DefinePK

Abstract

Data analytics in mechanical engineering ,in particular, has changed the way things are made and maintained by giving people a way to make decisions based on data and predictions.  In the past few years, mechanical systems have been able to collect a lot of operational data from sensors, controllers, and monitoring devices. This raw data can then be turned into useful information.  This review compiles the open access research on applying data analytics to improve mechanical systems with the help of predictive maintenance, problem detection, and process optimization. This research focuses on data preparation, feature extraction, statistical model development, and the application of predictive frameworks to enhance asset reliability and production efficiency. It also examines current trends such as digital twin integration, real time monitoring, and hybrid analytics, which integrate machine learning with physical modeling. This paper examines previous studies and identifies some major issues, such as data inconsistencies, insufficient error labels, and scaling difficulties. Then, it suggests ways to fix these problems. Overall, this research gives mechanical engineers a good overview of how to use data analysis to get better results, be more reliable, and become sustainable in factories.