Semantic Digital Twin for Product Anomaly Cause Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current manufacturing processes face challenges in seamlessly transferring data across lifecycle phases, leading to complexity, synchronization errors, and loss of knowledge, resulting in high costs and variability in product quality due to differing reactions from various actors involved in the fabrication process.
Innovation Solution
A method is introduced to create a semantically linked digital representation of physical products, incorporating design and fabrication features, which transmits and stores anomaly information to identify causes using semantic patterns, leveraging a knowledge graph and cloud computing for analysis, and utilizing interfaces like HMI and virtual reality trackers for data input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data is transferred between various lifecycle phases using current processes, then information exchange occurs, but complexity increases and synchronization errors occur
Solution Approach 1:
The patent merges data from design, fabrication, and service life phases into a single semantically linked digital representation (digital twin). This integration allows all lifecycle data to be accessed through one unified model, eliminating the need for multiple separate data gathering operations and reducing complexity while maintaining information completeness.
Solution Approach 2:
The patent introduces a semantic linking layer as an intermediary between different lifecycle phases. This layer uses standardized semantic relationships to translate and connect data from different sources and phases, enabling seamless information transfer without direct complex interactions between all actors and systems.
2Loss of information
If data is gathered multiple times for the same workpiece along its lifecycle, then comprehensive information is obtained, but time and resources are wasted
Solution Approach 1:
The patent creates a digital representation of the workpiece in advance that is designed to accommodate and integrate data from all lifecycle phases. This preliminary digital model serves as a ready framework that can continuously receive and store information without requiring repeated data gathering operations, thus saving time while maintaining completeness.
Solution Approach 2:
The patent establishes a continuous data flow into the digital representation throughout the workpiece lifecycle. Instead of discrete, repeated data gathering operations, the system continuously integrates new information from design, fabrication, and service phases into the same digital model, eliminating time losses while maintaining comprehensive information.
3Reliability
If knowledge is stored in individual experts, then specialized knowledge is preserved, but scalability is limited and knowledge is lost when experts change position
Solution Approach 1:
The patent creates a digital copy (digital twin) of the physical workpiece that contains and preserves all knowledge about its lifecycle. This digital representation captures expert knowledge in a structured, accessible format that can be replicated and shared across the organization, making knowledge independent of individual experts and enabling scalability when personnel change.
Solution Approach 2:
The patent makes the digital representation a universal knowledge repository that serves multiple functions: storing design information, fabrication data, service history, and expert knowledge. This multi-functional system can be accessed and utilized by any actor in the lifecycle, enhancing both reliability and scalability of knowledge across the organization.
4Measurement precision
If semantic patterns are identified in the digital representation, then anomaly causes are recognized, but computational resources are required
Solution Approach 1:
The patent performs preliminary organization of data into a semantically linked digital representation with structured relationships between design, fabrication, and service data. This pre-structured format enables more efficient pattern matching and anomaly detection, reducing the computational energy required for analysis while maintaining high detection accuracy.
Data Source
AI summary
Various embodiments of the teachings herein include a method for recognizing causes of anomalies in a physical product during the design, fabrication, and/or service life thereof. An example method includes: creating a semantically linked digital representation of the physical product, the representation including design features and fabrication features of the physical product; transmitting information about anomalies from a quality test to the digital representation and storing said information; and identifying semantic patterns to recognize causes of anomalies using the information, the design features, and/or the fabrication features. The information about anomalies is transmitted in the form of attributes of the product. The attributes contain the location and the time point of the anomaly and machine codes and product regions. The product is assigned location coordinates in a spatial coordinate system for specific time points.