roadway. The results suggest that vehicle delay increased at PHB sites compared to at RRFB sites, indicating a possible decrease in traffic efficiency at PHB sites. | |
