Person-Based Adaptive Priority Signal Control with Connected-Vehicle Information
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
The goal for transit signal priority (TSP) strategies is to improve the efficiency of urban transportation systems by promoting fast passage of system users. However, because conventional vehicle detection technologies require TSP strategies to be vehicle based, TSP may not lead to optimal results for person delay. This paper proposes a signal control model called PAPSCCI (person-based adaptive signal priority control with connected-vehicle information). First, by using vehicle speed and location information available from connected-vehicle technologies, the model explicitly computes individual vehicle delay. In this way the model avoids assumptions about vehicle arrivals, which often are inevitable in a delay calculation derived from a queuing model. Furthermore, in the model approach, the computation of delays for private vehicles is no different from that for public buses except in the priority level and unifies the two types of vehicles. With onboard passenger information, the PAPSCCI model computes person delay for every vehicle running through the intersection and offers a more accurate basis for person delay minimization. The performance of the PAPSCCI model is evaluated in a traffic simulation environment. Compared with the optimized timing from SYNCHRO, the PAPSCCI model produces 39%, 49%, and 30% decreases in bus passenger delays for one, two, and three conflicting bus routes, respectively. In addition, general automobiles experience about an 8% to 11% decrease in person delays, showing the potential of PAPSCCI as a general adaptive signal control model. Finally, a penetration rate study shows that the PAPSCCI model can consistently perform reasonably well even when only about 30% of vehicles are equipped with connected-vehicle technology.
