The association of socio-demographic characteristics towards driver behaviour and traffic fatality in Selangor, Malaysia

Radzuan, N. Q. and Mohd Hasnun Ariff, Hassan and Abu Kassim, K. A. and Ab. Rashid, A. A. and Mohd Razelan, Intan Suhana and Othman, N. A. (2022) The association of socio-demographic characteristics towards driver behaviour and traffic fatality in Selangor, Malaysia. In: Proceedings of HUMENS 2021. Human Engineering Symposium, HUMENS 2021 , 22 February 2021 - 22 February 2021 , Virtual Online. pp. 395-405.. ISSN 2195-4364 ISBN 978-981164114-5

Preview	Pdf The association of socio-demographic characteristics towards.pdf Download (146kB) | Preview
	Pdf 01_The Association of Socio-Demographics Background towards Traffic Fatality in Selangor, Malaysia.pdf Restricted to Repository staff only Download (309kB) | Request a copy

Abstract

There are more than 6,000 road traffic fatalities each year in Malaysia since 2004, although the index of road traffic fatality per 100,000 population is decreasing steadily each year. Selangor, one of the states in Malaysia, has recorded the highest road traffic fatalities in 2019. Since traffic accidents and driver’s behaviour is correlated, this study aims to understand the association between the socio-demographic characteristics and driver’s behaviour, and the correlation between these parameters and road fatality. Road traffic accidents in Selangor in the year 2019 were used as the input data in this study. The selected socio-demographic characteristics are gender, age group, ethnicity, level of education, job specification, and type of driving licence. The statistics show that most of the fatalities involve employed young Malay drivers with more than five years driving experience. A classification model was formulated to classify the driver’s behaviour based on the socio-demographic characteristics and vehicle type data. In addition, another classification model was developed, combining the aforementioned parameters as independent variables to classify the occurrence of fatal accidents. The supervised machine learning analysis was conducted using classifiers such as Random Forest, Classification Tree, Neural Network, Support Vector Machine, and Naïve Bayes using Orange data mining software. Random Forest was found to produce the most accurate classification of both driver’s behaviour based on the socio-demographic characteristics, and the occurrence of fatal accidents as compared to other classifiers. This study shows that socio-demographic characteristics are indeed associated with driver’s behaviour, and both socio-demographic characteristics and driver’s behaviour are among the causes of fatal accidents.

Item Type:	Conference or Workshop Item (Paper)
Additional Information:	Indexed by Scopus
Uncontrolled Keywords:	Driving behaviour; Socio-demographics; Supervised machine learning; Traffic fatality
Subjects:	T Technology > TJ Mechanical engineering and machinery T Technology > TK Electrical engineering. Electronics Nuclear engineering T Technology > TP Chemical technology
Faculty/Division:	Faculty of Electrical & Electronic Engineering Faculty of Civil Engineering Technology Faculty of Mechanical and Automotive Engineering Technology
Depositing User:	Pn. Hazlinda Abd Rahman
Date Deposited:	05 Jul 2024 02:48
Last Modified:	05 Jul 2024 02:51
URI:	http://umpir.ump.edu.my/id/eprint/33364
Download Statistic:	View Download Statistics

Actions (login required)

View Item