Development and laboratory testing of an intelligent approach for providing priority to buses at traffic-signalized intersections

The purpose of this dissertation was to develop and test in a laboratory setting an intelligent algorithm for providing priority to buses at traffic signalized intersections without disrupting progression. The algorithm used bus position information to predict when in the cycle a bus would arrive at the bus stop and stop line of a signalized intersection. It also automatically determined whether or not a bus needed priority. The strategy used to provide priority was selected based on the estimated arrival time of the bus at the stop line. In addition to using the phase extension and early return strategies, the buses could also be granted priority by inserting a special bus phase into the normal background cycle plan, a strategy not commonly used in existing bus priority systems.; Three hardware-in-the-loop simulation studies were performed to evaluate the effectiveness of the algorithm with real traffic signal controllers. The first simulation study was designed to verify that the algorithm was functioning as designed at three volume-to-capacity levels: 0.5, 0.8, and 0.95. The algorithm was able to provide priority to buses at all three signalized intersections about 92 to 95 percent of the time, with most failures caused by errors in the traffic simulation model. The second study was designed to evaluate the impacts of using the algorithm on both bus and non-transit vehicles at the three volume-to-capacity levels. Significant reductions in bus travel times were achieved at all dim volume-to-capacity levels by using the algorithm. The algorithm had little effect on the non-priority traffic except at the 0.95 volume-to-capacity level. A break-even analysis showed the algorithm was cost-effective (in terms of passenger delay savings) up to a volume-to-capacity level of 0.87 for fully loaded regular transit buses and 0.90 for fully loaded articulated buses. In the third simulation study, the researcher found that there was little difference in the performance of the algorithm at intersection spacings of 782, 391, and 195.5 meters (2565, 1283, and 691 feet). From these studies, the researcher concluded that the intelligent bus priority concept presented in this dissertation was viable for continued expansion and potential field operations.