Effect of transportation parameters on traffic accident in urban areas comparison study of anfis with statistical analysis

Ghassan M. Suleiman, Mohammad K. Younes*, Murat Ergun, Khaled Al Omari

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Citations (Scopus)

Abstract

Traffic accidents present a serious problem for both developed and developing countries and have become an urgent matter to tackle in all large metropolitan areas. This study aims to perform a deep comprehensive analysis of the traffic accidents issue in Istanbul, one of the world’s most populous cities. The accidents were classified and its intensities were presented on Istanbul map using a GIS tool. Furthermore, the performance of Negative Binomial Regression analysis and Adaptive Neuro-Fuzzy Inference System (ANFIS) model was assessed. Data collection of independent variables included distribution of trips, percentage of street parking, rate of car ownership, street density and population density. Trips were divided into three categories, passenger car, minibus and bus trips. The results showed that four legs intersection got the highest proportion of accidents among the other types with (40%). It also demonstrated that increasing both the percentage of bus trips and the percentage of street parking will decrease the traffic accident rate. Furthermore, the implementation of ANFIS model increased the accuracy of forecasts and reduced errors more than the regression model.

Original languageEnglish
Pages (from-to)129-134
Number of pages6
JournalInternational Journal of Safety and Security Engineering
Volume11
Issue number2
DOIs
Publication statusPublished - Apr 2021

Bibliographical note

Publisher Copyright:
© 2021 WITPress. All rights reserved.

Keywords

  • ANFIS
  • GIS and traffic
  • Traffic accidents
  • Traffic modeling
  • Traffic regression analysis

Fingerprint

Dive into the research topics of 'Effect of transportation parameters on traffic accident in urban areas comparison study of anfis with statistical analysis'. Together they form a unique fingerprint.

Cite this