A Survey on Predictive Maintenance Through Big Data

Modern manufacturing systems use thousands of sensors retrieving information at hundreds to thousands of samples per second. The real time data being generated is mostly used for monitoring the processes and the equipment condition. Data processing techniques applied to this data to detect anomalies and thus applying preventive maintenance have been used in the industry. Currently available technologies which were developed during the last two decade for scanning the Internet and providing computational services, working at very large scale can be re-targeted to fulfil the requirements of maintenance of complex systems. These systems can support storage and processing of current as well as historical data. Ability to access and process these large data sets will lead from preventive to predictive maintenance and eventually to smart manufacturing. © 2016, Springer International Publishing Switzerland.