Citation

Cogollos-Adrián H, Porras-Alfonso S, Baruque-Zanón B. Transp. Res. Proc. 2021; 58: 654-661.

Copyright

(Copyright © 2021, Elsevier Publications)

DOI

10.1016/j.trpro.2021.11.086

PMID

unavailable

Abstract

The density of traffic within urban areas depends on multiple factors, and among those; one that has major impact is the weather. This paper presents a study that empirically analyzes the traffic flow velocity depending on the atmospherically conditions and the day schedule. The objective is to assess systematically to which extent there is a correlation between the vehicle velocity on an urban environment and the weather, depending on the time of day. The presented case of study uses a set of real data, specifically the trips made by taxis in the city of Porto (Portugal). First, the vehicle GPS trajectories are analyzed to identify departure and arrival points in key locations of the city and estimate the route average velocity; while they are enriched by adding weather and time conditions. The data is processed with different regression techniques to calculate the influence of the variables on the velocity. The results show differences between workdays and weekends, as well as differences in speeds with a climate conditions favourable for driving compared to other more adverse ones.

Language: en

Keywords

GPS Trajectories; MARS; PLS Regression; Traffic Analysis