Abstract: Due to the unique environmental characteristics of expressway tunnels, driving risks in these sections are significantly higher than those in regular road segments, with variations in driving behavior being a key factor contributing to the uneven distribution of risks. To investigate the influence of tunnel environments and driving experience on driving risk, this study focuses on typical expressway tunnels of varying lengths. Real-world vehicle experiments are conducted to collect speed data from different driver groups, including metrics such as average speed, speed deviation from the limit, sectional running speed difference, and acceleration. Factor analysis is employed to construct a quantitative model of driving risk, analyzing the distribution patterns of driving risk across different tunnel segments and variations among driver groups. The results indicate that driving risk exhibits a spatial distribution pattern of "middle segment>exit segment>entrance segment", and tunnel length is positively correlated with risk levels. Experienced drivers demonstrate higher driving risks compared to inexperienced drivers due to overconfidence. By quantifying the distribution patterns of driving risks and behavioral differences among drivers, this study provides a scientific basis for differentiated speed limit design and targeted driver training in expressway tunnels.

Key words: expressway tunnels; driving risk; factor analysis; driving experience

摘要： 高速公路隧道路段因其环境的特殊性，行车风险高于普通路段，而驾驶行为差异是导 致风险分布不均的关键因素。为探究隧道环境与驾驶经验对行车风险的影响机制，以典型高 速公路不同长度隧道为研究对象，通过实车试验收集不同类别驾驶人的车速数据，包括平均 车速、运行车速与限速差、区间运行车速差及加速度等指标，运用因子分析法构建行车风险 量化模型，分析隧道不同路段的行车风险分布规律及不同驾驶员行车风险差异。研究结果表 明，隧道路段行车风险呈现“中间段>出口段>入口段”的空间分布特征，且隧道长度与风险 值呈正相关；熟练驾驶人行车风险高于非熟练驾驶人。通过分析行车风险分布规律及驾驶人 行为差异，可为高速公路隧道差异化限速设计和驾驶人针对性培训提供科学依据。

关键词: 高速公路隧道；行车风险；因子模型；驾驶经验