Selection Dominant Features Using Principal Component Analysis for Predictive Maintenance of Heave Engines

Abstract

This article aims to identify the dominant features that have a significant impact on the health of a heavy machine that relates to the digital infrastructure of a company. The importance of this research is that the authors define predictive maintenance based on Principal Component Analysis (PCA), which is the novelty of this article. The novel contribution of this research lies in the application of Principal Component Analysis (PCA) for predictive maintenance of heavy machinery, which has not been integrated into the Scheduled Oil Sampling (SOS) procedures. The recorded data are called Scheduled Oil Sampling (SOS) and historical data from an equipment called CoreDataQ, which works for recording many features from heavy machine activities. The data contain two sets data. The method is Principal Component Analysis (PCA). This method leads to obtain a maximum of 20 significant features on data based on SOS. The results have been confirmed and agreed upon by the manager who owned CoreDataQ to consider the selected dominant features for further related maintenance.

Keywords

• pca
• sos
• coredataq
• predictive maintenance.