An Approach Towards the Integration of Bus Priority, Traffic Adaptive Signal Control, and Bus Information/Scheduling Systems

Abstract

This paper addresses the integration of adaptive traffic signal control and bus priority. In bus priority, several possible approaches are used for giving more “weight” to the buses: (1) passive priority, when signal timings are set, ahead of time, so that buses incur less delays, (2) active priority, where buses are detected at approaches to the intersection and phase splits are adjusted to accommodate the bus, and (3) “optimization-based” priority where the current state of the system is estimated and the signals are changed as per active priority schemes. Our work is related to the last approach where the signals are set based on real-time optimization of the phasing that considers all the vehicles on the network, the passenger counts in the buses, and the schedule status of the buses.

The architecture for phase optimization is based on the RHODES ™ traffic adaptive signal control system developed at the University of Arizona. RHODES ™ (Real-time, Hierarchical, Optimized, Distributed, and Effective System) takes second-by-second data from loop-detectors at the intersections as input, and outputs the durations of the phases. Objectives for optimizing phase durations include, among others, “minimize average delay per vehicle.”

When bus priority, referred to as “BUSBAND,” is introduced into RHODES ™, it is assumed that exact locations of buses are available in the network, as well as passenger counts through an advanced communication/information system. In this way, the bus is given a weight that depends on the number of passengers, and whether the bus is behind schedule. The RHODES™ /BUSBAND scheme was analyzed using a micro-simulation modeling package known as CORSIM.

RHODES ™, with and without bus priority, significantly increases average travel speeds and decreases total traffic delays as well as average and variance of bus delays. With BUSBAND there is an additional decrease in bus delays and passenger travel times with little effect on the rest of the traffic.

On-time performance of the buses will depend on how well the route travel times and ridership are estimated when the bus schedules are developed. Future efforts are planned on developing the bus schedules with consideration of ridership and traffic adaptive control.