Modalflow: Cross-Origin Flow Data Visualization for Urban Mobility

Abstract

Pervasive data have become a key source of information for mobility and transportation analyses. However, as a secondary source, it has a different methodological origin than travel survey data, usually relying on unsupervised algorithms, and so it requires to be assessed as a dataset. This assessment is challenging, because, in general, there is not a benchmark dataset or a ground truth scenario available, as travel surveys only represent a partial view of the phenomenon and suffer from their own biases. For this critical task, which involves urban planners and data scientists, we study the design space of the visualization of cross-origin, multivariate flow datasets. For this purpose, we introduce the Modalflow system, which incorporates and adapts different visualization techniques in a notebook-like setting, presenting novel visual encodings and interactions for flows with modal partition into scatterplots, flow maps, origin-destination matrices, and ternary plots. Using this system, we extract general insights on visual analysis of pervasive and survey data for urban mobility and assess a mobile phone network dataset for one metropolitan area.

Spatial Analysis of Subway Passenger Traffic in Saint Petersburg

“…They are related to personal activity and SoBigData initiative (Andrienko et al, 2020), spatial connectedness by public transport (Andrienko et al, 2019) transport network graphs visualization (Andrienko et al, 2016). Origin-destination (OD) data visualization and passenger flow simulation are also very popular (Pérez-Messina et al, 2020;Massobrio & Nesmachnow, 2020). Some authors present a micro-prediction approach to predict individual passenger's destination station and arrival time (Lin et al, 2017).…”

Influence of Socioeconomic Factors on Transit Demand During the COVID-19 Pandemic: A Case Study of Bogotá’s BRT System

“…We leveraged smartcard data from Bogota's Bus Rapid Transit (BRT) system to assess the differential impact of the COVID-19 pandemic restrictions by socioeconomic groups. Although the data does not contain any socioeconomic variables, we use inference methods (Pappalardo et al, 2019;Pérez-Messina et al, 2020) to obtain a probability distribution for one socioeconomic variable in the context of Bogota. Our research objectives are 1) to infer socioeconomic characteristics of frequent BRT users based on smartcard data, and 2) to use this information to expand our understanding of the COVID-19 lockdowns impacts and trends on transit demand by strata.…”