This sub-project called BBI4.0 will develop signal processing modules bounded with data-driven Virtual Sensors and novelty/anomaly detection Machine Learning solutions, specifically addressed for the manufacturing industry. These building blocks for Predictive Maintenance applications are key tools for improving manufacturing Key Performance Indicators (KPI) such as the Overall Equipment Effectiveness (OEE). Developing powerful, tailored, and user-friendly tools is crucial to make Industry 4.0 a reality. BBI40 project is designed and executed by the Applied Artificial Intelligence (AAI) Unit of Eurecat Technology Centre. AAI develops innovative solutions (algorithms, methods, platforms) based on a combination of artificial intelligence and knowledge management technologies, especially for the industrial, energy and sustainability sectors.

​

Main objectives of the sub-project are: 

​

BBI4.0 propose to implement novel data-driven Predictive Maintenance (PM) solutions tailored to industrial applications. The proposed objectives and modules will extend the EFPF platform functionalities and demonstrate their capabilities by developing bespoke solutions for pilot use cases. The specific objectives are:

​

• Design and develop signal processing and pattern recognition modules and solutions customized to manufacturing sensor data.

• Design, develop and extend novelty/anomaly detection Machine Learning (ML) and Deep Learning (DL) based solutions for PM applications.

• Development of Virtual Sensors (VS) to contribute to a better understanding of the process as well as providing stability to the system.

 

Demonstrate and validate the BBI4.0 solutions developments through bespoke solutions for specific pilot datasets (both EFPF and EUT data).