Scholarlink

<p>Compared to non-occupational drivers, taxi drivers face an increased risk of road accidents given their prolonged driving, heavy workloads, and rotating shifts that lead to increased fatigue. This study aims to investigate the risk factors causing road accidents among taxi drivers and to develop machine-learning models that classify drivers on a risk scale based on these factors. A survey was administered to 790 taxi drivers in the United Arab Emirates, covering demographics, work environment, and well-being factors. The survey responses were validated using a comprehensive accident history database collected from a taxi organization. Data mining techniques, such as cross-tabulation, were employed along with statistical analysis using the chi-square test to identify the key risk factors. The results identified age, employment status, experience, paid leave, stress, satisfaction with income and life, and fatigue as key factors leading to accidents. Moreover, four binary machine learning classification models have been trained and tested to classify the taxi driver profiles and to analyze their risk factors. These models achieved 91.8% accuracy in identifying taxi drivers with a history of at least one accident. The high accuracy underscores the models’ effectiveness in identifying high-risk driver profiles, allowing taxi corporations to develop targeted interventions and training aimed at reducing accident rates and fatalities. Future research could explore psychophysiological factors and evaluate the effectiveness of targeted interventions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A data-driven framework for urban transportation safety using data mining and machine learning

  • 期刊论文

  • 发表日期：

    2026年6月20日

查看全文
  • Imad Alsyouf,
  • Fakhariya Ibrahim,
  • Khaled Obaideen,
  • Ali Cheaitou,
  • Amir Shikhli,
  • Meera Almaazmi,
  • Alaa Abdelrahman,
  • Ahmad Alzghoul,
  • Salah Haridy,
  • Hamid A. Alhaj

摘要

Compared to non-occupational drivers, taxi drivers face an increased risk of road accidents given their prolonged driving, heavy workloads, and rotating shifts that lead to increased fatigue. This study aims to investigate the risk factors causing road accidents among taxi drivers and to develop machine-learning models that classify drivers on a risk scale based on these factors. A survey was administered to 790 taxi drivers in the United Arab Emirates, covering demographics, work environment, and well-being factors. The survey responses were validated using a comprehensive accident history database collected from a taxi organization. Data mining techniques, such as cross-tabulation, were employed along with statistical analysis using the chi-square test to identify the key risk factors. The results identified age, employment status, experience, paid leave, stress, satisfaction with income and life, and fatigue as key factors leading to accidents. Moreover, four binary machine learning classification models have been trained and tested to classify the taxi driver profiles and to analyze their risk factors. These models achieved 91.8% accuracy in identifying taxi drivers with a history of at least one accident. The high accuracy underscores the models’ effectiveness in identifying high-risk driver profiles, allowing taxi corporations to develop targeted interventions and training aimed at reducing accident rates and fatalities. Future research could explore psychophysiological factors and evaluate the effectiveness of targeted interventions.