INSE6310Project

In predictive maintenance previous data of the equipment is analyzed to monitor its state. The main objective of this type of maintenance is to predict and eventually prevent system failures. Using machine learning techniques in predictive maintenance can help us to reduce the guesswork and predict future failure of the equipment with more accuracy. Usage history data is an important indicator of equipment condition that is essential. In this project, we used synthetic data to use it for training our machine learning algorithm for predictive maintenance. First, we described and visualized the data using Exploratory Data Analysis to get a good understanding of it. We used PyCaret library to find the best performing machine learning algorithm. Light gradient Boosting Machine came out as the best classifier and regressor.