Understanding Human Mobility Flows from Aggregated Mobile Phone Data

TL;DR

This paper presents a method to analyze human mobility patterns using aggregated mobile phone data and Wasserstein distance, enabling insights into daily flows and traffic management without tracking individual devices.

Contribution

The paper introduces a novel approach using Wasserstein distance to extract mobility flows from coarse-grained aggregated data, avoiding individual tracking.

Findings

Effective extraction of main movement directions from density profiles

Application to monitor daily commuter flows and traffic patterns

Potential for large event organization and traffic control

Abstract

In this paper we deal with the study of travel flows and patterns of people in large populated areas. Information about the movements of people is extracted from coarse-grained aggregated cellular network data without tracking mobile devices individually. Mobile phone data are provided by the Italian telecommunication company TIM and consist of density profiles (i.e. the spatial distribution) of people in a given area at various instants of time. By computing a suitable approximation of the Wasserstein distance between two consecutive density profiles, we are able to extract the main directions followed by people, i.e. to understand how the mass of people distribute in space and time. The main applications of the proposed technique are the monitoring of daily flows of commuters, the organization of large events, and, more in general, the traffic management and control.