Experimental Analysis of Machine Learning Algorithms used in Predictive Maintenance

In the field of engineering, various systems and components are added with cutting edge technology and complexity. Hence, it is important to know how these systems and components currently work but also how their performance degrades over time. This points towards the field of prognostics, that makes an attempt to predict the future of any system or component using present and past data of that system or component. The main hurdle in this field is to estimate the remaining useful life (RUL) of a system or component. RUL means how long the system's or component's functionality will last. Fault detection is one of the aspects of Prognostics or also referred to in Predictive Maintenance Field. In the proposed paper, for RUL estimation an appropriate algorithm has been put forth, and for Fault detection, the appropriate classification algorithm has been identified, for the respective mentioned datasets.