Probabilistic risk-based pollution prevention model for a foundry: a case study of casting

Hazardous air pollutants from industrial activities have long been associated with serious health effects. Traditional health risk assessment uses point estimates of inhalation concentrations based on standard Gaussian diffusion models with steady-state emission rate assumptions. This traditional approach was criticized because it does not account for variability and thus leading to a potential overestimate of the health risk from the batch processes. To overcome this deficiency, a probabilistic risk assessment model is proposed. The foundry industry with processes typically associated with several hazardous air pollutants is identified and iron casting is chosen as a case study to compare risk estimates. Existing data, representing historical proprietary information of the case study, were used to deliver representative risk values and help identify potential replacements or interventions in the manufacturing process. A probability distribution function of emitted concentrations was simulated to model the batch process emissions from mold and core resin binders, a major source of pollution. The same method was applied to exposure factors to feed into the risk model resulting in a probabilistic risk evaluation. Several alternative resin binders in commercial use were examined to offer a risk-based substitute to the resin binder in use. The risk results provided an opportunity to consider newer and environmentally friendlier options. A comparison of the results from this approach and those from the point estimate analysis reveals a gross over estimation of risks. The point estimate risk values were about eight time larger than the mean value and about twice the 95th percentile values of the probabilistic risk approach. The wide range of variability among resin binders associated risk results, close to two orders of magnitude in some cases, presented opportunities to select from a variety of binders with lesser emissions and lower risk. Optimal selection will depend on several pollutants emitted from this process to help address cumulative impacts of multiple pollutants. Investigations are underway for a multi-pollutant strategy including trade-offs, and other quality controls vital to the decision-making.