In 2017, Metro began a pilot where it allowed employee shuttles from Microsoft and Seattle Children’s Hospital to use nine public bus stops scattered around Seattle, in exchange for a fee. Aside from money, sharing stops might improve inter-system transfers and foot traffic. But beyond global concerns about how the shuttles impact public transit ridership, the local concern is that shuttles might cause delays at these stops. A UW/UC Davis study used Metro’s real-time data to measure the impact of the pilot on reliability data six weeks before and after the pilot began.
You can see that none of these stops are particularly saturated to begin with:
The quickest summary is that shuttles didn’t make much of a difference at all. The raw on-time data suggest things got a bit worse, but that data did not control for other reliability issues. When the study applies those controls, the effect disappears. When the study applied advanced statistical techniques to individual stops, they found a small negative impact on Stop B. Notably, this stop has only one bus length of space.
The study recommends expanding the program at other “low volume” stops. It’s good news that sharing stops provides a few benefits without causing significant problems. But it’s intuitive that quiet stops could stand to take a few more buses. It would be far more interesting, if riskier, to try this out and some of the really busy stops in the system and understand what the limits of this program are.