Automatic model for detecting anomalies and reading errors with machine learning, which identifies conditions that may lead to machine stoppages and/or manufacturing in conditions classified as non-optimal.


Detection of anomalous behaviors unknown until now..


Increase the competitiveness of production processes through technological modernization and automation. Greater customer satisfaction, a better brand image and an increase in the competitiveness of the companies that apply it and, by extension, of the industry in the global market.


The “Predictive Maintenance of Industrial Machinery from Sensory Data – MADAM” project works with rolling machinery data from the point of view of the machinery manufacturer and associated sensors and users. The objectives of the project are the reduction of waste, the improvement of product quality, the optimization of machinery, the increase of productivity and the saving of time and costs associated with repairs, maintenance, interruptions and machine shutdowns .


Data has been collected from several rolling machines that have made it possible to identify the main causes of stoppages and implement preventive measures. Thanks to this information, an interface has been developed that allows you to view the data and status of the devices. Finally, a validation of the system and its results has been carried out through industrial tests at the RotorPrint facilities.

Press release