Universal model of individual and population mobility on diverse spatial scales

Xiao-Yong Yan,Ying-Cheng Lai,Wen-Xu Wang,Zi-You Gao

doi:10.1038/s41467-017-01892-8

Xiao-Yong Yan, Ying-Cheng Lai + Show 2 more

Open Access

https://doi.org/10.1038/s41467-017-01892-8

Copy DOI

Abstract

Studies of human mobility in the past decade revealed a number of general scaling laws. However, to reproduce the scaling behaviors quantitatively at both the individual and population levels simultaneously remains to be an outstanding problem. Moreover, recent evidence suggests that spatial scales have a significant effect on human mobility, raising the need for formulating a universal model suited for human mobility at different levels and spatial scales. Here we develop a general model by combining memory effect and population-induced competition to enable accurate prediction of human mobility based on population distribution only. A variety of individual and collective mobility patterns such as scaling behaviors and trajectory motifs are accurately predicted for different countries and cities of diverse spatial scales. Our model establishes a universal underlying mechanism capable of explaining a variety of human mobility behaviors, and has significant applications for understanding many dynamical processes associated with human mobility.

Highlights

Studies of human mobility in the past decade revealed a number of general scaling laws
In China or the United States, the size of the region can be on the order of millions of square kilometers, while in small countries such as Belgium, the domain size is only about 10 km. (Here we regard international travel as atypical and exclude it from our consideration.) Comparing Belgium with China or the United States, the difference in the spatial scale in terms of areas is at least two orders of magnitude
To understand the dynamics of human movements and to uncover scaling laws underlying human mobility are of fundamental importance as they are relevant to problems such as disease control, social stability, congestion alleviation, information propagation, and e-commerce[21,22,23,24,25,26,27,28]

Summary

Introduction

Studies of human mobility in the past decade revealed a number of general scaling laws. Existing models of human mobility are tailored at understanding and characterizing scaling laws either at large (e.g., big country) or small (e.g., city) scales—we lack a universal model capable of quantifying human movements across all spatial scales. Another deficiency of existing models is that they can explain human mobility patterns either at the individual or at the population level, but not both.

Objectives

Methods

Results

Conclusion