This study aims to assess the metal fume exposure of welders and to determine exposure rates for similar exposure groups in a shipyard through the use of Near-field/Far-field (NF/FF) mathematical model and Bayesian decision analysis (BDA) technique. Emission rates of various metal fumes (i.e., total chromium (Cr), iron (Fe), lead (Pb), manganese (Mn), and nickel (Ni)) were experimentally determined for the gas metal arc welding and flux cored arc welding processes, which are commonly used in shipyards. Then the NF/FF field model which used the emission rates were further validated by welding simulation experiment, and together with long-term operation condition data obtained from the investigated shipyard, the predicted long-term exposure concentrations of workers was established and used as the prior distribution in the BDA. Along with the field monitoring metal fume concentrations which served as the likelihood distribution, the posterior decision distributions in the BDA were determined and used to assess workers' long-term metal exposures. Results show that the predicted exposure concentrations (Cp) and the field worker's exposure concentrations (Cm) were statistically correlated, and the high R2 (= 0.81-0.94) indicates that the proposed surrogate predicting method by the NF and FF model was adequate for predicting metal fume concentrations. The consistency in both prior and likelihood distributions suggests the resultant posterior would be more feasible to assess workers' long-term exposures. Welders' Fe, Mn and Pb exposures were found to exceed their corresponding action levels with a high probability (= 54%), indicating preventive measures should be taken immediately. The proposed approach provides a universal solution for conducting exposure assessment with usual limited number of personal exposure data.
© 2022. The Author(s).