The SOON project proposes to investigate the impact of using autonomous social agents to optimize industrial production processes within Industry 4.0. By the term "Social" it is understood that cyber-physical entities will belong to a network of connected entities that will act as autonomous agents with relationships similar to human socio-professional entities. This concept is inspired by human social networks and professional relationships. Currently, there are intelligent entities in Industry 4.0. In today's state-of-the-art approaches, intelligence, if implemented, is located at the individual level, and heterogeneous intelligent entities cannot effectively communicate and cooperate even if they have a close location. The main motivation of our research comes from the reason that if we want to create an authentic Internet of Everything that brings together processes, data, things and people, all these entities must be connected and follow a shared and easy to understand paradigm.
In this project, we propose a holistic multi-agent paradigm that includes cars and people. The presence of human operators is essential for both self-learning and learning from software agencies, through deep learning and data extraction algorithms. Agents will make decisions regarding the collection, fusion and analysis of large heterogeneous amounts of data produced by sensors (vibration, temperature, etc.), automation and information systems (such as enterprise resource planning and production systems) and people in real time. The main objective of the project will be to solve the difficult tasks of predictive maintenance. The design and evaluation of the SOON system will be carried out through predictive maintenance scenarios in collaboration with three industrial companies (from Slovakia, Spain and Switzerland). The choice of these three industrial partners, with a completely different industrial profile, was based on increasing the generality of our approach and proving the suitability for a greater diversity of industrial manufacturing processes. This collaboration will allow the project consortium to evaluate the concrete improvement of the specific industrial processes. We intend to demonstrate that the introduction of the Industry 4.0 revolution, combined with the recent improvements in machine learning, and the application of an autonomous multi-agent cooperative architecture can ultimately bring significant innovation in the optimization and modeling of industrial processes.
Main Objective
The objective of the SOON project is to investigate the impact of using autonomous social agents to optimize industrial production processes within Industry 4.0. The main direction of research of the project will be to solve difficult tasks of predictive maintenance.