Research On Perception And Evaluation Of Psychological Situation Of Unsafe Behavior Of Construction Workers Based On User Portraits

At present,the frequent safety accidents in China ’s construction industry have not been effectively controlled,most of which are caused by the unsafe behavior of workers.In view of the fact that the psychological heterogeneity of construction workers leads to the differentiation of their unsafe behaviors,there is a significant conflict between the current extensive and universal safety management methods and the personalized safety management needs of workers,and it is difficult to solve the safety management deviation caused by the psychological heterogeneity of workers.Therefore,based on the psychological cognitive perspective of construction workers,this thesis describes,perceives and evaluates the psychological situation of workers’ unsafe behaviors from two psychological dimensions of personality traits and emotional state by using user portrait technology and artificial intelligence technology,and realizes the personalized safety management of construction workers according to the generated user psychological portrait.To this end,this thesis made the following research :(1)Building a psychological situation portrait label system for unsafe behavior.Semi-structured in-depth interviews were used to obtain the interview data of workers’ unsafe behavior.Nvivo software was used to analyze the psychological situation of unsafe behavior.By constructing the psychological cause model of unsafe behavior of construction workers,the psychological situation label was extracted and the portrait label system was built.(2)Constructing the mechanism for the role of psychological situation on unsafe behavior.Based on the label system of psychological situation portrait,a hazard cognition experiment was designed to explore the specific path of personality traits and emotions affecting the process of hazard cognition and thus acting on unsafe behavior,and to construct a mechanism model of psychological situation on unsafe behavior of construction workers from a cognitive perspective.(3)Depicting the portrait of the risk level of the psychological situation of unsafe behavior.Based on the mechanism of unsafe behavior psychological situation,a threedimensional risk matrix model was constructed to evaluate the risk level of unsafe behavior psychological situation.By exploring the mapping relationship between each risk level and unsafe behavior psychological situation,the risk level portrait of unsafe behavior psychological situation was depicted and modeled.(4)Constructing a prewarning mechanism for psychological situation assessment of unsafe behavior.Based on the psychological situation characteristics of each risk level portrait,taking the psychological situation label as the input end and the risk level category as the output end,the optimal prediction classification algorithm model was obtained by comparing the prediction performance of eight supervised learning algorithms.The evaluation and prewarning mechanism of unsafe behavior psychological situation was constructed,and a personalized safety management scheme integrating personality traits and emotions was proposed.(5)Constructing a psychological situation assessment system for unsafe behavior.Based on the prewarning mechanism of psychological situation assessment of unsafe behavior,the personality files,emotion recognition,assessment and prewarning,portrait generation and corresponding personalized control measures are intelligently integrated to construct the psychological situation assessment system of unsafe behavior.Through intelligent perception of the psychological situation of unsafe behavior of construction workers,this thesis carries out differentiated behavior control,focuses on solving the imbalance of behavior management caused by the heterogeneity of individual psychological situation,and provides new ideas for realizing accurate prewarning,personalized intervention and correction of unsafe behavior of high-risk workers,which is conducive to improving the safety management level of construction enterprises.There are 93 pictures,69 tables and 133 references in this thesis.