Skip to main navigation Skip to search Skip to main content

Traffic Conflicts-Based Crash Risk Assessment at Intersections Using Extreme Value Theory Approach

  • Chuanyun Fu
  • , Jiaming Liu
  • , Ayinigeer Wumaierjiang
  • , Huahua Liu
  • , Zhaoyou Lu
  • , Wei Bai*
  • *Corresponding author for this work
  • School of Transportation Science and Engineering, Harbin Institute of Technology
  • Sichuan Police College

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

To assess traffic risk at intersections during snowy weather, this study introduces a collision risk determination method based on the extreme values of Post-Encroachment Time (PET). Subsequently, the SUMO software is utilized to simulate traffic flow during peak hours under continuous snowfall conditions at the intersection, and the Markov Chain Monte Carlo (MCMC) method is employed to fit the parameters of the extreme value theory model. The results indicate that the MCMC method performs better in handling parameter estimation for the generalized extreme value (GEV) model, and the risk of traffic conflict events in snowy environments is significantly higher than in clear weather.

Original languageEnglish
Title of host publicationThe Proceedings of 2024 International Conference on Artificial Intelligence and Autonomous Transportation - Volume VI
EditorsJun Liu, Wang Li, Xiongfei Geng, Ke Zhang, Honghai Ji, Kailong Li
PublisherSpringer Science and Business Media Deutschland GmbH
Pages236-243
Number of pages8
ISBN (Print)9789819639762
DOIs
StatePublished - 2025
Externally publishedYes
EventInternational Conference on Artificial Intelligence and Autonomous Transportation, AIAT 2024 - Beijing, China
Duration: 6 Dec 20248 Dec 2024

Publication series

NameLecture Notes in Electrical Engineering
Volume1394
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceInternational Conference on Artificial Intelligence and Autonomous Transportation, AIAT 2024
Country/TerritoryChina
CityBeijing
Period6/12/248/12/24

Keywords

  • Crash Risk
  • Extreme Value Theory
  • Markov Chain Monte Carlo
  • Post Encroachment Time
  • Snow Weather
  • Traffic Conflict

Fingerprint

Dive into the research topics of 'Traffic Conflicts-Based Crash Risk Assessment at Intersections Using Extreme Value Theory Approach'. Together they form a unique fingerprint.

Cite this