A quality-oriented digital twin modelling method for manufacturing processes based on a multi-agent architecture

The quality of a product is highly dependent on manufacturing processes. The recent development of industrial information technologies, such as Cyber-Physical Production Systems, Industrial Internet of Things, and Big Manufacturing Data Analytics has empowered the digitalization of manufacturing processes and promoted the concept of Digital Twin (DT). As one of the fundamental enabling technologies for Industry 4.0, DT enables the convergence between a physical system and its digital representation. DT modelling is the basis of implementing DT in practice. In this paper, we propose a DT modelling method based on a multi-agent architecture. It focuses on quality control during manufacturing processes and provides solutions to gather relevant information and analyze the corresponding influences on product quality. The MPFQ-model (Material, Production Process, Product Function/Future, Product Quality) is adopted to support the analysis of main influential factors related to the final product quality during the manufacturing phase. The five-dimension architecture is used as the basis for the DT models, including (i) physical entities, (ii) virtual models, (iii) DT data, (iv) services and (v) connections. Based on this architecture a Multi-Agent System (MAS) component and a semantic engineering component are integrated to create a quality-oriented DT framework.