Citation

Aiash A, Robusté F. Transp. Res. Proc. 2021; 58: 678-682.

Copyright

(Copyright © 2021, Elsevier Publications)

DOI

10.1016/j.trpro.2021.11.089

PMID

unavailable

Abstract

Working hours in Spain is generally last for 8 hours per day with a maximum of 40 hours per week. Working hours, therefore, can impact the traffic flow and characteristics due to the intensity of road usage by different road users. On the other side, traffic accidents can also be impacted by several temporal factors that may lead to a higher number of traffic accidents or may increase the level of injury alongside other risk factors. In this study, these timings are examined to comprehend the influence of the temporal factors represented by the working hours scheme on traffic accident injuries in Barcelona. Another temporal factor which is the season of the year is included to provide a wider and clearer image for the conducted results and the current situation. The data is collected from the open data service provided by the City Hall of Barcelona. Data preparation and segregation to include the categories of working hours that may lead to a different level of injury resulted from a traffic accident is, firstly, carried. Then, a machine learning model is applied to classify the correlations for both the temporal factors and traffic accidents. Eventually, a Tree Augmented Naïve Bayes model is applied. The results showed that both working hours timing and summer season have higher probabilities of having traffic accidents slight injuries with different medical care assistance provided compared to other timings.

Language: en

Keywords

Barcelona; Bayesian network; Traffic accidents; Working hours