Skip to main navigation Skip to search Skip to main content

Predicting crashes based on artificial neural networks and identifying the hazardous crash type at intersections

  • Xianghai Meng*
  • , Hongfei Sheng
  • , Xiaoning Wang
  • , Yuejing Lv
  • *Corresponding author for this work

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

Abstract

Research works to predict the traffic crash and to identify the hazard plane intersections are reviewed firstly. Then, this paper studies the use of a well-known artificial neural network (ANN), the multilayer perception (MLP), in predicting the number of each type of crash. The crash data include 4340 fatal/injury crashes occurred on 197 at-grade intersections in Harbin city from 2000 to 2004. Modeling result showed that the MLP is capable of predicting the crashes by type, and the testing accuracy is up to 89%. Lastly, a method to identify the hazardous crash patterns for a single intersection is presented based on the "overrepresented" concept. Copyright ASCE 2007.

Original languageEnglish
Title of host publicationInternational Conference on Transportation Engineering 2007, ICTE 2007
Pages1451-1456
Number of pages6
DOIs
StatePublished - 2007
EventInternational Conference on Transportation Engineering 2007, ICTE 2007 - Chengdu, China
Duration: 22 Jul 200724 Jul 2007

Publication series

NameInternational Conference on Transportation Engineering 2007, ICTE 2007

Conference

ConferenceInternational Conference on Transportation Engineering 2007, ICTE 2007
Country/TerritoryChina
CityChengdu
Period22/07/0724/07/07

Keywords

  • Artificial neural networks
  • Crash pattern
  • Hazardous crash type
  • Multilayer perceptron
  • Overrepresented

Fingerprint

Dive into the research topics of 'Predicting crashes based on artificial neural networks and identifying the hazardous crash type at intersections'. Together they form a unique fingerprint.

Cite this