A Hybrid Peer-to-Peer Architecture for Agent-Based Steel Manufacturing Processes

• Published in: IFAC-PapersOnLine Vol. 54; no. 1; pp. 528 - 533
• Main Authors: Iannino, Vincenzo, Mocci, Claudio, Colla, Valentina
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 2021
• Subjects: agents; applications using mediators; engineering applications of artificial intelligence; factory of the future; hybrid peer-to-peer; industry 4.0; multi-agent simulation; multi-agent systems; steel industry; multi-agent systems; hybrid peer-to-peer; multi-agent simulation; industry 4.0; factory of the future; steel industry; engineering applications of artificial intelligence; agents; applications using mediators
• ISSN: 2405-8963; 2405-8963
• DOI: 10.1016/j.ifacol.2021.08.167

The new generation of steel manufacturing processes shaped by Industry 4.0 are more digitalized, networked, flexible and adaptable. Production processes use distributed information and communication structures, are more autonomous and capable to react to dynamic changings of the environment. Agent-based systems represent a paradigm, which is well suited to address these new generation of smart processes. The paper presents a hybrid peer-to-peer architecture for agent-based steel production processes. The architecture exploits a central database server for storing and retrieving updated information from peers about a cold rolling manufacturing process. The cold rolling process is modeled as a multi-agent system composed of four types of autonomous agents, each playing a different role in the steel production chain. Agents are designed to take autonomous decisions, and to coordinate and collaborate with each other, by ensuring the dynamic plant resources allocation even if unforeseen interruptions of the production flow may happen. The proposed approach is designed for the steel strip manufacturing process but can be easily readapted to any flat production process. The test of the design multi-agent system with the proposed architecture is supported though the simulation of the dynamic plant resources allocation under changing dynamic conditions.