Fraunhofer Institute for Production Technology IPTResearch project "Blockchain4DMP"
Great potential for added value through data-based applications
The networking of devices, machines and systems in the sense of the Industrial Internet of Things (IIoT) promises enormous value-added potential for production. Data-driven applications enable manufacturing companies to optimize their processes and products and significantly reduce costs:
Predictive quality and maintenance are intelligent manufacturing solutions based on artificial intelligence (AI) for the automated monitoring of production processes. These tools enable companies to predict and eliminate product defects early on, thus avoiding productivity losses and increasing product quality. The wear of critical machine components can be forecast, enabling manufacturers to carry out targeted maintenance and minimize downtime.
By developing data-based optimization and maintenance services, companies can innovate their business models and tap into additional sources of revenue.
Challenges of digitalization
However, for the digitalization of production to succeed, companies need extensive expertise in digital technologies and access to high-quality, diverse data. Training robust AI models requires sufficient data from diverse sources, which most companies lack. There is also a shortage of IT specialists in firms who can effectively integrate this data and develop digital applications.
Realizing data potential: more efficient and sustainable production thanks to IIoT data and service platform
In the “Blockchain4DMP” research project, we are working with our partners to develop a blockchain-based data and service marketplace that brings together historical production, quality and usage data from various manufacturing processes and entire process chains and makes it directly available to data scientists and industry experts. The secure coupling with the blockchain ensures comprehensive availability of IIoT data for analysis, ultimately facilitating data integration for digital service offerings on this platform.
The data supports the development of AI algorithms, e.g. for predicting tool wear and component quality, and helps companies identify optimization potential and improve the efficiency and sustainability of their production processes. We work closely with partners to implement specific use cases, including solutions for predictive maintenance and predictive quality, as well as models for smart city applications.
Use case predictive quality: AI optimization of milling aerospace components
Our AI model for predicting product quality is tailored to specific milling data and increases efficiency for a wide range of users. We use digital twin technology and physical modeling to predict the dynamics of workpieces, stabilize milling processes and improve the quality of thin-walled aero engine components. This involves simulations that take into account the changing thickness and thus the stiffness of thin-walled workpieces.
This application aims to reduce undesirable surface features caused by chatter. Chatter refers to self-excited vibrations in machining systems that commonly occur during cutting processes, affecting both efficiency and quality in production. By measuring profile deviations and surface roughness, we link vibrations to the accuracy of the produced parts and develop an AI algorithm to recognize chatter vibrations. As a result, manufacturers can adjust process parameters such as changing the spindle speed and reducing the cutting depth to make the machining process more stable and improve surface quality.
Ultimately, users can submit data, build models for their manufacturing processes, and execute quality predictions, ensuring real-time access to insights from milling operations.
Project partners
- dataMatters GmbH, Cologne
- Fraunhofer Institute for Production Technology IPT, Aachen
- gemineers GmbH, Aachen
- Chair for Information Systems and Databases at RWTH Aachen University
- Manufacturing Technology Institute – MTI at RWTH Aachen University
- Stadtmarketing- und Wirtschaftsförderungsgesellschaft Dormagen mbH
Funding note
The "Blockchain4DMP" project is funded by the Ministry of Economic Affairs, Industry, Climate Protection and Energy (MWIKE) of the state of North Rhine-Westphalia as part of the "STARK - Strengthening the Transformation Dynamics and Departure in the Mining Areas and at the Coal-Fired Power Plant Sites" funding program.
Funding reference: 46SK0233B
Project management agency
German Aerospace Center (DLR)