Citation

Goenaga B, Shane Underwood B, Castorena C, Cantillo V, Arellana J. Lat. Am. Transp. Stud. 2023; 1: e100005.

Copyright

(Copyright © 2023, Elsevier Publishing)

DOI

10.1016/j.latran.2023.100005

PMID

unavailable

Abstract

Lockdowns to deal with the COVID-19 outbreak affected peoples' life in different dimensions. In particular, we are interested in evaluating the effects on traffic flows. For this purpose, it is necessary to accurately estimate the temporal variation of traffic flows during the pandemic period. New data collection techniques, including information from smartphones, can be used to collect this information at multiple locations of a road network. A key step in using this new data collection is the validation against more traditional measures to ensure consistency in traffic volume interpretation. This paper presents a case study whose main goals are to compare the smartphone-based traffic count predictions from the StreetLight data source against the values reported by traditional methods of traffic quantification and estimate reductions and recovery rates on traffic volumes in North Carolina during the COVID-19 pandemic. The results show that the largest reductions in traffic flows occurred mainly during the first three months of lockdown.

Language: en

Keywords

COVID-19; Non-parametric methods; Time series; Traffic disruptions