Title

Scheduling Buses to Take Advantage of Transit Signal Priority

Document Type

Journal Article

Publication Date

2009

Subject Area

operations - scheduling, operations - traffic, infrastructure - bus/tram priority, infrastructure - bus/tram priority, infrastructure - bus/tram lane, infrastructure - bus/tram lane, infrastructure - traffic signals, policy - congestion, mode - bus

Keywords

Traffic signal priority systems, Traffic signal preemption, Traffic delay, Traffic congestion, Simulation, Schedules and scheduling, Running time, Preemption (Traffic signals), Optimization, Optimisation, Microsimulation, Headways, Gridlock (Traffic), Dwell time, Departure time, Computer simulation, Bus transit operations, Bus scheduling, Bus priority, Bus lanes

Abstract

Transit signal priority can improve bus operations when it is applied to a route without making any changes to its route design or management; however, benefits can be greater if service design and management policies are purposely altered to take advantage of transit signal priority. Service design issues include generating carefully constructed schedules for use with conditional priority and deciding whether to locate stops on the near side or far side of intersections. Management issues include using conditional priority as a means to give priority only to late buses and deciding whether to hold buses at bus stops until a scheduled departure time, in case conditional priority itself does not offer the desired level of operational control. An optimal level of aggressiveness in the running time schedule is sought as a balance between mean running time with and without priority, assuming that priority will be conditional. With a more aggressive schedule, buses will be late and therefore will request (and get) priority more often; however, a more aggressive schedule also offers less slack to compensate for random delays. Analysis uses both a spreadsheet-based simulation model and a traffic microsimulation model that account for random delays at dispatch and at traffic signals, the effect of crowding on dwell time, and choice of priority and operational control tactics. It is found that optimal performance uses conditional priority and holding, but with an aggressive schedule, leading to substantial reductions in mean running time, standard deviation of running time, headway irregularity, and crowding.