Dozens of major research labs dot the streets of Kendall Square, a Cambridge, Massachusetts, neighborhood in which MIT partially sits. But for Andres Sevtsuk’s City Form Lab, the streets of Kendall Square themselves, and those in other cities, are subjects for research.
Sevtsuk is an associate professor of urban science and planning at MIT and a leading expert in urban form and spatial analysis. His work examines how the design of built environments affects social life within them. The way cities are structured influences whether street-level retail commerce can thrive, whether and how much people walk, and how much they encounter each other face to face.
“City environments that allow us to get more things done on foot tend to not only make people healthier, but they are more sustainable in terms of emissions and energy use, and they provide more social encounters between different members of society, which is fundamental to democracy,” Sevtsuk says.
However, many things Sevtsuk studies do not come with much pre-existing data. While some aspects of cities are studied extensively — vehicle traffic, for instance — fewer people have studied how urban planning affects walking and cycling, which most city governments seek to increase.
To counter this trend, several years ago Sevtsuk and some research assistants began studying foot traffic in several cities, as well as Kendall Square — how much people walk, where they go, and why. Most urban walking trips are destination-driven: People go to offices, eateries, and transit stops. But a lot of pedestrian activity is also recreational and social, such as sitting in a square, people-watching, and window-shopping. Eventually Sevtsuk emerged with an innovative model of pedestrian activity, which is based around these spatial networks of interaction and calibrated to observed people counts.
He and his colleagues then scaled up their model and took it to major cities around the world, starting with the whole downtown of Melbourne, Australia. The model now includes detailed street characteristics — sidewalk dimensions, the presence of ground floor businesses, landscaping, and more — and Sevtsuk has also helped apply it to Beirut and, most recently, New York City.
The project is typical of Sevtsuk’s research, which creates new ways to bring data to urban design. In 2023, Sevtsuk and his colleagues also released a novel open-source tool, called TILE2NET, to automatically map city sidewalks from aerial imagery. He has even studied interactions on the MIT campus, in a 2022 paper quantifying how spatial relatedness between departments and centers affects communications among them.
“Applying spatial analytics to city design is timely today because when it comes to cutting carbon emissions and energy consumption, or improving public health, or supporting local business on city streets, they relate to how cities are configured,” Sevtsuk says. “Urban designers have historically not been very focused on quantifying those effects. But studying these dynamics can help us understand how social interactions in cities work and how proposed interventions may impact a community.”
For his research and teaching, Sevtsuk received tenure at MIT earlier this year.
Growing and living in cities
Sevtsuk is originally from Tartu, Estonia, where his experiences helped attune him to the street life of cities.
“I do think where I come from enhanced my interest in urban design,” Sevtsuk says. “I grew up in public housing. That very much framed my appreciation for public amenities. Your home was where you slept, but everything else, where you played as a child or found cultural entertainment as a teenager, was in the public sphere of the city.”
Initially interested in studying architecture, Sevtsuk received a BArch degree from the Estonian Academy of Arts, then a BArch from the Ecole d’Architecture de la Ville et des Territoires, in Paris. Over time, he became increasingly interested in city design and planning, and enrolled as a master’s student at MIT, earning his SMArchS degree in 2006 while studying how technology could help us better understand urban social processes.
“MIT had a very strong research orientation for even masters-level students,” Sevtsuk says. “It is famous for that. I came because I was drawn to the opportunity to get hands-on into research around city design.”
Sevtsuk stayed at MIT for his doctoral studies, earning his PhD in 2010, with the late William Mitchell as his principal advisor. “Bill was interested in the influence of technology on cities,” says Sevtsuk, who appreciated the wide-ranging intellectual milieu that sprang up around Mitchell. “A lot of fascinating and intellectually experimental people gravitated around Bill.”
With his PhD in hand, Sevtstuk then joined an MIT collaboration at the new Singapore University of Technology and Design, a couple of years after it first opened.
“That was a lot of fun, building a new university, and we were teaching the first cohort and first courses,” Sevtsuk says. “It was an exciting project.”
Living in Asia also helped open doors for some hands-on research in Singapore and Indonesia, where Sevtsuk worked with city governments and the World Bank on urban planning and design projects in several cities.
“There was not a lot of data, and yet we had to think about how spatial analyses could be deployed to support planning decisions,” Sevtsuk says. “It forced you to think how to apply methods without abundant data in the traditional sense. In retrospect some of the software around pedestrian modeling we developed was influenced by these constraints, from understanding the minimum data inputs needed to capture people’s mobility dynamics in a neighborhood.”
From Melbourne to the Infinite Corridor
Returning to the U.S., Sevtsuk took a faculty position at Harvard University’s Graduate School of Design in 2015. He then joined the MIT faculty in 2019.
Throughout his career, Sevtsuk’s projects have consistently added insight to existing data or created all-new repositories of data for wider use. His team’s work in Melbourne leveraged a rare case of a city with copious pedestrian data of its own. There, Sevtsuk found the model not only explained foot traffic patterns but could also be used to forecast how changes in the built environment, such as new development projects, could affect foot traffic in different parts of the city.
In Beirut, the modeling work on improving community streets is part of post-disaster recovery after the Beirut port explosion of 2020. In New York, Sevtsuk and his colleagues are studying the largest pedestrian network in the U.S., covering all five boroughs of the city. The TILE2NET project, meanwhile, provides information for planners and experts in an area — sidewalk mapping — which most places do not have data on either.
When it came to studying the MIT campus, Sevtsuk brought new a new approach to a subject with an Institute legacy: An earlier campus professor, Thomas Allen of the MIT Sloan School of Management, did pioneering research about workspace design and collaboration. Sevtsuk and his team, however, looked at the larger campus as a network.
Linking spatial relations and email communication, they found that not only does the level of interaction between MIT departments and labs increase when those units are spatially closer to each other, but it also increases when their members are more likely to walk past each other’s offices on their daily routes to work or when they patronize the same eateries on campus.
Urban design for the people
Sevtsuk thinks about his own work as being not just data-driven but part of a larger refashioning of the field of urban design. In American cities, urban design may still be associated with the large-scale redevelopment of neighborhoods that took place in the first few postwar decades: massive freeways tearing through cities and dislocating older business districts, and large housing and office projects undertaken in the name of modernization and tax revenue increases but not in the interests of existing residents and workers. Many of these projects were disastrous for urban communities.
By the 1960s and 1970s, urban planning programs around the country attempted to quell the inadequacy of large-scale urban design and instead focused on the social and economic needs of communities first. The role of urban design was somewhat sidelined in this transition. But instead of giving up on urban design as a tool for community improvement, Sevtsuk thinks that planning and urban design research can help uncover the important ways in which design can support communities in their daily lives as much as community development initiatives and policies can.
“There was a turn in the field of planning away from urban design as a central area of focus, toward more sociologically grounded community-driven approaches,” Sevtsuk says. “And for good reasons. But during these decades, some of the most anti-urban, car-oriented, and resource-intensive built environments in the U.S. were created, which we now need to deal with.”
He adds: “In my work I try to quantify effects of urban design on people, from mobility outcomes, to generating social encounters, to supporting small local businesses on city streets. In my research group we try to connect urban design back to the qualities that people and communities care about. Faced with the profound climate challenges today, we must better understand the influence of urban design on society — on carbon emissions, on health, on social exchange, and even on democracy, because it’s such a critical dimension.”
A dedicated teacher, Sevtsuk works with students with broad backgrounds and interests from across the Institute. One of his main classes, 11.001 (Introduction to Urban Design and Development), draws students from many departments — including computer science, civil engineering, and management — who want to contribute to sustainable and equitable cities. He also teaches an applied class on modeling pedestrian activity, and his research group draws students and researchers from many countries.
“What resonates with students is that when we look closely at the complex organized systems of cities, we can make sense of how they work,” Sevtsuk says. “But we can also figure out how to change them, how to nudge them toward collective improvement. And many MIT students are eager to mobilize their amazing technical skills towards that quest.”
