Who rides Uber anyway? A census-tract level analysis and clustering of ride-shares for the city of Chicago during the era of the pandemic

The COVID-19 pandemic has led to an unprecedented change in transportation, including shared mobility services. This study attempted to identify the user group of ride-share services by leveraging daily ride-sharing trip data for the year of 2020 associated with other socio-demographic and built-environment attributes of Chicago, Illinois. The study employed K-means clustering for user group segmentation. Results show: i) the cluster with the largest share of census tracts generate lowest average trips which is clearly an impact of the pandemic; ii) The high-income cluster generates short trip and coupled with high population, land-use, and employment density; iii) The low-income cluster generates longer trips coupled with diversity of land-use mx, employment and population density. Results of this study provide insights for policymakers and ride- sharing operators to ensure access to the services among the population irrespective of spatial diversity.