Multi-model traffic accident clearance time prediction framework

Anyi Zhang; Qianqian Wang; Zhejun Huang; Jiyao Yin; Lili Yang

doi:10.1117/12.3030789

16 May 2024 Multi-model traffic accident clearance time prediction framework

Anyi Zhang, Qianqian Wang, Zhejun Huang, Jiyao Yin, Lili Yang

Proceedings Volume 13160, Fourth International Conference on Smart City Engineering and Public Transportation (SCEPT 2024); 131600O (2024) https://doi.org/10.1117/12.3030789
Event: 4th International Conference on Smart City Engineering and Public Transportation (SCEPT 2024), 2024, Beijin, China

Abstract

Accurate traffic accident clearance times prediction can help road managers make effective decisions and reduce property damage. This paper aims to develop a framework for traffic accident clearance time prediction and find the best prediction model. We propose a multi-model prediction framework for traffic accident severity. This framework consists of three parts: preprocessing of imbalanced data, variable selection and establishment of hybrid models: RF-SVM, RFBPNN, and RF-BN. Four highways in Shandong Province's traffic accident data are used as a case study in this paper. Based on the data used in this paper and previous literature exploration, three mixed models are constructed. Comparing the outcomes, we discover that the RF-SVM model has the highest prediction accuracy, up to 0.98, for the oversampled data set. This framework can be used to forecast the clearance time for traffic accidents, allowing for prompt emergency response and a reduction in fatalities and property damage.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Anyi Zhang, Qianqian Wang, Zhejun Huang, Jiyao Yin, and Lili Yang "Multi-model traffic accident clearance time prediction framework", Proc. SPIE 13160, Fourth International Conference on Smart City Engineering and Public Transportation (SCEPT 2024), 131600O (16 May 2024); https://doi.org/10.1117/12.3030789

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available