The iconic scene from the 1969 film "Midnight Cowboy," where Dustin Hoffman’s character, Ratso Rizzo, famously berates an oncoming taxi with the declaration, "I’m walking here!", encapsulates a perennial tension in urban environments: the complex and often fraught interaction between pedestrians and vehicular traffic. For decades, urban planners and government officials have meticulously tracked the flow of cars, a quantifiable metric that directly influences infrastructure decisions and economic development. However, the movement of people on foot, a fundamental aspect of city life, has largely remained an unmapped territory, an invisible force shaping urban landscapes. Now, a groundbreaking research initiative from the Massachusetts Institute of Technology (MIT) has produced the first comprehensive, routable dataset of sidewalks, crosswalks, and footpaths for all of New York City. This ambitious mapping project represents a significant leap forward, offering the first complete model of pedestrian activity in any major U.S. city and promising to revolutionize how urban spaces are designed and managed.
The implications of this new model are far-reaching. It provides city planners with an unprecedented tool to inform decisions regarding pedestrian infrastructure and public space investments. By understanding precisely where and when people walk, officials can better allocate resources to enhance safety, improve accessibility, and create more vibrant urban environments. Furthermore, the model illuminates how development decisions, from new building projects to changes in street design, can significantly impact non-motorized travel. Perhaps most critically, it allows for the identification of specific locations across the city where high pedestrian volumes intersect with significant safety hazards, such as traffic crashes, highlighting areas in urgent need of upgrades and interventions.
"We now have a first view of foot traffic all over New York City and can check planning decisions against it," stated Andres Sevtsuk, an associate professor in MIT’s Department of Urban Studies and Planning (DUSP) and the lead researcher on the study. "New York has very high densities of foot traffic outside of its most well-known areas." This observation challenges the conventional wisdom that dense pedestrian activity is confined to iconic business districts.
A key finding from the MIT research is that while Manhattan, particularly Midtown, indeed exhibits the highest foot traffic per block, the city’s other boroughs also contain substantial stretches of pedestrian-heavy sidewalks that may be underserved by current investment strategies. This suggests a potential Manhattan-centric bias in urban planning and infrastructure development, overlooking the needs and realities of pedestrian life in Queens, Brooklyn, and the Bronx.
"Midtown Manhattan has by far the most foot traffic, but we found there is a probably unintentional Manhattan bias when it comes to policies that support pedestrian infrastructure," Sevtsuk elaborated. "There are a whole lot of streets in New York with very high pedestrian volumes outside of Manhattan, whether in Queens or the Bronx or Brooklyn, and we’re able to show, based on data, that a lot of these streets have foot-traffic levels similar to many parts of Manhattan." This revelation could prompt a recalibration of resource allocation, ensuring more equitable development and improved walkability across all five boroughs.
Beyond mapping pedestrian flows, the MIT model introduces a novel approach to analyzing pedestrian safety. It quantifies vehicle crashes involving pedestrians not merely as raw totals but on a per-pedestrian basis. This shift in perspective is crucial for understanding actual risk.
"A lot of cities put real investments behind keeping pedestrians safe from vehicles by prioritizing dangerous locations," Sevtsuk explained. "But that’s not only where the most crashes occur. Here we are able to calculate accidents per pedestrian, the risk people face, and that broadens the picture in terms of where the most dangerous intersections for pedestrians really are." This nuanced analysis moves beyond simply counting accidents to understanding the likelihood of a pedestrian encountering danger in a given area, a vital distinction for effective safety interventions.
The comprehensive study, titled "Spatial Distribution of Foot-traffic in New York City and Applications for Urban Planning," was published today in the prestigious journal Nature Cities. The research team comprised Sevtsuk, the Charles and Ann Spaulding Associate Professor of Urban Science and Planning in DUSP and head of the City Design and Development Group; Rounaq Basu, an assistant professor at Georgia Tech; Liu Liu, a PhD student at the City Form Lab in DUSP; Abdulaziz Alhassan, a PhD student at MIT’s Center for Complex Engineering Systems; and Justin Kollar, a PhD student at MIT’s Leventhal Center for Advanced Urbanism in DUSP.
Charting the Unseen: The Genesis of a Pedestrian Model
The current research builds upon years of work by Sevtsuk and his colleagues who have been dedicated to charting and modeling pedestrian traffic globally, from the bustling streets of Melbourne to the academic hub of MIT’s Kendall Square neighborhood in Cambridge, Massachusetts. While many cities collect some pedestrian count data, it is often fragmented and insufficient for comprehensive analysis. Moreover, while vehicle traffic impact assessments are standard for new development plans, the impact on pedestrians is rarely scrutinized with the same rigor.
New York City, however, presents a unique case study. It boasts a dedicated portion of its Department of Transportation (DOT) focused on pedestrian issues, and notably, approximately 41 percent of all trips within the city are made on foot, compared to just 28 percent by vehicle. This ratio is likely the highest among major U.S. cities, underscoring the critical importance of understanding pedestrian behavior.
To calibrate their model, the MIT team leveraged pedestrian count data meticulously recorded by the New York City DOT in 2018 and 2019. This data captured pedestrian activity on up to 1,000 city sidewalk segments during weekdays and roughly 450 segments on weekends. The model, which integrates a wide array of influencing factors, was rigorously tested against this real-world data, ensuring its accuracy and reliability. Once calibrated, the model’s predictive power allowed researchers to extrapolate foot-traffic estimates across the entire city, extending beyond the specific points where direct observations were made.
Mapping the City’s Pulse: Foot Traffic Patterns Revealed
The findings from the calibrated model paint a detailed picture of pedestrian movement across New York City. During the evening peak of foot traffic, Midtown Manhattan emerged as the undisputed epicenter, with an average of 1,697 pedestrians per sidewalk segment per hour. This figure dwarfs other areas, highlighting its status as a major hub for work, entertainment, and retail. The Financial District in lower Manhattan followed, registering a substantial 740 pedestrians per hour, with Greenwich Village securing the third position at 656 pedestrians per hour.
However, the data also revealed that other parts of Manhattan experience significantly lower pedestrian densities. For instance, Morningside Heights and East Harlem recorded pedestrian counts of 226 and 227 per block per hour, respectively. These figures are comparable to, and in some cases lower than, those observed in neighborhoods outside of Manhattan. Brooklyn Heights, for example, saw an average of 277 pedestrians per sidewalk segment per hour. In the Bronx, University Heights registered 263 pedestrians per hour, while areas like Borough Park in Brooklyn and the Grand Concourse in the Bronx averaged 236. Even a section of Queens in the Corona area recorded 222 pedestrians per hour, with many other locations in these boroughs consistently exceeding 200 pedestrians per hour.
This distribution of foot traffic suggests that while Manhattan’s central business districts are undeniably busy, other boroughs possess significant pedestrian activity that may not be adequately supported by current infrastructure. The model’s ability to overlay different types of pedestrian journeys for various time periods further enriches this understanding. It demonstrates that morning commutes are primarily driven by proximity to jobs and transit stops. However, as the day progresses into midday and evening, travel patterns become more diverse, reflecting a search for amenities, social engagements, and recreational activities.
"Because of jobs, transit stops are the biggest generators of foot traffic in the morning peak," noted Liu Liu, a PhD student involved in the research. "In the evening peak, of course people need to get home too, but patterns are much more varied, and people are not just returning from work or school. More social and recreational travel happens after work, whether it’s getting together with friends or running errands for family or family care trips, and that’s what the model detects too." This insight is invaluable for urban planners seeking to design public spaces that cater to a wider range of daily activities and needs.
Re-evaluating Safety: Beyond Raw Numbers
The safety analysis conducted by the MIT team offers a critical new perspective on pedestrian risk. The model reveals that danger is not solely concentrated in intersections with the highest raw number of accidents. Instead, many parts of the city present a higher risk on a per-pedestrian basis, even if their total crash numbers are lower.
"Places like Times Square and Herald Square in Manhattan may have numerous crashes, but they have very high pedestrian volumes, and it’s actually relatively safe to walk there," explained Rounaq Basu, an assistant professor at Georgia Tech and a co-author of the study. "There are other parts of the city, around highway off-ramps and heavy car-infrastructure, including the relatively low-density borough of Staten Island, which turn out to have a disproportionate number of crashes per pedestrian." This finding is particularly significant, as it suggests that interventions focused solely on high-crash locations might miss areas where pedestrians face a greater individual risk due to a less forgiving built environment.
This granular understanding of risk allows for more targeted and effective safety improvements. By identifying streets and intersections where the ratio of crashes to pedestrian volume is high, city officials can prioritize investments in traffic calming measures, improved pedestrian crossings, and better street lighting in areas that might otherwise be overlooked.
A Blueprint for the Nation: Expanding the Model’s Reach
The MIT model is poised to have a significant impact on policy and planning within New York City, as officials are actively engaged with the research team and exploring its applications. However, the ambition of this project extends far beyond the five boroughs. The researchers believe their methodology can be replicated and adapted for cities and towns across the United States.
Currently, the MIT team is collaborating with municipal officials in two other locations. In Los Angeles, city leaders are utilizing the model to enhance pedestrian and public transit mobility, not only for daily commutes but also in preparation for the influx of visitors expected for the 2028 Summer Olympics. This application highlights the model’s utility in planning for major events and large-scale urban transformations.
Simultaneously, the state of Maine is working with the MIT team to assess pedestrian movement across over 140 of its cities and towns. This initiative aims to provide a comprehensive understanding of pedestrian needs and identify opportunities for upgrades and safety improvements statewide. Sevtsuk expresses hope that these examples will inspire other municipalities to adopt similar approaches to pedestrian data analysis, contributing to the broader goal of decarbonizing cities and addressing the historical overemphasis on automobile-centric urban planning.
"I hope this can inspire other cities to invest in modeling foot traffic and mapping pedestrian infrastructure as well," Sevtsuk concluded. "Very few cities make plans for pedestrian mobility or examine rigorously how future developments will impact foot-traffic. But they can. Our models serve as a test bed for making future changes." The advent of this comprehensive pedestrian model marks a pivotal moment in urban planning, offering a data-driven pathway toward creating safer, more equitable, and more livable cities for all residents.
