The unsafe behavior of miners seriously affects the safety of deep mining. A comprehensive evaluation of miners’ unsafe behavior in deep coal mines can prevent coal mine accidents. This study combines HFACS-CM, SEM, and SD models to evaluate miners’ unsafe behaviors in deep coal mining. First, the HFACS-CM model identifies the risk factors affecting miners’ unsafe behavior in deep coal mines. Second, SEM was used to analyze the interaction between risk factors and miners’ unsafe behavior. Finally, the SD model was used to simulate the sensitivity of each risk factor to miners’ unsafe behavior to explore the best prevention and control strategies for unsafe behavior. The results showed that (1) environmental factors, organizational influence, unsafe supervision, and unsafe state of miners are the four main risk factors affecting the unsafe behavior of miners in deep coal mines. Among them, the unsafe state of miners is the most critical risk factor. (2) Environmental factors, organizational influence, unsafe supervision, and the unsafe state of miners have both direct and indirect impacts on unsafe behaviors, and their immediate effects are far more significant than their indirect influence. (3) Environmental factors, organizational influence, and unsafe supervision positively impact miners’ unsafe behavior through the mediating effect of miners’ unsafe states. (4) Mental state, physiological state, business abilities, resource management, and organizational climate were the top five risk factors affecting miners’ unsafe behaviors. Taking measures to improve the adverse environmental factors, strengthening the organization’s supervision and management, and improving the unsafe state of miners can effectively reduce the risk of miners’ unsafe behavior in deep coal mines. This study provides a new idea and method for preventing and controlling the unsafe behavior of miners in deep coal mines.

Based on prior research on the relationship between personality and safety behavior, we construct a moderated mediation model that tests the effects of each of the Big Five personality traits (openness, conscientiousness, agreeableness, extraversion, and neuroticism) on the safety behavior of gold miners in Ghana. The model included safety competency as a mediator between the five personality traits and safety behavior. Management commitment to safety was used as a moderator to test the strength of the mediation of safety competency. Data was collected from 344 individuals employed across six large-scale gold mining companies in Ghana using a questionnaire survey. Amos 26 was used to conduct confirmatory factor analysis. The hypotheses were tested using Hayes PROCESS macros models 4 and 7 on SPSS 26. Findings show that openness and extraversion have an insignificant direct influence on safety behavior. Neuroticism negatively affects safety behavior. In contrast, conscientiousness and agreeableness positively affect safety behavior. Mediation analysis revealed that safety competency partially mediates the relationships between (1) conscientiousness and safety behavior and (2) agreeableness and safety behavior. The relationship between extraversion and safety behavior is fully mediated by safety competency. Additionally, we discovered that management safety commitment has a significant conditional indirect effect (Index of moderated mediation = 0.168 95% CI = [0.122;0.222]) on the relationship between conscientiousness and safety behavior through safety competency. Management safety commitment also significantly moderated (Index of moderated mediation = 0.075 95% CI = [0.021;0.120]) the relationship between agreeableness and safety behavior through safety competency.