2016 Boston Bicycle Counts
In 2016, we took our first steps to create an annual program using automated technology to count the number of people bicycling. During three days in September, we counted nearly 30,000 bike trips per day. In some locations during peak commuting times, bikes were more than one in six of all vehicles passing by. These trips represent a small percentage of the many trips people take by bike every year. We’ll use this data to better understand how many people are already biking in Boston, and what we can do to encourage more people to go by bike.
We made the switch to automated bike counts so that we could capture more, and better, data.
To identify best practices, we:
- researched available technologies,
- learned about potential drawbacks, and
- reviewed count programs in several sister cities.
With the annual automated counts program, we are improving the quality of our data by:
- collecting information about trips that happen outside of peak commuting hours
- diversifying the types of locations that we count
- standardizing the dates and hours of our counts, year-over-year, and
- ensuring that we do our counts, regardless of bad weather.
In June, we piloted 24 locations and used a variety of counting technologies. We used video technology to execute the full program in September.
In previous years, teams of volunteers counted the number of people they saw on bikes at a variety of locations. These counts took place during peak commute times on one of several days in September. While this information has been helpful, we knew it had real limitations:
- counts weren’t always done during the same day of the year and didn’t align year-over-year
- we missed everyone who was biking outside the peak commute times
- our volunteers, as steadfast as they were, could not be at every location each year, and
- we also didn’t want to send them out in very poor weather.
Better data helps us do better in planning and designing our bike network.
We’ll be able to develop factors to understand how many people are biking on certain types of routes in certain contexts. (“If X people bike on this street today, we can expect Y people to bike on a similar street today.”) We can begin to normalize other counts taken at other times of the year. (“If X people ride here on a colder morning in November, we can expect Y people to ride here on a warmer evening in June.”)
We may also be able to project how many people would ride on new routes or different facilities as those projects are being planned and designed. (“If we add a separated bike lane here, we can expect Z more people to bike on this route.”)
Better data also helps us in looking back and assessing our progress in promoting bicycling in Boston. We can better answer questions such as:
- Has ridership changed on key routes?
- Have we encouraged more people to ride by building a more connected network?
- Has the presence of new bike share stations influenced bike-riding on nearby streets?