Predictive maintenance of electrical machining from the MANTIS project

Nugget 9 is about predictive maintenance of electrical machining from the Mantis project.

Predictive maintenance is and emerging need from the industry. There is no avoiding of maintenance because of physical reasons, but maintenance should be done right before a machine breaks. Five steps in
predictive maintenance are explained: gathering data, feature extraction, model selection, training models and integration. After gathering of data, reducing noise is important, because no data is perfect. An exercise is part of the nugget to do this.

Types of models (predict or detect) are explained, where no one model fits all purposes. For predictive and analytical models there is the risk of under- and over-fitting.Python practice is part of the learning.