Transit Trips with Bunching or Gaps Between Vehicles

Detail Transportation Icon

San Francisco Municipal Transportation Agency


Target: Less than 1.8% bunching and 8.8% gaps on the Rapid Network (FY18 goal)

Result: 5.9% bunching and 17% gaps on the Rapid Network

The regularity of vehicle arrivals is the most important indicator of customer experience and reliability for the Muni Rapid Network, which carries almost 70 percent of all passengers and forms the backbone of Muni’s network, and enables the San Francisco Municipal Transit Agency (SFMTA) to track its ability to deliver frequent, reliable service. Data on where bunching or gaps in transit service occur also helps the SFMTA understand where potential transit priority projects will be most effective.

In addition, regularity of vehicle arrivals is the best measure of customer experience with transit reliability. Most Muni passengers ride routes that are scheduled to run frequently, and as such are unlikely to consult a printed schedule when planning their transit trips. For frequent routes, what is most important for customer experience is that the time between buses and trains (“headway”) is regular and close to the headways in the schedule. For example, if every single trip on the 5-Fulton were exactly 6 minutes late, on-time performance would be 0 percent (not a single bus arrived at the scheduled time), but customer experience with wait times would be excellent (the time between buses was exactly as scheduled, meaning no one waited longer than the scheduled headway).

Measuring bunching and gaps shows how Muni is performing on providing regular, reliable service on frequent lines. Regularity of vehicle arrivals has other ancillary benefits aside from minimizing passenger wait time. When buses are evenly spaced, crowding is also reduced, since gaps in service exacerbate crowding on the first bus at the end of a gap in service.


Note: January and February 2017 on-time performance, gaps, and bunching data cannot be reported due to a network issue resulting from the phase-out of the 2G cellular network. This network issue limited NextBus predictions and prevented systemwide on-time performance data from being collected.


Muni Forward is the SFMTA’s program of service changes and transit priority street projects to improve safety and reliability, and establish a Rapid Network. Gaps increased in 2015 due to service increases starting in Spring 2015 as well as the hiring of more than 700 new operators. However, perhaps in part due to the service increases, bunching has slightly increased as more service is being delivered. Service increases were implemented again in Spring 2016, and other Muni Forward capital projects, such as the 14-Mission Rapid Project, have been implemented to improve travel speed, reliability, and safety.

Going forward, ongoing fleet replacement is expected to improve vehicle reliability and contribute to reduced bunching and gaps as less service is disrupted by fewer mechanical failures. More Muni Forward service improvements and transit priority projects are proposed for routes across the city, which should also continue to improve performance.

How Performance is Measured

The SFMTA monitors the regularity of vehicle arrivals by measuring the actual arrival times at designated points along transit routes and comparing them to the scheduled headways (time between vehicles). Time point arrival data from the NextBus system are used to compare actual versus scheduled headways (time between vehicles). Bunching occurs if a vehicle arrives less than two minutes after a preceding vehicle, or less than one minute for routes with vehicles scheduled to arrive every 5 minutes or less. A gap occurs if a vehicle arrives 5 or more minutes later than the scheduled headway.

The Rapid Network includes the 5R-Fulton Rapid, 7R-Haight/Noriega Rapid, 9R-San Bruno Rapid, 14R-Mission Rapid, 28R-19th Avenue Rapid, 38R-Geary Rapid, J-Church, KT-Ingleside/Third Street, L-Taraval, M-Ocean View and N-Judah lines.

The number displayed on the scorecard page represents a fiscal year average of the values in the charts above.

Additional Information


Please visit DataSF for the scorecard data.