Predictive Maintenance using Clustering Methods for the use-case of Bolted Connections in the Automotive Industry

Abstract

The paper explores the feasibility of using clustering for the predictive maintenance of nutrunners in a modern high-volume manufacturing environment. During the period of the study, the failure rate of one nutrunner was significantly higher than that of the others. The clustering algorithms that were evaluated were agglomerative hierarchical clustering (AHC), density-based spatial clustering of applications with noise (DBSCAN), and a self-organising feature map (SOFM). The performance metrics used to compare and evaluate the clusters were the silhouette coefficient score (SC) and the variation rate criterion (VRC). It was found that it is feasible to use clustering to improve the maintenance strategy of nutrunners in the automotive industry.