Year: 2025
Type: article
Abstract: Traffic accidents are a major global public health issue, causing around 1.19 million deaths annually, with the highest impact on individuals aged 5 to 29. This paper reviews recent advances in using Machine Learning (ML) and Deep Learning (DL) for traffic accident prediction and analysis. Covering 191 studies from the past five years, it explores models for predicting accident risk, frequency, severity, and duration, along with general statistical analysis. This is the first broad review covering multiple domains in traffic incident prediction. It highlights the value of integrating streaming analytics and diverse data sources to enhance prediction accuracy and handle traffic data complexity. By identifying current trends and research gaps, the study supports efforts to reduce road traffic fatalities by 2030 in line with WHO targets. Keywords: Traffic Incident Management, Machine Learning, Deep Learning, Streaming Analytics, Predictive Modeling
DOI: 10.55041/ijsrem47533
Author: PIDIKITI JAHNAVI
Language: English
Topic: Traffic and Road Safety
Subfield: Safety, Risk, Reliability and Quality
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4410380608