Project description

CityWays is a research to reveal human’s recreational patterns through visualizing datasets from self-tracking applications, quantifying the effects of temperature, precipitation, and other environmental factors. Through finding factors that affect pedestrian activities, we also found implications for street design and zoning policies, possibly leading to a re-definition of well-known static metrics of walkability.

Tens of millions of self-tracking apps are installed annually by people eager to monitor themselves – from how many calories they eat to how many hours they sleep to how many steps they take in the course of a day. While these devices and applications help people realize their fitness goals, researchers, including me at the MIT Senseable City Lab, in collaboration with Liberty Mutual, have found another use for the Big Data they produce. As part of a new study called CityWays, we have analyzed anonymized data to better understand activity patterns in cities, including how pedestrians, runners, and cyclists move in the urban environment. The innovative approach and result of this project shows the reason why CityWays is dedicated to the improvement of people’s daily experience in cities. 

By comparing the dataset and external factors especially weather visualized on the CityWays web application, users (urban planners/ designers/researchers) can have their own suppositions and inquiries about the human’s seasonal trip trends of cities, particularly in Boston and San Francisco.

Through exploring the landing website, users can have an overall understanding of this research. By investigating the web application that is comprised of two-part, map, and dashboard, they can scrutinize data having deeper insights.

What makes this project innovative?

Currently, fitness tracking applications are present in 60% of smartphones, and constitute a rapidly growing market, with sales of $800 million projected by 2020. They are also behind many ad-hoc devices, such as fitness electronic gear contained in watches and bracelets, among others. While the individual benefits of such apps are well known, the aggregated data they produce has never been studied to better understand cities – at least until now.

In addition, until recently, movement data was difficult to obtain and required tedious and expensive surveys. But, information from self-tracking apps is a game changer because it allows people to discover how a diverse group of people from different communities move around an urban space – and what are the key factors influencing it. Thus, as the first step, we challenged in finding factors that affect pedestrian activities to redefine metrics of walkability in a city.

Using data collected from personal tracking applications helps us better understand how people move around cities - and can lead to more human-centered urban design, grounded in actual data.

What was the impact of your project? How did you measure it?

Researchers focused initially on two cities, Boston and San Francisco, investigating thousands of walking, running and cycling trips recorded with self-tracking apps over the course of more than one year. As a first step, they were able to quantify the effects of temperature, precipitation, and other environmental factors on outdoor activities. Not surprisingly, Bostonians are resistant to cold weather, and keep moving when temperature falls below -15oC —although males seem to be more cold-resistant.

In Boston and San Francisco, people tend to travel within the central business districts, mostly walking between transport hubs, jobs, and commercial areas to complete short errands. Runners prefer open spaces—in Boston, for example, the Charles River banks are highly frequented, as are the Chestnut Hill reservoir and Jamaica Pond. In San Francisco, runners populate the Golden Gate Park and bay area. Finally, in these cities, bicycle trips are longer and span outside of the cities, crossing the Golden Gate bridge in San Francisco, and along the Minute Man trail in Boston.

Another interesting finding is that presence of restaurants and grocery stores seems to have a stronger effect on pedestrian activity than the presence of street elements that are explicitly designed for pedestrians, such as benches, trees, or sidewalks. Thus, the study suggests that the destination of your trip, especially if related to eating, is relatively more important than the amenities available along the road. This finding can have implications on street design and zoning policies, possibly leading to a re-definition of well-known static metrics of walkability such as the walk score.

These results and insights of the novel project reveal human's recreational patterns clearly thanks to datasets collected with self-tracking applications even though, in general, integrating movement data is an expensive and difficult process. This study will be a foundation to improve the better people’s living standard in cities using mobility data.

Source and methodology

We collected data from self-tracking apps over the course of more than one year. The names of the sources/ organizations are confidential.

Technologies Used

Javascript, d3js, HTML, CSS, Mapbox, Leaflet, Python, Tableau, and Excel were used for the front-end development/design, data cleaning, and analysis.

Project members

Senseable City Lab: Carlo Ratti, Hyemi Song, Fábio Duarte, Ruixian Ma, Anthony Vanky, Paolo Santi
Liberty Mutual: Santosh Verma,Theodore Courtney


Additional links

Project owner administration

Contributor username


Click Follow to keep up with the evolution of this project:
you will receive a notification anytime the project leader updates the project page.