Objective To establish a comprehensive quantitative assessment method for occupational hazard based on back propagation (BP)neural network.
Methods A comprehensive assessment index system containing 11 indicators was established after analyzing the key factors of occupational health status in chemical industry. The indicators were graded. Twenty-three samples of chemical projects in the city were collected, of which 18 samples were used to train the neural network and established a BP neural network model.
Results Five samples were used as test samples and fed into the input layer of the trained neural network. Furthermore, the output results of the 5 test samples showed good coincidence with the results rated by experts (100% correct).
Conclusion The neural network model can be used for quantitative assessment of occupational risk in chemical projects. The developed method provides new ideas in risk assessment of occupational hazard and an effective way for ranking management and risk control in chemical projects.
DownLoad: