中国危险品车辆驾驶人驾驶行为影响因素分析

An Analysis of Factors Influencing Driving Behaviors of Drivers of Dangerous Goods Vehicles in China

为探索中国危险品车辆驾驶人风险驾驶行为影响因素以及与小车驾驶人的差异性,采集50名男性危险品车辆驾驶人的DBQ问卷、PDBS问卷、DSI问卷、DAIS问卷、驾驶人基本信息问卷数据。借助Spearman相关性分析方法探索问卷中各变量的相关性,使用Mann-Whitney检验不同年龄、驾龄、攻击性、积极性4个维度在DBQ和DSI上的差异性。结果表明,相对于年龄,驾龄对驾驶技能的影响程度较小。年轻男性危险品车辆驾驶人的平均驾驶经验虽不如中年驾驶人,但行车更为谨慎,安全技能水平相对较高。与小车驾驶人不同,危险品车辆驾驶人的驾驶技能、异常驾驶行为与积极性没有明显相关性。攻击性行为频率较高的驾驶人出现违规操作、错误行为、失误行为的频率要高于低频组。利用Logistic二元回归分析建立了驾驶人攻击性预测模型,研究表明,限速50 km/h城区道路倾向行驶速度、违规操作总分、过去3年主动事故数可以作为驾驶人攻击性强弱预测模型的输入变量,依次输入模型后,模型卡方值改进为10.513,11.529,14.566,预测正确率依次为76.10%,80.40%,87.00%。整体模型Hosmer-Lemeshow检验值为4.415,最终驾驶人攻击性预测模型的预测正确率为87.00%。

To explore factors influencing risky driving behaviors of drivers of dangerous goods vehicles in China and their differences with drivers of sedan cars,the Driver Behavior Questionnaire(DBQ),Positive Driver Behavior Scale(PDBS),Driver Skill Inventory(DSI),Driver Aggression Indicators Scale(DAIS),and basic driver information questionnaire data of 50 male drivers of dangerous goods vehicles are collected.The Spearman correlation analysis method is used to study the correlation of each variable of the questionnaire.In addition,Mann-Whitney is used to identify the differences of age,driving experience,aggressiveness,and positivity between DBQ and DSI.The results indicates that,different from drivers of sedan cars,there is no obvious correlation among the driving skills,abnormal driving behaviors,and positivity of drivers of dangerous goods vehicles,and young male drivers of dangerous goods vehicles have better safety skills than middle-aged drivers.A logistic binary regression analysis is used to build a driver aggressiveness prediction model.The results shows that the preferred driving speed on the urban road with a speed limit of 50 km/h,the total score of illegal operations,and the number of active accidents in the past three years can be used as the input variables of driver aggressiveness prediction model.The chi-square values can be improved to 10.513,11.529,14.566,and the correct rate of predictions is 76.10%,80.40%,87.00%.The overall Hosmer-Lemeshow test value of this model is 4.415,the accuracy of prediction is 87.00%.