| In the traffic system composed of human vehicle road environment,the driver is the main body of the system and the core of ensuring road traffic safety.In recent years,the research on active traffic safety management system has gradually become the mainstream,especially for the behavior characteristics of large operational drivers such as heavy trucks.Therefore,this paper takes the behavior and safety risk of heavy truck drivers as the research object,designs the driving behavior system scale suitable for heavy truck drivers in China,analyzes the driving behavior characteristics and main influencing factors of heavy truck drivers,studies the evaluation method of driving behavior safety level of heavy truck drivers,and the traffic accident risk prediction model of heavy truck drivers is constructed.In order to improve the driving behavior of heavy truck drivers,the control of traffic accident probability of heavy truck drivers is transformed into the control of its main influencing factors.It enriches the theory of driver’s driving behavior,and provides ideas and experimental methods for the analysis of heavy truck driver’s safe driving behavior and safety risk research.Firstly,on the basis of summarizing and analyzing the research progress of heavy truck operation safety,driving behavior characteristics,driving behavior safety evaluation method and traffic accident risk prediction model,the research purpose,method and innovative ideas were clarified.The definition of driving behavior was expanded,the driving behavior was classified,and the definition relationship between various related concepts was sorted out;Based on traffic psychology and Heinrich’s "domino" causal chain theory,combined with the statistical data of heavy truck traffic accidents,this paper deeply analyzed the driving behavior characteristics of heavy truck drivers and the causes of traffic accidents,and constructs the representation framework of heavy truck drivers’ driving behavior.Secondly,based on the theory of statistics and psychology,used the methods of literature analysis,questionnaire survey,exchange interview,observation and exploratory factor analysis,this paper designed a heavy truck driver driving behavior system scale suitable for the characteristics of heavy truck drivers in China,analyzed the performance characteristics of heavy truck drivers’ driving behavior,and verified the rationality and effectiveness of the scale.Then,correlation analysis,multiple correspondence analysis and one-way ANOVA were combined to analyze the driving behavior characteristics of heavy truck drivers with different attributes.Based on the theory of planned behavior,used the path prediction method of structural equation model(SEM),this paper constructed the path integration model of driving behavior influencing factors of heavy truck drivers,comprehensively analyzed the driving motivation factors of heavy truck drivers,and found out the key factors and influence degree of driving behavior of heavy truck drivers.Furthermore,based on the driving behavior scale,the driving behavior safety index system of heavy truck drivers was constructed from five aspects: positive behavior,general violation behavior,wrong driving behavior,careless driving behavior and risk / aggressive driving behavior.This paper proposes the method of combining the principal component analysis and entropy method to weight the driving behavior safety evaluation indexes,and the method of combining the grey correlation degree and TOPSIS comprehensive evaluation method to build the evaluation model of driving behavior safety level of heavy truck drivers,so as to realize the comprehensive evaluation of driving behavior safety level of different groups of heavy truck drivers.Finally,based on the heavy truck driving behavior scale and its statistical characteristics,the attributes and driving behavior of heavy truck drivers were taken as independent variables,and whether heavy truck accidents occur was taken as dependent variable.Correlation analysis,multiple linear regression analysis and single factor Logistic regression analysis were used to screen the independent variables for traffic accident risk modeling of heavy truck drivers.A binomial logistic regression model was proposed to predict the traffic accident risk of heavy truck drivers,and a binomial logistic regression model was constructed to explain the influence of the correlation characteristics between drivers and their driving behavior parameters on the traffic accident risk prediction results of heavy truck drivers.It also analyzed the factors that affect the traffic accidents of heavy truck drivers,and put forward the measures and suggestions to improve the driving behavior of heavy truck drivers and prevent accidents.The research results enrich the theory of driving behavior of heavy truck drivers,help to improve the importance of the research on driving behavior of heavy truck drivers,provide theoretical basis for the correction of improper driving behavior and the cultivation of positive driving behavior of heavy truck drivers,and provide guidance for the formulation of relevant safety measures policies or regulations and the development of heavy truck collision warning system.It can improve the driving behavior of heavy truck drivers,improve driving safety,and prevent or reduce heavy truck traffic accidents. |