Transit signal priority deployment along a 4 mile corridor can reduce bus travel times by 5 percent.

Simulation results of deploying TSP along the Columbia Pike corridor in northern Virginia.

Date Posted
04/29/2014
Identifier
2014-B00911
TwitterLinkedInFacebook

Evaluation of Potential Transit Signal Priority Benefits Along a Fixed-Time Signalized Arterial

Summary Information

This study used the INTEGRATION simulation model to estimate the impact of implementing transit signal priority (TSP) on a 3.95 mile section of Columbia Pike in Arlington, Virginia. Baseline corridor and cross street traffic flows were determined from field data collected between June 12 and 14, 2000. Saturated traffic flow parameters were estimated based on corridor geometry.

The model was designed to reflect the geometric location and configuration of multiple bus stops and intersections on the corridor. The transit signal priority scheme gave priority to buses traveling on Columbia Pike or entering the corridor from major cross streets. The signal priority logic enhanced the existing optimized fixed-time signal control system by enabling transit vehicles to receive a five second green extension if approaching an intersection at the end of a green cycle. The model considered vehicle movements only, and did not account for pedestrian clearance intervals or the impact of emergency vehicle signal preemption. In addition, if two transit vehicles called for priority simultaneous at the same intersection, no priority was given.

Methodology
The simulation model was run to determine the impacts of TSP on AM peak and midday traffic flows. Two time periods were examined: 7-9AM (AM Peak, heavily directional travel towards Washington, D.C.) and 11AM-1PM (Mid-day, off peak, balanced travel). The two periods were divided into 15 minute intervals for the model runs. Four scenarios were examined for the AM Peak period:
  • Base case - no priority
  • TSP for express buses only
  • TSP for express and other buses traveling along the Columbia Pike
  • TSP for all buses at the intersections in the corridor, regardless of direction.
For the Midday time period, only the base case and full priority scenarios were tested. Due to the stochastic nature of the model, dozens of runs for each scenario were averaged to get the results provided in the Findings section below.

Peak period person delays were calculated based on the assumption that passenger cars carried 1.2 persons per vehicle, and transit vehicles carried 23 persons per vehicle. During non-peak hours, transit vehicle were assumed to carry 16 persons per vehicle.

Findings
For the AM peak, the initial simulation results indicated that the provision of transit priority at all intersections between Dinwiddie and Queen was generally beneficial to buses.
  • For the Columbia Pike express buses, statistically significant reductions in travel time (-2.3 to -2.5 percent), delays (-3.7 to -4.1 percent), vehicle stops (-1.3 to -2.7 percent), and fuel consumption (-1.1 to -2.7 percent) were obtained when priority was provided exclusively to them or in conjunction with other buses running along the arterial.
  • Regular Columbia Pike buses also experienced statistically significant reductions in travel time (-4.8 percent), delay (-7.6 percent), stops (-1.8 percent), and fuel consumption (-1.9 percent) when priority was granted to them.
However, other vehicles experienced 18 percent increases in average travel time under the scenario where all buses were granted priority, so the simulation was examined more closely to see if there were specific intersections that were the major causes of this increase in travel time. After determining that two intersections (Columbia Pike & George Mason; Columbia Pike & Walter Reed) carry significant cross-street traffic and were the major causes of additional travel time along the corridor, the simulation was run again without TSP deployed at those two intersections. This change marginally reduced the benefits to buses but significantly reduced the negative network-wide impacts of transit priority.
  • In this case, maximum increases in travel time, delays, stop and fuel consumption incurred by the general traffic were reduced from 19.3 to 5.7 percent, 26.4 to 8.7 percent, and 4.7 to 1.8 percent, and 4.8 to 2.4 percent, respectively.
Goal Areas
Deployment Locations