Attractive locations work like large planets

Human mobility impacts many aspects of a city, from its spatial structure to its response to an epidemic. It is not only key to social interactions, innovation, and productivity, but also gives rise to traffic congestion, and fuels the spread of contagious diseases. Despite its importance, however, our understanding of what determines the flow of individuals to a location has remained incomplete. 

Researchers from ETH Zurich, MIT and the Santa Fe Institute have discovered a law that governs the number of visitors to any location based on how far they are traveling and how often they are visiting. It is based on the intuition that that people visit places more frequently when they have to travel shorter distances to get there. The visitation law opens up unprecedented possibilities for accurately predicting flows between locations, which could ultimately have applications in everything from city planning to preventing the spread of the next major pandemic. 

Analyzing mobile phone data in cities across the world

“Imagine you are standing on a busy plaza in Zurich and you see people coming and going. This may look pretty random and chaotic, but the law shows that these movements are surprisingly structured and predictable. It basically tells you how many of these people are coming from 1, 2 or 10 kilometers away and how many are visiting once, twice or 10 times a month”, says lead author Markus Schläpfer of ETH Zurich’s Future Cities Laboratory. What is surprising is that this regularity holds not only in Zurich, but across cities worldwide.

The researchers’ findings are a result of an analysis of aggregated mobile phone data from millions of anonymized cell phone users in highly diverse urban regions across the world, including Greater Boston in the United States, Lisbon in Europe, Singapore in Asia, and Dakar in Africa.

Universal law to predict human mobility

Like the gravitational pull of a large planet, attractive locations like famous shopping streets attract more visitors from far away than less prominent places, though less frequently than those coming from nearby locations. More precisely, the scientists found that the number of travelers to any location in a city decreases as the inverse square of their visiting frequency and the distance to their home location.

Schläpfer says the new paper can give urban planners “a baseline for understanding which locations in their cities are over- or under-performing,” in terms of the number of people they attract. It can inform planners about where to add amenities like parks and restaurants, or how much public transportation is needed for new urban developments.

A longer and slightly adapted version was previously published on Santa Fe News.