Intersection Signal Systems with Intelligent Pedestrian Accommodation: Dynamic Pedestrian Timing

Perhaps the most critical issue facing urban transportation engineers is the optimization of signalized intersections, with the goal of increasing both safety and efficiency for all modes without additional infrastructure. Furthermore, accommodating pedestrians at high volume intersections is becoming more challenging as increasing vehicular and pedestrian demand requires more green time. To safely accommodate pedestrians, the Highway Capacity Manual prescribes that the parallel vehicular green must be at least equivalent to " WALK " plus the " pedestrian clearance interval (PCI) ". A fixed walking speed is used to calculate the PCI. Despite its effectiveness, this practice is not optimal from both and operations and safety perspective since some pedestrians walk more slowly or quickly than the design 4.0 feet/second pedestrian. Variability in the required PCI is not captured in the current design process. A problem also arises when the required PCI length exceeds the parallel vehicular green requirement. Additional green for longer PCI is unnecessary for vehicular flow efficiency and takes green time away from the conflicting phase(s), potentially increasing intersection delay. In this research, the concept of dynamic PCI timing in traffic signal operations was explored. Two signal systems, the extended NEMA system and a fuzzy logic controller (FLC), were developed and evaluated, with dynamic PCI control, against current signalization methods. Both the extended NEMA and FLC proved to significantly improve operations with most cases evaluated. This research shows that traffic signal control can remove the dependence on a design walking speed and the ongoing debate on what this speed should be. Given there are reportedly 325,000 signalized intersections in the U.S., the potential impact of this research could be significant in traffic safety and operational efficiency.