Urban planners can now pinpoint exactly where in a city increased housing density will make the biggest difference on shortening car commutes. That’s the promise of a new study in which researchers used urban big data and AI to hone densification strategies for six cities around the world.

It’s pretty well established that the best way to get city dwellers to drive less is to change characteristics of the built environment such as city shape, size, and density, rather than simply hectoring them to reduce their carbon emissions. But past studies have been unable to establish causal relationships between specific aspects of urban form and car travel. They also miss the smaller picture—neighborhood-level differences—and the bigger one—how these patterns differ across various regions of the world.

The new study puts all these pieces together for the first time, the researchers say.  They gathered 10 million data points on morning car commute distances from six metropolises worldwide: Berlin, Boston, Los Angeles, the San Francisco Bay Area, Rio de Janeiro, and Bogotá.

Using a machine learning algorithm, the researchers analyzed how different aspects of urban form —the distance to jobs and the city center, population density, income, and the pattern and interconnectedness of streets—influence the length of car commutes in different neighborhoods in each city.

Because all of the data points are from people traveling to work by car, the study can’t say anything about what makes people abandon their cars entirely and commute to work by bike, on foot, or via public transit. But it does provide hints about how to reduce the length of trips that are made by car.

Across cities, the distance to jobs and the city center matters more than population density or street connectivity in determining the length of car commutes. But where and how strongly these effects occur varies from city to city, revealing new urban planning strategies. What’s more, metropolises themselves aren’t monoliths: some policies are helpful in particular parts of a city, but not city-wide.

“The importance of high access implies that new housing should be located as close to the center as possible, highlighting the relevance of compact development,” the researchers write. “While this strategy is relevant for all cities, it requires context-specific adaptation.”

In urban regions with a single, defined core such as Berlin and Boston, the best place to increase housing density is in a ring around the center where there’s room for infill development but the city center is still easily accessible. In the case of Boston, for example, this zone occurs about 10-21 kilometers from the city center.

Meanwhile, in cities with multiple centers like Los Angeles and the San Francisco Bay Area, the best strategy is to build more housing in areas with high concentration of jobs.

In each city, the researchers identified specific areas where lack of density is a bigger factor than distance to the center in increasing car commutes. Those are the places where strategic densification will make the biggest difference, the researchers argue.

Using a similar methodology to analyze other modes of transport and trips throughout the day, not just during the morning commute, would build a fuller picture of opportunities to reduce carbon emissions from urban transport, the researchers say.

Source: Wagner F. et al. “Refining urban typologies: causal insights into urban form, car commuting, and related CO₂ emissions.” Environmental Research Letters 2026.

Image: Getty Images for Unsplash+.