Model-Driven Approach for Realization of Data Collection Architectures for Cyber-Physical Systems of Systems to Lower...

Area of Science:

• Engineering
• Computer Science
• Industrial Automation

Background:

• Data collection is crucial for Industrie 4.0 initiatives like product quality optimization.
• Implementing data collection architectures faces challenges due to system heterogeneity and complexity.
• Significant manual effort is currently required for developing these architectures.

Purpose of the Study:

• To present a model-driven approach for automated generation of data collection architectures.
• To reduce the manual implementation effort in setting up data collection systems.
• To facilitate broader access to industrial data for analysis and optimization.

Main Methods:

• Developed a model-driven approach using formalized models and a graphical domain-specific language.
• Utilized model transformations to automatically generate source code for data collection architectures.
• Integrated support for various Industrial Internet of Things (IIoT) protocols into the generated architectures.

Main Results:

• Quantified significant effort savings compared to manual programming through lab-scale evaluation and extrapolation.
• Demonstrated the capability of generated architectures to support diverse IIoT protocols.
• Validated the approach's effectiveness in mitigating challenges in industrial data collection.

Conclusions:

• The model-driven generation approach effectively lowers manual implementation efforts for data collection architectures.
• The approach successfully addresses scientific and industrial challenges, enabling scalable industrial data access.
• This methodology is key to unlocking the full potential of data analysis in Industrie 4.0.