[Study on causative model of acute occupational poisoning accidents based on interpretative structural model-Bayesian network].

Objective: To further study the causes of acute occupational poisoning accidents, and to provide scientific basis and decision support for the prevention of accidents in advance. Methods: From September 2022 to May 2023, the literature was searched and 232 cases of acute occupational poisoning cases occurred from 2013 to 2022 were collected. The causal nodes of the accident were determined according to the expert score, and the interpretative structural model (ISM) was used to construct the correlation model between the causal nodes to obtain the hierarchical relationship between the factors. The influence of each causal node on the occurrence of acute occupational poisoning accidents was studied by using Bayesian network (BN), and the relationship and influence among the causal nodes were analyzed by Netica 5.18 software to establish the pre-prevention model of acute occupational poisoning accidents and identify the key causal factors. Results: A total of 23, 203, and 6 cases of significant, large, and medium acute occupational poisoning accidents were included, of which 179, 29, and 24 cases were asphyxiating gas, irritating gas, and mixed gas, respectively. ISM of acute occupational poisoning accidents divided the causal factors into a 7-layer and 3-level hierarchical structure model. Among them, operation conditions, protective measures, ventilation equipment, hidden trouble investigation, emergency management, illegal operation, equipment and facilities, and blind rescue were the direct causes of the occurrence and expansion of accidents. Warning devices, inspection situation, safety education situation, safety operation procedures, and technology in the production process were indirect influences that lead to the occurrence and expansion of accidents. Safety production responsibility system, enterprise supervision and management and government supervision were the deep-rooted influences. BN reasoning showed that the maximum probability causal chain of acute occupational poisoning accidents was as follows: safety production responsibility system→enterprise supervision and management→safety education and training→protective measures→accident occurrence. The key factors leading to the occurrence of acute occupational poisoning accidents were inadequate protective measures, equipment and facility failures, operational errors, ventilation equipment not being used properly and improper emergency management. Conclusion: In the prevention of acute occupational poisoning accidents, it is necessary to correctly use protective measures, test equipment and facilities before operation, operate according to regulations, ensure the normal use of ventilation equipment, and strengthen emergency management, so as to reduce the incidence of acute occupational poisoning accidents.