Analysis of Machine Learning Techniques for Predictive Maintenance in Cooler Condition

Abstract

By exploiting the potential that machine learning has in predicting failures before they occur, a more robust maintenance plan can be planned, increasing operational efficiency, and saving expenses. Hence, utilizing machine learning techniques for predictive maintenance has become a primary focus in the field of facility management in the construction industry optimizing building efficiency with better decision-making. Nonetheless, to have an efficient system utilizing machine learning techniques, initially, an in-depth analysis of the common algorithms needs to be conducted to determine the efficacy of the available options. Therefore, this research focuses on analyzing common machine learning algorithms for supervised learning to predict cooler conditions for both classification and regression problems to determine the efficacy of the techniques.