Digital transformation of manufacturing. Industry of the future with cyber-physical production systems

• Autor(a) principal: Borangiu, Theodor
• Data de Publicação: 2020
• Outros Autores: Morariu, Octavian, Răileanu, Silviu, Trentesaux, Damien, Leitão, Paulo, Barata, José
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/148749
• topic: Cloud manufacturing; Cloud services; Cyber Physical Production System; Digital manufacturing; Digital twin; Holonic manufacturing control; Industrial Internet of Things; Machine learning; Multi-Agent System; Real-time data analysis; Resource virtualization; Computer Science(all)

This paper analyses the main research direction for the digital transformation of manufacturing and its important drivers: cloud services and resource virtualization that have led to the new Cloud Manufacturing (CMfg) model – an industrial replica of Cloud Computing. This model is adopted for the higher layer of the Manufacturing Execution System (e.g. the centralised, hierarchical System Scheduler), while its lower layers distribute intelligence through agent- and service orientation in the holonic paradigm. In this approach, Intelligent Manufacturing Systems are assimilated to Cyber-Physical Production Systems in which informational and operational technologies are merged, the shop floor physical reality being mirrored by virtual counterparts – the digital twins that represent abstract entities specific for the manufacturing domain: products, orders and resources. Industry 4.0 represents the vision for the Industry of the Future which is based on Cyber-Physical Production Systems that assure the flexible, dynamically reconfigurable control of strongly coupled processes. In this picture of the future manufacturing industry, the Industrial Internet of Things framework provides connectivity and interoperability to integrate communications between different kinds of things: products, orders and resources and legacy manufacturing devices to the web service ecosystem. The paper describes the scientific issues related to big data processing, analytics and intelligent decision making through machine learning in predictive resource maintenance, optimized production planning and control, lists solutions and proposes new research directions.