Placing detectors and deploying transit signal priority should be done to maximize benefits for transit vehicles while minimizing delay for other vehicles.

TSP bus and light rail deployments at intersections in San Francisco and Salt Lake City, Utah

Date Posted
02/27/2017
TwitterLinkedInFacebook
Identifier
2016-L00744

NCHRP Synthesis 447: Active Traffic Management for Arterials

Summary Information

Active traffic management (ATM) is the adaptation of facility configuration and controls in response to or anticipation of variations in demand, incidents, and weather to optimize facility operation. The objective of ATM strategies on arterials is to maximize cost-effectiveness of the facility. The purpose of this synthesis was to document the state of the practice associated with designing, implementing, and operating ATM on arterials. This was accomplished through a literature review of advanced ATM methods for arterials and an in-depth telephone survey of agencies that had indicated through a pre-screening process that they had a high degree of knowledge and experience implementing ATM on their arterial streets.

Lessons Learned

Transit Signal Priority at Two Agencies

The San Francisco Municipal Transportation Agency (SFMTA) uses TSP for both light rail and buses at many intersections in the city. TSP systems used in San Francisco include or have included: video-based detection, inductive loop technology, and infrared emitters/detectors. SFMTA switch to a wireless radio communications-based system over the next few years. The system will be GPS based and fully integrated with signal controllers, NextBus, and the transit management center. The system will know if a bus is behind schedule, ahead of schedule, loaded, and other factors, which will then be used intelligently by the signal controllers to prioritize the signals on the transit vehicles route. SFMTA is also setting up a peer-to-peer communication system that will operate mostly wirelessly so the traffic signal controllers can communicate with each other. When a transit vehicle arrives at one signal, it will communicate this downstream so timing can begin to be adjusted downstream for the bus’s arrival.



SFMTA had difficulty obtaining funding for new radios that will make the buses operate on the new integrated system. The agency also had institutional issues with the FCC and the requirement for 5.9 GHz for vehicle-to-traffic signal communications. This standard is so new that it is not fully vetted. There are also not a lot of vendors making radio equipment at this frequency range. Additionally, many traffic signal controllers needed to be updated to handle the new wireless technology. San Francisco has approximately 1,200 signals in total, and 600 of them are located on the high-priority transit routes. Currently, 350 have been upgraded and the other 250 are in the process of being upgraded.



The main measure of effectiveness SFMTA uses to determine the effectiveness of TSP is service reliability. The agency hopes the upgraded TSP system will improve reliability and make SFMTA a transit service the city can be proud of because it is extremely reliable.



The Utah Department of Transportation is implementing transit signal priority (TSP) on two corridors around Salt Lake City. In one corridor, TSP is being used for bus rapid transit (BRT). The BRT corridor has 10 to 15 intersections, with TSP implemented using a built-in feature on the signal controllers. Each BRT bus has an infrared transmitter broadcasting to the signal controllers. When the infrared transmitter gets within range of the signal, the signal controller extends the length of the green to accommodate the bus. The second corridor uses TSP on a section of the light rail system. This corridor uses inductive loops on the track bed at most intersections along the light rail route, except at major intersections where the disruption to operations was too great. The light rail TSP detects the train at the signal and performs different functions depending on where the signal is in the cycle. If the train is approaching near the end of the cycle, it will extend the green. If the train is waiting for the signal, the signal controller will give the train a green indication before the auto green indication. This allows the train to begin moving first (queue jump) so vehicles do not try to turn in front of the train. Another feature of the light rail system TSP is its use of special software to communicate the presence of a train between one to two signals downstream. This allows the downstream signals to know a train will be approaching soon so that it can improve progression through modes such as phase rotation. Because the light rail had been in place for several years without TSP, signal controllers were not set up for it and had to be replaced at those intersections where TSP was going to be allowed.



The next iteration of TSP intends to be based more on GPS and timing, these systems will be able to determine if a vehicle is ahead or behind schedule and the occupancy of a vehicle. If the transit vehicle is ahead, it will not get priority. If the vehicle is behind schedule, it will be given the priority needed to get it back on schedule. This should allow for a more finely tuned balanced between quickly serving transit vehicles while maintaining auto LOS.



Lessons Learned

 

  • Remain patient with the infrastructure. All systems have flaws and it is important to understand the system.
  • Placement of TSP activation detectors is important, ensure any priority maximizes benefit to transit vehicles while minimizing delay to others.
  • Not all signals need the same amount of priority. A more balanced approach for all users could be allowing less priority at major intersections and more at minor intersections. The theory was that a little more delay for the transit vehicle at major intersections is fine if they will move faster through the minor ones.
  • Do not forget about the associated maintenance. There is more than the initial capital cost of the system. If a system is not maintained, it will not work.