Abstract: In order to improve the sense of occupational safety responsibility of coalmine workers and reduce the accident rate,the thesis take advantage of the rough set theory and SPSS software,which were used for correlation analysis to select eight significant threelevel human factor risk factors.A human factor risk prediction model based on logistic regression analysis was established,and three significant risk factors (cultural level,mood swings and technical proficiency) were analyzed.On the same time,the results show that the significant level of goodness of fit coefficient of the model is far greater than 0.05,and the prediction accuracy of the measured data is 80.5%.With the help of scientific calculation,the application necessity of emotion management in safety work is obtained,which provides valuable reference for enterprise management.

Key words: Dynamic timevarying system,Unsafe behavior,Human factors Logistic,Regression model