Geometrical effects on mobility

TL;DR

This study investigates how random street deletions in synthetic tessellated cities affect displacement statistics, revealing that street connectivity significantly influences the shape of displacement distribution functions.

Contribution

It introduces the novel observation that street network connectivity determines the form of displacement CCDFs, including a previously unreported linear behavior.

Findings

Displacement CCDFs transition from stretched exponential to linear and exponential forms as streets are deleted.

Connectivity of street networks critically influences displacement distribution shapes.

Linear CCDF behavior observed for certain street deletion proportions, a new finding in the literature.

Abstract

In this paper we analyze the effect of randomly deleting streets of a synthetic city on the statistics of displacements. Our city is constituted initially by a set of streets that form a regular tessellation of the euclidean plane. Therefore we will have three types of cities, formed by squares, triangles or hexagons. We studied the complementary cumulative distribution function for displacements (CCDF). For the whole set of streets the CCDF is a stretched exponential, and as streets are deleted this function becomes a linear function and then two clear different exponentials. This behavior is qualitatively the same for all the tessellations. Most of this functions has been reported in the literature when studying the displacements of individuals based on cell data trajectories and GPS information. However, in the light of this work, the appearance of different functions for displacements CCDF can be attributed to the connectivity of the underlying street network. It is remarkably that for some proportion of streets we got a linear function for such function, and as far as we know this behavior has not been reported nor considered. Therefore, it is advisable to analyze experimental in the light of connectivity of the street network to make correlations with the present work.