The optimization of blasting patterns involves the strategic adjustment of blast design parameters with the goal of achieving optimal fragmentation, thereby minimizing operational costs in mining and mitigating associated environmental impacts. The objective is to concurrently minimize operating costs from the mine to the crusher and address the repercussions of blasting, encompassing fly-rock and back-break. To fulfill the study’s objectives, a multi-variable regression model was developed to depict total costs spanning drilling to crushing. Beyond cost considerations, multilayer perception neural networks were implemented to predict blast-induced back-break and fly-rock. The main novelty of this work is the unified integration of cost prediction and MLPNN-based consequence prediction within a multi-objective GOA to deliver Pareto-optimal blast designs that explicitly quantify trade-offs between mine-to-crusher costs and blast-induced fly-rock/back-break. The precision of estimations for both back-break and fly-rock reached an average coefficient of determination of 99% across training, testing, and validation datasets. Subsequently, the Grasshopper Optimization Algorithm is used to determine the optimal blast design while adhering to practical constraints. The results of the optimization model yielded a Pareto set of solutions, allowing the mining operation management team to select any solution based on their strategic preferences. Notably, the blast pattern with the lowest cost exhibited relatively high fly-rock and back-break, while opting for a pattern with minimal fly-rock and back-break resulted in a 20.13% increase in costs compared to the minimum cost blast design.
Jian et al. (Mon,) studied this question.