A literature review: AI models for road safety for prediction of crash frequency and severity
Abstract Artificial intelligence and machine learning have brought a new paradigm in road safety, moving from the traditional approach to adopting data-driven techniques for predicting the frequency and severity of crashes. This review gives the main research contributions and highlights works towar...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-05-01
|
| Series: | Discover Civil Engineering |
| Online Access: | https://doi.org/10.1007/s44290-025-00255-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Artificial intelligence and machine learning have brought a new paradigm in road safety, moving from the traditional approach to adopting data-driven techniques for predicting the frequency and severity of crashes. This review gives the main research contributions and highlights works toward integrating different data sets into one comprehensive safety assessment, starting from traffic patterns, environmental conditions, and driver behavior analytics. This research paper reviews how AI and ML are revolutionizing road safety: from basic statistical methods, such as Ordered Probit, to advanced Neural Networks and Deep Learning techniques. It provides rich detail on the dynamics of crashes to serve as a basis for intelligent traffic management and policy decisions. That's because what has transpired with the confluence of AI, ML, and road safety initiatives is a major step forward to reduce traffic incidents and improve roadway safety, from early statistical models to sophisticated systems capable of predicting crashes and identifying opportunities for intervention Fig. 2. |
|---|---|
| ISSN: | 2948-1546 |