USING MACHINE LEARNING TO EVALUATE COAL GEOCHEMICAL DATA WITH RESPECT TO DYNAMIC FAILURES - SME Annual Conference 2023

In this study, NIOSH researchers conducted a machine learning analysis to examine whether a model could be constructed to assess the probability of dynamic failure occurrence based on geochemical and petrographic data. Random forest, cluster analysis and dimension reduction were all applied. The objective of dimensionality reduction was to explore patterns and groupings in the data and search for relations between compositional parameters. Cluster analyses were performed to determine if an algorithm could find clusters with given class memberships and to what extent misclassifications of dynamic failure status occurred. A random forest analysis performed on data from the Pennsylvania Coal Sample Databank cross-referenced with accident data from the Mine Safety and Health Administration (MSHA) determined that 7 parameters of the 18 examined exerted the most influence on results. Cluster analysis on data after dimensionality reduction resulted in a hierarchal clustering algorithm finding four clusters, with one relatively distinct dynamic failure cluster, and three clusters consisting mostly of control group members but with a small number of dynamic failure members.