Evaluating the Impact of Midblock Pedestrian Signals on Vehicle Delays

The Florida Department of Transportation has recently introduced midblock pedestrian signals (MPSs) to improve pedestrian safety and traffic efficiency at midblock crossings. This study evaluates the impact of MPSs by analyzing vehicle delay at each pedestrian–vehicle interaction and comparing them with other midblock crossing systems. To conduct the evaluation, CCTV video data were collected from 14 MPS-equipped locations and five reference sites. Advanced computer vision techniques, including the real-time detection transformer model for object detection and the ByteTrack algorithm for tracking, were used to process the video data. The study employed both cross-sectional and before–after evaluation methods, incorporating the comparison group and empirical Bayesian approaches to estimate delay modification factors. To account for repeated measures and within-site correlation, delay functions were developed using generalized estimating equations with a gamma distribution, offering improved model robustness over traditional generalized linear models. Results across all evaluation methods indicate that MPSs increase vehicle delays by 8% to 14% compared to RRFBs and flashing beacons, due to mandatory stopping phase. However, when compared to pedestrian hybrid beacons, which share similar functionalities but differ in signal phase management, MPSs reduced vehicle delays by 5% to 7%. Given their traffic efficiency, ease of implementation, and safety benefits, MPSs can serve as a viable alternative for upgrading existing crossings or installing new pedestrian signal systems.