To explore the influencing factors of occupational stress and their interrelationships among frontline workers in the secondary industry, so as to provide a reference basis for targeted intervention of occupational stress.

1 227 frontline workers in the secondary industry of Nanchong City were selected as study subjects by stratified random sampling. The Core Occupational stress and the Occupational Health Literacy Questionnaire of National Key Population were used to investigate occupational stress and occupational health literacy. SPSS 27.0 and R 4.1.2 were used to screen the influencing factors, Netica 7.01 constructed a Bayesian network model with inference and sensitivity analyses, and the area under the curve (AUC) assessed the model fit.

370(30.2%) were found to have occupational stress among frontline workers in the secondary industry in Nanchong City. LASSO regression analysis screened out 12 influencing factors, which were gender, marital status, health status, age, education, monthly income, exercise, occupational health literacy, enterprise size, nature of the enterprise, weekly working hours, and night shifts. The occupational stress Bayesian network model fit well (AUC=0.865), and backward inference yielded six key influences, ranked by sensitivity analysis: monthly income (57.9%), weekly working hours (55.4%), health status (52.6%), occupational health literacy (33.4%), enterprise size (20.4%), and exercise (15.5%).

The Bayesian network model constructed in this study has good prediction performance for occupational stress of frontline workers in secondary industry. It should focus on helping small and micro-enterprises, pay attention to low-income and poor health workers, and suggest rationalizing working hours, improving occupational health literacy, and strengthening physical exercise to prevent and control the occurrence of occupational stress.