Scholarlink

<p>This study aims to enhance road safety by understanding pedestrian-vehicle dynamics, analysing 452 pedestrian-vehicle interactions at a midblock crossing in New Delhi, India using Binary Logistic Regression. The results show that drivers yield in 67.92% of total interactions. Key findings indicate that several pedestrian characteristics are highly predictive of a driver’s decision to yield, highlighting strong associations with yielding behaviour. Pedestrians using gestures (<i>β</i> = 1.619, OR = 5.048) to signal their intention had 5 times higher odds of being yielded to than those who did not use gestures. Similarly, female pedestrians (<i>β</i> = 0.919, OR = 2.506), those wearing coloured clothing (<i>β</i> = 0.914, OR = 2.494), crossing in a group (<i>β</i> = 0.214, OR = 1.239) and pedestrians carrying baggage (<i>β</i> = 1.119, OR = 3.074) had significantly higher odds of being yielded to than their respective reference categories. Pedestrians who were distracted (<i>β</i> = −&#xa0;0.436, OR = 0.647) and &gt; 60&#xa0;years old (<i>β</i> = −&#xa0;1.524, OR = 0.217) had lower odds of being yielded to than other variables. The model’s mean cross validated F1-score was 0<i>.</i>80 ± 0<i>.</i>04 for the Not Yield class and 0<i>.</i>86 ± 0<i>.</i>02 for the Yield class. These findings highlight the critical role of non-verbal communication and pedestrian vulnerability in driver yielding behaviour. The study provides evidence-based information for policy makers, urban planners, and law enforcement agencies, suggesting that interventions will aim at encouraging positive pedestrian behaviour and driver awareness of vulnerable road users that could lead to increased safety at midblock crossings.</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

The impact of pedestrian characteristics on driver yielding behaviour in mixed traffic conditions

  • 期刊论文

  • 发表日期：

    2026年2月17日

查看全文
  • Raja Adnan Khan,
  • Farhan Ahmad Kidwai,
  • Shamshad Ahmad

摘要

This study aims to enhance road safety by understanding pedestrian-vehicle dynamics, analysing 452 pedestrian-vehicle interactions at a midblock crossing in New Delhi, India using Binary Logistic Regression. The results show that drivers yield in 67.92% of total interactions. Key findings indicate that several pedestrian characteristics are highly predictive of a driver’s decision to yield, highlighting strong associations with yielding behaviour. Pedestrians using gestures (β = 1.619, OR = 5.048) to signal their intention had 5 times higher odds of being yielded to than those who did not use gestures. Similarly, female pedestrians (β = 0.919, OR = 2.506), those wearing coloured clothing (β = 0.914, OR = 2.494), crossing in a group (β = 0.214, OR = 1.239) and pedestrians carrying baggage (β = 1.119, OR = 3.074) had significantly higher odds of being yielded to than their respective reference categories. Pedestrians who were distracted (β = − 0.436, OR = 0.647) and > 60 years old (β = − 1.524, OR = 0.217) had lower odds of being yielded to than other variables. The model’s mean cross validated F1-score was 0.80 ± 0.04 for the Not Yield class and 0.86 ± 0.02 for the Yield class. These findings highlight the critical role of non-verbal communication and pedestrian vulnerability in driver yielding behaviour. The study provides evidence-based information for policy makers, urban planners, and law enforcement agencies, suggesting that interventions will aim at encouraging positive pedestrian behaviour and driver awareness of vulnerable road users that could lead to increased safety at midblock crossings.