Cities are increasingly lowering speed limits to 30 km/h to enhance road safety, reduce noise pollution, and promote sustainable modes of transportation, such as walking and cycling.
In Slowing by Design, we used artificial intelligence to analyze millions of street view images and vehicle mobility data across Milan, Amsterdam, and Dubai—revealing that street design, in synergy with signage, plays a critical role in shaping how fast people drive.
Our findings reveal that narrow, enclosed streets tend to slow drivers down— while wide roads, open visibility, and expansive sky views encourage higher speeds.
Street view images from Dubai
High Compliance Streets = Slower Speeds
Low Compliance Streets = Faster Speeds
In Slowing by Design, we introduce a scalable method for
understanding urban speed compliance by combining computer vision
and large-scale mobility data. Using artificial intelligence to
analyze over 51 million vehicle telemetry points and thousands of
street view images across Milan, Amsterdam, and Dubai, we identify
the spatial features that influence how fast people drive.
Our findings reveal that street design, in synergy with signage,
is a major influence on speed compliance. Roads with narrow lanes,
dense buildings, and limited sky visibility lead to slower
driving, while wide, open streets can lead to higher speeds.
We developed an AI model trained on street view imagery and
spatial data that can predict speed limit compliance based on
built environment features. This tool provides urban planners with
actionable insights to design more effective interventions for
traffic calming, optimize traffic management strategies, and
create safer, more livable cities.
Orsi G., Venverloo T., La Grotteria A., Fugiglando, U., Duarte F., Santi P., Ratti C. (2025). Street design and driving behavior: evidence from a large-scale study in Milan, Amsterdam, and Dubai. arXiv
Hu, S., Orsi, G., Santi, P., Wang, A., Fugiglando, U., Ratti, C. (2025). Evaluating the impact of zone 30 policies on citywide road traffic emissions and efficiency: A big data-driven approach. Transportation Research Part A
The material on this website can be used freely in any publication provided that:
For more information,
senseable-contacts@mit.edu
