Multiple correspondence analysis applied to the study of the relationship between traffic crashes and precipitation on a highway in Brazil

Autor: Milena Arruda Silva, Leandro Canezin Guideli, Anselmo Chaves Neto, Larissa de Brum Passini, Alessander Christopher Morales Kormann
Jazyk: English<br />Spanish; Castilian<br />Portuguese
Rok vydání: 2020
Předmět:
Zdroj: Transportes, Vol 28, Iss 3 (2020)
Druh dokumentu: article
ISSN: 2237-1346
DOI: 10.14295/transportes.v28i3.2044
Popis: The incorporation of meteorological data in road accident research has been considered essential in studies that aim to analyze traffic crashes behavior. Thus, this paper intended to explore the relationship between the occurrence of traffic crashes and weather conditions, especially daily precipitation, in a selected segment of the BR-376 Highway (state of Paraná, Brazil). A multiple correspondence analysis (MCA) was applied (Statgraphics software) in order to determine the relationship among crash data variables and precipitation. The MCA analysis allowed the creation of a perceptual map, grouping variables that are similar among themselves, which revealed that rainfall intensity was not significant in determining the accident characteristics. However, the presence/absence of rain was a conditioning circumstance. Concerning road safety, the study gives practical implications that can be used by traffic agencies as the implementation of traffic and educational signs in specific points of the highway.
Databáze: Directory of Open Access Journals