Impact of Wind Turbine Distraction on Crash Severity: Assessment and Prediction Study

https://doi.org/10.28991/CEJ-2025-011-04-018
Wind Turbines Driver Distractions Mixed-Effects Logit Model Machine Learning in Traffic Analysis Bagged Tree Classifier Crash Severity.

Authors

  • Fadi Alhomaidat
    fadi.alhomaidat@ahu.edu.jo
    Department of Civil Engineering, College of Engineering, Al Hussein Bin Talal University, Ma'an 71111,, Jordan
  • Mu'ath Al-Tarawneh Civil and Environmental Engineering Department, College of Engineering, Mutah University, Mutah-Karak, 61710, P.O. BOX 7,, Jordan
  • Mousa Abushattal Department of Civil Engineering, College of Engineering, Al Hussein Bin Talal University, Ma'an 71111,, Jordan
  • Renad Al-Alaya Department of Civil Engineering, College of Engineering, Al Hussein Bin Talal University, Ma'an 71111,, Jordan
Vol. 11 No. 4 (2025): April
Research Articles

Downloads

Wind turbines are increasingly installed near highways, yet their potential role as external distractions impacting traffic crashes remains underexplored. This study investigates the effect of wind turbines on crash severity and frequency along Jordan's King's Highway, analyzing data through a mixed-effect logit model and machine learning techniques. Key factors, including driver demographics, road geometry, and environmental conditions, were incorporated to provide a comprehensive analysis. The findings indicate a 117.4% increase in severe injury crashes (KAB) and a 25.7% rise in property damage only near wind turbines. Using MK models such as Bagged Tree classifiers and SMOT-balanced datasets, the study achieved a high prediction accuracy of 89.6% for crash severity. Shapley value analysis identified crash type and wind turbine proximity as critical predictors, while other influential factors included younger drivers, poorly separated roads, and higher speed limits. By integrating statistical and ML approaches, this research provides actionable insights into the relationship between wind turbines and road safety. The results underscore the need for regulatory policies to optimize wind turbine placement and reduce their potential as driver distractions. This study also demonstrated the potential of ML techniques to enhance traffic safety analysis, paving the way for future research to address multi-class crash severity predictions and other external roadside distractions.

 

Doi: 10.28991/CEJ-2025-011-04-018

Full Text: PDF