 This paper presents a novel approach to predicting blast-induced ground vibrations at the Micurana quarry. It uses a Bayesian neural network, BNN, which is a type of deep learning algorithm to analyze data from eight input parameters and one output. The results show that the BNN outperforms other machine learning algorithms, such as gradient boosting, k-nearest neighbor, decision tree, and random forest. Additionally, the SHAP values indicate that the BNN is more accurate than the other algorithms, because it takes into account all the features of the dataset. This article was authored by Yoho Halashet Fischa, Hajime Ikeda, Hisatoshi Torea, and others.