Design and validation of a flexible framework for developing generic digital twins for Industry 4.0

Area of Science:

• Industrial Engineering
• Computer Science
• Manufacturing Systems

Background:

• Digital Twins (DT) are increasingly vital for optimizing industrial processes.
• Existing architectures often lack flexibility and scalability for multidisciplinary applications.
• Integrating Industry 4.0 concepts is crucial for advanced manufacturing.

Purpose of the Study:

• To present an innovative Digital Twin methodology and prototype.
• To develop a 5-Dimensional Digital Twin architecture combined with Service-Oriented Architecture (SOA) and Industry 4.0 Components.
• To validate the architecture's effectiveness in improving industrial efficiency and enabling predictive maintenance.

Main Methods:

• Developed a Digital Twin prototype separating business logic (Digital Twin Brain) from domain logic.
• Utilized Asset Administration Shell (AAS) Industry 4.0 Components as stateless services for domain functionalities.
• Implemented machine learning algorithms for efficiency improvements and downtime reduction.
• Validated the architecture with real data from an industrial packaging line.

Main Results:

• Demonstrated a flexible and scalable Digital Twin architecture.
• Achieved improved efficiency and reduced production downtime on a packaging line.
• Successfully integrated shared technical functionalities across multiple Digital Twins via AAS.
• Validated the capability for predictive maintenance and seamless integration of additional domain functions.

Conclusions:

• The proposed methodology and architecture facilitate efficient Digital Twin deployment in multidisciplinary settings.
• The architecture supports enhanced industrial efficiency, predictive maintenance, and scalability.
• This work lays the foundation for broader applications within Industry 5.0.