Proposed autonomous agent by using process mining techniques in a Virtual Architecture based on Multi-Agent Systems

The increase of complexity in control production systems, currently requires more flexible and adaptable architectures to comply with demands during runtime. In this way, to apply the concepts of Industry 4.0 around the process management will have a significant impact on the added value, effectiveness and efficiency of the organizations involved with production. Therefore, the purpose of this research is to present an architecture proposal that uses the concept of Process Mining to enable the delivery, automatically, the improvement indicators, such as: reduction of machine downtime, readjustment of the production flow, maintenance costs and assertiveness in fault detection and prediction. This approach foresees the use of a distributed architecture, adhering to the RAMI 4.0 architecture. The advantage of this architecture is the flexibility and distributed e multi-agent control systems. Another advantage is the integration of the physical world and the virtual world with the concept of CPPS (Cyber-Physical Production Systems), thus, bringing flexibility, interoperability and adaptability, that are necessary for the production system.