Problems and challenges regarding the digital twin

However, the data occurring within the software and hardware components has so far been used almost exclusively to fulfill the individual task of the individual component. Process design, manufacturing and quality assurance nowadays work on different, only extremely incomplete digital twins. An integrated and cross-system solution for linking planning, production and quality in the sense of the Industrial Internet of Things (IIoT) and a profound digital twinning based on it, which is data-consistent and comprehensive, does not exist so far.

A digital twin framework for the machining domain

dPart® describes a domain-specific framework for digital twinning in the metalworking industry and its technologies. It includes state of the art big data approaches (e.g. lambda architecture) and considers current national and international activities in the area of standardization (e.g. ISO23247 – A digital twin framework for manufacturing). The framework as well as the implementation are divided into four main development areas:

• Process design: Within dPart a new generation of CAM systems for process planning in the metalworking industry is being developed. The system is able to perform extensive technology simulations, to instantiate additional computing power in the edge or cloud, to add suitable microservices such as AI functionalities. The data generated within the CAM system is fed into the end-to-end digital twin of the value chain and stored in the cloud in suitable database solutions.
• Manufacturing: In the area of manufacturing, dPart focuses on the implementation of suitable data acquisition, data processing and data storage solutions. Control-specific acquisition solutions are developed for the most important control types. The acquired machine and sensor data is transferred to suitable data processing solutions on edge or cloud level. The data is translated into process and quality data via domain specific technology microservices and stored in the end-to-end digital twin of the value chain.
• Quality assurance: The digital twin of the planning and manufacturing phase is transferred to quality assurance together with the physical product after completion of production. The data determined within quality assurance is compared with process and quality data determined within planning and manufacturing. AI-based long-term data analyses enable the identification of correlations and causalities for possible productivity and quality problems.
• IIoT, data economy and data sovereignty: The main innovation driver for the digital twinning along the value chain of planning, production and quality assurance is their networking in the sense of the IIoT. The fourth development area covers aspects such as hardware infrastructure, software orchestration and data semantics. It addresses the development of digital business models based on the digital twin as well as secure data handling in terms of data sovereignty.

Benefit of the framework for the manufacturing industry

dPart enables the creation of a complete and data-consistent digital twin along the entire value-added chain of planning, production and quality assurance. dPart thus reduces interface problems of the metalworking industry within manufacturing, increases the digital resilience of the industry and prepares it for the Industrial Internet of Things and a future data economy.

The Fraunhofer IPT and the MTI of the RWTH Aachen University are continuously developing dPart together with small and medium-sized enterprises as well as large companies in the metalworking industry. The developments are supported by suppliers and subcontractors of software and hardware components for manufacturing systems in metalworking. The digital infrastructure is covered by providers and suppliers of digital infrastructure such as edge or cloud platform providers.