An integrated life cycle inventory and artificial neural network model for mining air pollution management

It is necessary to trade off environmental impact and mining performance, preferably using an integrated life cycle modeling approach, while quantifying the air emissions. An integrated framework comprises of a life cycle inventory, and artificial neural network model for mining (LCAMM) is developed and applied to the real case study of an open-pit gold mine. This study aims to develop the air pollution inventory using inverse matrix method and estimation of midpoint impact assessment using defined characterization methodologies. Furthermore, this paper explores the feasibility of back-propagation artificial neural networking model to develop the carbon footprints analysis for the mine in terms of CO₂ equivalent by constructing different scenarios. The inventory result shows that TSP (49.6%), PM₁₀ (20%), PM_2.5 (14.7%), N₂O (7.1%), CO (6.1%), SO₂ (1.3%), HCN (0.6%), CH₄ (0.4%), CaO (0.2%) and VOCs (0.06%) are responsible for the total environmental load in this gold mining. Subsequently, particulate matter formation (kg PM₁₀ eq.) significantly contributes to the midpoint impact in contrast to the other environmental impacts, mainly because of hauling, handling of ore, milling/grinding, drilling and stockpiling. The study reveals that average 13,992.25–13,993.75 tons of CO₂ equivalent is produced as the carbon footprints for this mine. This study confirms that LCAMM framework can serve the basis for further analysis of pollutant's dispersion and assist decision makers to select an appropriate remedial technique in the mining industry.