Exploring the Influence of Driving Automation in Vehicle Recurrent Crashes: A Random Forest Approach

Hala A. Eljailany; Helai Huang; Hanchu Zhou; Ali M.A. Ibrahim

doi:10.3233/ATDE250440

Exploring the Influence of Driving Automation in Vehicle Recurrent Crashes: A Random Forest Approach

Hala A. Eljailany
, Helai Huang
, Hanchu Zhou^*
, Ali M.A. Ibrahim

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

This study assesses the impact of driving automation on the frequency of repeated vehicle crashes. Using random forest modeling on a comprehensive crash dataset including vehicles with SAE Level 2 and Level 3 automation, we analyzed the relationship between these existing automation features, safety systems, and the time intervals between crashes. The analysis indicates that the presence of Level 2 and 3 automation is significantly associated with longer periods between crash recurrences. Further investigation via simulation explored better scenarios, specifically the potential effects of widespread adoption of SAE Level 3 automation and improved infrastructure. These simulations project that Level 3 systems, especially combined with infrastructure enhancements, could lead to substantially greater increases in time between crashes than observed with current L2/L3 technologies. Our findings quantify the safety benefits of current automation levels while underscoring the significant future potential of higher automation spreading (Level 3) and synergistic infrastructure improvements to enhance road safety, providing valuable guidance for policy and industry.

Original language	English
Title of host publication	Intelligent Transportation Engineering - Proceedings of the 9th International Conference, ICITE 2024
Editors	Guoqiang Mao
Publisher	IOS Press BV
Pages	408-417
Number of pages	10
ISBN (Electronic)	9781643686028
DOIs	https://doi.org/10.3233/ATDE250440
State	Published - 17 Jul 2025
Externally published	Yes
Event	9th International Conference on Intelligent Transportation Engineering, ICITE 2024 - Xi'an, China Duration: 18 Oct 2024 → 20 Oct 2024

Publication series

Name	Advances in Transdisciplinary Engineering
Volume	72
ISSN (Print)	2352-751X
ISSN (Electronic)	2352-7528

Conference

Conference	9th International Conference on Intelligent Transportation Engineering, ICITE 2024
Country/Territory	China
City	Xi'an
Period	18/10/24 → 20/10/24

Bibliographical note

Publisher Copyright:
© 2025 The Authors.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure
SDG 11 Sustainable Cities and Communities

Keywords

Automated driving
Random Forest
Road safety
Vehicle's Repeated Crashes

ASJC Scopus subject areas

Software
Algebra and Number Theory
Computer Science Applications
Strategy and Management
Industrial and Manufacturing Engineering

Access to Document

10.3233/ATDE250440

Cite this

Eljailany, H. A., Huang, H., Zhou, H., & Ibrahim, A. M. A. (2025). Exploring the Influence of Driving Automation in Vehicle Recurrent Crashes: A Random Forest Approach. In G. Mao (Ed.), Intelligent Transportation Engineering - Proceedings of the 9th International Conference, ICITE 2024 (pp. 408-417). (Advances in Transdisciplinary Engineering; Vol. 72). IOS Press BV. https://doi.org/10.3233/ATDE250440

@inproceedings{de3fe9f3ce88406d9d9bd389c9cb6e4a,

title = "Exploring the Influence of Driving Automation in Vehicle Recurrent Crashes: A Random Forest Approach",

abstract = "This study assesses the impact of driving automation on the frequency of repeated vehicle crashes. Using random forest modeling on a comprehensive crash dataset including vehicles with SAE Level 2 and Level 3 automation, we analyzed the relationship between these existing automation features, safety systems, and the time intervals between crashes. The analysis indicates that the presence of Level 2 and 3 automation is significantly associated with longer periods between crash recurrences. Further investigation via simulation explored better scenarios, specifically the potential effects of widespread adoption of SAE Level 3 automation and improved infrastructure. These simulations project that Level 3 systems, especially combined with infrastructure enhancements, could lead to substantially greater increases in time between crashes than observed with current L2/L3 technologies. Our findings quantify the safety benefits of current automation levels while underscoring the significant future potential of higher automation spreading (Level 3) and synergistic infrastructure improvements to enhance road safety, providing valuable guidance for policy and industry.",

keywords = "Automated driving, Random Forest, Road safety, Vehicle's Repeated Crashes",

author = "Eljailany, \{Hala A.\} and Helai Huang and Hanchu Zhou and Ibrahim, \{Ali M.A.\}",

note = "Publisher Copyright: {\textcopyright} 2025 The Authors.; 9th International Conference on Intelligent Transportation Engineering, ICITE 2024 ; Conference date: 18-10-2024 Through 20-10-2024",

year = "2025",

month = jul,

day = "17",

doi = "10.3233/ATDE250440",

language = "English",

series = "Advances in Transdisciplinary Engineering",

publisher = "IOS Press BV",

pages = "408--417",

editor = "Guoqiang Mao",

booktitle = "Intelligent Transportation Engineering - Proceedings of the 9th International Conference, ICITE 2024",

address = "Netherlands",

}

Eljailany, HA, Huang, H, Zhou, H & Ibrahim, AMA 2025, Exploring the Influence of Driving Automation in Vehicle Recurrent Crashes: A Random Forest Approach. in G Mao (ed.), Intelligent Transportation Engineering - Proceedings of the 9th International Conference, ICITE 2024. Advances in Transdisciplinary Engineering, vol. 72, IOS Press BV, pp. 408-417, 9th International Conference on Intelligent Transportation Engineering, ICITE 2024, Xi'an, China, 18/10/24. https://doi.org/10.3233/ATDE250440

Exploring the Influence of Driving Automation in Vehicle Recurrent Crashes: A Random Forest Approach. / Eljailany, Hala A.; Huang, Helai; Zhou, Hanchu et al.
Intelligent Transportation Engineering - Proceedings of the 9th International Conference, ICITE 2024. ed. / Guoqiang Mao. IOS Press BV, 2025. p. 408-417 (Advances in Transdisciplinary Engineering; Vol. 72).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Exploring the Influence of Driving Automation in Vehicle Recurrent Crashes

T2 - 9th International Conference on Intelligent Transportation Engineering, ICITE 2024

AU - Eljailany, Hala A.

AU - Huang, Helai

AU - Zhou, Hanchu

AU - Ibrahim, Ali M.A.

PY - 2025/7/17

Y1 - 2025/7/17

N2 - This study assesses the impact of driving automation on the frequency of repeated vehicle crashes. Using random forest modeling on a comprehensive crash dataset including vehicles with SAE Level 2 and Level 3 automation, we analyzed the relationship between these existing automation features, safety systems, and the time intervals between crashes. The analysis indicates that the presence of Level 2 and 3 automation is significantly associated with longer periods between crash recurrences. Further investigation via simulation explored better scenarios, specifically the potential effects of widespread adoption of SAE Level 3 automation and improved infrastructure. These simulations project that Level 3 systems, especially combined with infrastructure enhancements, could lead to substantially greater increases in time between crashes than observed with current L2/L3 technologies. Our findings quantify the safety benefits of current automation levels while underscoring the significant future potential of higher automation spreading (Level 3) and synergistic infrastructure improvements to enhance road safety, providing valuable guidance for policy and industry.

AB - This study assesses the impact of driving automation on the frequency of repeated vehicle crashes. Using random forest modeling on a comprehensive crash dataset including vehicles with SAE Level 2 and Level 3 automation, we analyzed the relationship between these existing automation features, safety systems, and the time intervals between crashes. The analysis indicates that the presence of Level 2 and 3 automation is significantly associated with longer periods between crash recurrences. Further investigation via simulation explored better scenarios, specifically the potential effects of widespread adoption of SAE Level 3 automation and improved infrastructure. These simulations project that Level 3 systems, especially combined with infrastructure enhancements, could lead to substantially greater increases in time between crashes than observed with current L2/L3 technologies. Our findings quantify the safety benefits of current automation levels while underscoring the significant future potential of higher automation spreading (Level 3) and synergistic infrastructure improvements to enhance road safety, providing valuable guidance for policy and industry.

KW - Automated driving

KW - Random Forest

KW - Road safety

KW - Vehicle's Repeated Crashes

UR - https://www.scopus.com/pages/publications/105016844155

U2 - 10.3233/ATDE250440

DO - 10.3233/ATDE250440

M3 - Conference contribution

AN - SCOPUS:105016844155

T3 - Advances in Transdisciplinary Engineering

SP - 408

EP - 417

BT - Intelligent Transportation Engineering - Proceedings of the 9th International Conference, ICITE 2024

A2 - Mao, Guoqiang

PB - IOS Press BV

Y2 - 18 October 2024 through 20 October 2024

ER -

Exploring the Influence of Driving Automation in Vehicle Recurrent Crashes: A Random Forest Approach

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this