Person-based Adaptive Priority Signal Control with Connected-vehicle Information

Abstract

This thesis proposes a TSP (transit signal priority) strategy of person-based adaptive priority signal control with connected-vehicle information (PAPSCCI). By minimizing the total person delay at an isolated intersection, PAPSCCI can assign signal priorities to transit vehicles due to their high occupancies, while minimize the negative impact to the auto traffic. With the accurate vehicle information provided by connected-vehicle technology, PAPSCCI can estimate person delay for each passenger directly and form a MILP (mixed-integer linear program) for the optimization.

Performances of PAPSCCI were evaluated through simulations. Results show decreases of both vehicle delay and person delay of all vehicle types when there are up to three bus routes running through the intersection.

How different penetration rates of the connected-vehicle technology affect the performance of the PAPSCCI were tested. Necessary revisions were made to the PAPSCCI model considering different penetration rates. Results show that the effectiveness of PAPSCCI worsens with the lowering of penetration rate. The delay improvements, however, were still promising when the penetration rate is above 40%.

PAPSCCI model were also developed and tested with communication range of 2000 m, 1000 m, 500 m and 250 m. Expect that the 1000 m case has the best delay improvements after PAPSCCI optimization, the effectiveness of the model worsens when the communication range getting smaller. Even when the communication range is down to 250 m, PAPSCCI can still reduce the delay for all vehicle types.