Characterizing Urban Mobility Patterns: A Case Study of Mexico City

Abstract

Seamless access to destinations of value such as workplaces, schools, parks or hospitals, influences the quality of life of people all over the world. The first step to planning and improving proximity to services is to estimate the number of trips being made from different parts of a city. A challenge has been representative data available for that purpose. Relying on expensive and infrequently collected travel surveys for modeling trip distributions to facilities has slowed down the decision-making process. The growing abundance of data already collected, if analyzed with the right methods, can help us with planning and understanding cities. In this chapter, we examine human mobility patterns extracted from data passively collected. We present results on the use of points of interest (POIs) registered on Google Places to approximate trip attraction in a city. We compare the result of trip distribution models that utilize only POIs with those utilizing conventional data sets, based on surveys. We show that an extended radiation model provides very good estimates when compared with the official origin–destination matrices from the latest census in Mexico City.