Analysis of Traffic Safety Factors and Their Impact Using Machine Learning Algorithms - Research summary

Aug 4, 2026

Upload by Asher

Road traffic safety faces escalating challenges driven by interrelated human, vehicle, road, and environmental factors. This study develops a predictive and evaluative model using historical accident counts from the Pristina region over a 10-year period, classified by accident type and safety factors. Machine learning methods—Multiple Linear Regression, Artificial Neural Networks, and Random Trees—assess how human, vehicle, and road contributions relate to 36 categories of traffic accidents. Results show a very strong fit with Multiple Linear Regression and identify the road factor as the dominant influencer of traffic safety level.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Introduction

Safety factors: human, vehicle, and road

Human factors

Vehicle factors

Road and environmental factors

Analysis of Traffic Safety Factors and Their Impact Using Machine Learning Algorithms Liridon Sejdiu 1 , Tomaz Tollazzi 2 , Ferat Shala 1 , Halil Demolli 3* 1 Traffic and Transport Engineering, University of Prishtina “Hasan Prishtina”, Prishtina, Kosovo. 2 Department of Civil Engineering, University of Maribor, Maribor, Slovenia. 3 Engineering Design and Vehicles, University of Prishtina “Hasan Prishtina”, Prishtina, Kosovo. Received 18 April 2024; Revised 20 August 2024; Accepted 27 August 2024; Published 01 September 2024 Abstract The safety of road traffic is facing increasing challenges from a range of factors, and this study aims to address this issue. The paper describes the development of a model that assesses both the quantitative and qualitative aspects of the current traffic situation and can also predict future trends based on monthly data on traffic accidents over a period of years. The dataset is composed of the number of accidents that occurred in the Pristina region over a 10-year period, and these are categorized based on the type of accident and safety factors, including human, vehicle, and road factors. By using machine learning algorithms, a model has been developed that determines the factor with the greatest impact on traffic safety. To create the model, the algorithms Multiple Linear Regression (MLR), Artificial Neural Network (ANN), and Random Trees (RT) were used. The model evaluates the contribution of human, road, and vehicle factors to traffic accidents, using machine learning algorithms and 36 types of traffic accidents to analyze the relevant statistics. The results indicate a very good fit of the model according to the MLR algorithm, and this model also identifies the road factor as the main influencer of the traffic safety level. Keywords: Traffic Safety Factors; Traffic Accident; Machine Learning; Multiple Linear Regression. 1. Introduction Traffic, particularly road traffic, plays a crucial role in a country's economy, modern human life, and the globalization process by connecting different countries. However, as the number of vehicles increases, managing the growing number of accidents has become more challenging. Traffic accidents cause significant injuries and fatalities worldwide, with an economic impact that can cost up to 3% of a country's GDP. Traffic safety has recently faced serious challenges due to the high number of fatal accidents globally. According to the World Health Organization, approximately 1.35 million people die in traffic accidents each year, with over half of the victims being pedestrians and cyclists. Additionally, 93% of fatal accidents occur in developing countries. These alarming statistics highlight the urgent need for research to find solutions for preventing fatal accidents, especially in developing nations [1] . The global increase in the number of accidents can be attributed to several factors: the rising level of motorization, the large number of vehicles in poor technical condition, and the lack of adequate traffic knowledge and culture among both drivers and pedestrians. This issue is especially prevalent in developing countries, where, according to the World Health Organization, the number of accidents is significantly higher. * [Corresponding author: halil.demolli@uni-pr.edu](Corresponding author: halil.demolli@uni-pr.edu) [http://dx.doi.org/10.28991/CEJ-2024-010-09-06](http://dx.doi.org/10.28991/CEJ-2024-010-09-06) © 2024 by the authors. Licensee C.E.J, Tehran, Iran. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license ([http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/)). To mitigate traffic accidents, especially fatal ones, a thorough study must be conducted on the key factors that affect traffic safety and contribute to the development of traffic. These factors include human behavior, vehicles, and road

FAQs

What data and time span does the study use to model traffic safety?

The dataset uses monthly counts of traffic accidents in the Pristina region over a 10-year period. Accidents are categorized by type and by safety factors, including human, vehicle, and road factors.

Which machine learning algorithms are used in the proposed model?

The model is built using Multiple Linear Regression (MLR), Artificial Neural Network (ANN), and Random Trees (RT). These algorithms evaluate contributions of human, road, and vehicle factors to accidents.

What factor is identified as the main influencer of traffic safety in the results?

The results indicate a very good model fit with the MLR approach and identify the road factor as the primary influencer of the traffic safety level.

Aug 4, 2026

Upload by Asher