Predictive analysis of miners' group unsafe behavior based on group dynamics and institutional environment.

Coal mining is a typical high-risk industry, and the unsafe behavior of miners is the main reason for the high incidence of coal mining accidents. To prevent and control miners' group unsafe behavior, a research method combining survey design, Necessary Condition Analysis (NCA), and Fuzzy-set Qualitative Comparative Analysis (fsQCA) was used to explore the combined effects of multiple antecedent variables of group unsafe behavior and their complex causal relationships with outcomes from the perspectives of group dynamics and institutional environment. The results showed that individual group dynamics or institutional environmental factors do not constitute necessary conditions for group unsafe behavior among miners, but multiple equally effective causal configurations exist. High unsafe behavior among miner groups can be categorized into four types: multidimensional suppression type, lack of safety goals under high-pressure type, lack of safety culture under high-pressure type, and scattered type. Non-high unsafe behavior among miners can be categorized into two kinds: unity and cooperation, as well as cohesion, goal, and culture-driven behavior. The causal paths of high and non-high unsafe behavior exhibit asymmetry. The most influential path leading to high unsafe behavior among miner groups was configuration 1, which was ~ GC1 ×  ~ GC2 ×  ~ GPS ×  ~ GSO ×  ~ GSC; configuration 6b, which was GC1 ×  ~ GP × GPS × GSO × GSC, was the most effective in suppressing high unsafe behavior among miner groups, with a raw coverage rate of 0.444. Based on the results, intervention measures were proposed to address the key predictors of high unsafe behaviors in miners' groups. The study accurately identifies the grouping paths of "high unsafe behaviors" and "non-high unsafe behaviors" in the miners' group, providing new avenues for mine safety management.