Visualizing the structure of urban mobility with bundling: A case study of the city of São Paulo

Abstract

Visualization of urban mobility data can facilitate the analysis and the decision-making process by public managers. However, mobility datasets tend to be very large and pose several challenges to the use of visualization, such as algorithm scalability and data occlusion. One approach to solve this problem is trail bundling, which groups motion trails that are spatially close in a simplified representation. This paper presents the results of adapting and using a recent bundling technique on a big dataset of urban mobility in São Paulo. The results show that bundling allows the visualization of various mobility patterns in the city.

Using bundling to visualize multivariate urban mobility structure patterns in the São Paulo Metropolitan Area

“…This paper extends an earlier work presented in the IV Brazilian Workshop of Urban Computing (CoUrb 2020) [14] with more detailed analyses of the findings obtained by exploring the bundled visualizations of the SPMA data as well as a better coverage of related work and more detailed explanation about the techniques we used. It is organized as follows.…”

“…By suitably combining filtering (to reduce the amount of data and/or attributes to be explored) with bundling (to simplify the created visualizations and reduce visual clutter) and with the available visual channels (opacity, color, direction), we highlight different patterns in the OD17 dataset which would not have been easily obtainable by classical data mining and data analysis tools. In contrast to earlier work [14], this paper presents visual explorations of additional attribute combinations -density per social strata, mobility of young students per social strata, directions at peak hours, density by transportation mode, and trip distance per trip reasons. Together with earlier results [10,14], our results strengthen the claim that trail bundling is an useful and usable tool for the visual analysis of large OD trail-sets.…”

“…In contrast to earlier work [14], this paper presents visual explorations of additional attribute combinations -density per social strata, mobility of young students per social strata, directions at peak hours, density by transportation mode, and trip distance per trip reasons. Together with earlier results [10,14], our results strengthen the claim that trail bundling is an useful and usable tool for the visual analysis of large OD trail-sets. The bundled layout for the trajectory dataset highlights its centrality structure over the represented area.…”

“…Section 5 presents the results of our analysis and discusses the use of bundling in urban mobility visualization. In addition to the original explorations presented in [14], Section 5 shows the visual exploration of several other attribute combinations: density per social strata (Section 5.5), mobility of young students per social strata (Section 5.6), directions at peak hours (Section 5.7), density by transportation mode (Section 5.8), and trip distance per trip reasons (Section 5.9). Section 6 concludes the paper.…”

An Approach Inspired by Simulation Points to Accelerate Smart Cities Simulations

“…Urban mobility is an aspect of great interest of citizens and government. Urban mobility is key to ensure quality of life in large cities [Martins et al 2020]. Nevertheless, cities like São Paulo Tokyo, Paris and New York face issues that are consequence of mobility problems, such as pollution, insufficient public transportation and heavy urban traffic.…”