【作者】 王萍；

【导师】 席建锋；

【作者基本信息】 吉林大学 ， 交通信息工程及控制， 2023， 硕士

【摘要】 在公共交通不断完善和发展的过程中,由于公交司机不良心理导致的交通安全问题逐渐显现。在各个行业中,公交驾驶人是一种高压力、高风险的职业,该职业群体由于工作压力导致的不良心理问题日趋严重,而公交驾驶人将这种不良的心理压力带到工作中,不仅会降低公交行业服务质量,还会给公共交通安全营运带来隐患。因此,公交驾驶人存在的不良心理问题不容忽视,运用专业理论及方法分析不良心理症状和交通安全之间的关系,并找出导致公交驾驶人产生不良心理症状的工作压力来源,以针对性地帮助驾驶人排解心理压力和负面情绪,提高公交驾驶人的心理健康水平,这对于提高公交运营安全具有重大的现实意义。论文以吉林省某公交企业的驾驶人为研究对象,采用单因素分析和二元logistic回归确定公交驾驶人易发生事故的危险因素,提出了公交驾驶人心理健康与交通安全评估的贝叶斯网络模型,分析出人口学特征、不良心理和性格的不同严重程度对公交驾驶人发生交通事故的概率大小,以及发生交通事故的关键因素和组合因素,同时,为了找到导致公交驾驶人产生不良心理症状的工作压力因素,构建了工作压力与不良心理症状关系的结构方程模型,以确定工作压力与不良心理状态的关系,同时定量分析工作压力对公交司机不良心理症状的路径系数作用程度。首先,将人口学特征问卷、SCL-90量表和Y-G量表结合,从心理状态和心理调节能力两方面评估公交驾驶人易发生交通事故的因素,通过单因素分析初步找到公交驾驶人易发生交通事故的影响因素,使用二元logistic回归分析最终确定躯体化、抑郁、焦虑等七个因素易导致公交驾驶人发生交通事故。其次,将二元logistic回归分析的七个影响因子进行分级并作为节点变量,结合专家知识与k2算法构建公交驾驶人心理健康与交通安全评估的贝叶斯网络结构,使用Netica软件中Incorp Case File模块进行参数学习,采用Netica软件对模型进行推理分析,通过事故预测识别出显著增加事故概率的关键因素与组合因素。事故原因诊断的结果显示,最可能的事故原因是抑郁中度、焦虑重度、躯体化轻度。贝叶斯网络模型的ROC曲线下面积为0.88,表明该模型具有较高的使用价值。最后,为了分析出导致公交驾驶人不良心理的工作压力源,保证公交车的道路安全,构建公交驾驶人工作压力与不良心理症状关系的结构方程模型,分析工作压力与不良心理之间的作用关系,并且通过模型的路径显著性检验确定导致公交驾驶人产生不良心理的工作压力源以及工作压力源对不良心理症状的影响程度。论文研究方法及结论不仅有助于企业更好地判别公交驾驶人的极端性格与不良心理状态,而且可以从工作压力源的角度采取有效干预措施排解公交驾驶人的心理压力,提高公交驾驶人的心理健康水平,进而提高公共交通的安全性。更多还原

【Abstract】 In the process of continuous improvement and development of public transport,traffic safety problems caused by bus drivers’ bad psychology gradually appear.In various industries,the bus driver is a kind of high pressure,high risk occupation,the occupation group due to work pressure caused by the increasingly serious psychological problems,and the bus driver will bring this bad psychological pressure to work,not only will reduce the quality of public transport industry service,but also bring hidden dangers to the safety of public transport operation.Therefore,the bad psychological problems of bus drivers can not be ignored,the use of professional theories and methods to analyze the relationship between bad psychological symptoms and traffic safety,and find out the source of work pressure leading to bad psychological symptoms of bus drivers,in order to help the driver to relieve psychological pressure and negative emotions,improve the level of mental health of bus drivers.It is of great practical significance to improve the safety of bus operation.Taking the drivers of a bus enterprise in Jilin Province as the research object,the thesis uses univariate analysis and binary logistic regression analysis to determine the risk factors of bus drivers’ prone accidents,and puts forward the Bayesian network model of bus drivers’ mental health and traffic safety evaluation.The different severity of demographic characteristics,bad psychology and personality on drivers’ probability of traffic accidents as well as the key factors and combination factors of traffic accidents are analyzed.At the same time,in order to find out the work stress factors that cause bus drivers to produce bad psychological symptoms,a structural equation model of the relationship between work stress and bad psychological symptoms is constructed.The path coefficient effect of work stress on bus drivers’ bad psychological symptoms is quantitatively analyzed.Firstly,the demographic characteristics questionnaire,SCL-90 scale and Y-G scale are combined to evaluate the factors of bus drivers’ susceptibility to traffic accidents from the aspects of mental state and mental adjustment ability.Through single factor analysis,the influencing factors of bus drivers’ susceptibility to traffic accidents are preliminarily found.Binary logistic regression analysis is used todetermine seven factors,including somatization,depression and anxiety,that lead to traffic accidents of bus drivers.Secondly,the seven influencing factors of binary logistic regression analysis are graded and used as node variables.The Bayesian network structure of bus driver mental health and traffic safety assessment is constructed by combining expert knowledge and k2 algorithm.The Incorp Case File module in Netica software is used for parameter learning.Netica software is used to analyze the model,and the key factors and combination factors that significantly increase the probability of accidents are identified through accident prediction.The results of the diagnosis of the accident cause show that the most likely causes of the accident are moderate depression,severe anxiety and mild somatization.The area under ROC curve of Bayesian network model is 0.88,which indicates that this model has a high value of use.Finally,in order to analyze the work stressors that cause bus drivers’ bad psychology and ensure bus road safety,a structural equation model of the relationship between bus drivers’ work stress and bad psychological symptoms is constructed.The relationship between job stress and bad psychology is analyzed,and the route significance test of the model is used to determine the job stressors leading to bad psychology of bus drivers and the degree of influence of job stressors on bad psychological symptoms.The research methods and conclusions of the thesis not only help enterprises to better identify the extreme personality and bad psychological state of bus drivers,but also can take effective intervention measures from the perspective of work stress sources to relieve the psychological pressure of bus drivers,improve the level of bus drivers’ mental health,and improve the safety of public transport.更多还原