Failure Knowledge Coupling for Maintenance Diagnosis Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing maintenance operation assistance systems face a significant burden in creating and updating failure knowledge data, which is necessary for effective diagnosis assistance, due to the need for consistent management and large-scale knowledge extraction, leading to inefficiencies in identifying faulty parts and procedures.
Innovation Solution
A maintenance operation assistance system that reconstructs failure knowledge data by integrating partial instances, using a failure knowledge coupling unit to evaluate and adjust relatedness between different partial knowledge data, and an inspection procedure generation unit to present an efficient inspection procedure to maintenance operators, reducing the need for extensive knowledge management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If failure knowledge data is created with high coverage and consistency, then diagnosis accuracy is improved, but the burden of creating and updating knowledge data increases significantly
Solution Approach 1:
The patent segments the complex failure knowledge data into multiple partial instances, each describing a specific failure mode with its phenomenon, candidate faulty parts, and inspection means. This segmentation allows individual instances to be created and updated independently without requiring comprehensive knowledge extraction for the entire system, thereby reducing the management burden while maintaining high coverage through aggregation of multiple instances.
Solution Approach 2:
The patent implements a feedback mechanism where the system evaluates the relatedness between new partial failure knowledge instances and existing ones, automatically adjusting and integrating them into the knowledge base. This feedback loop ensures consistency and high coverage without manual intervention for every update, reducing the complexity of knowledge data management while maintaining reliability.
2Measurement precision
If comprehensive failure knowledge data is created, then faulty parts can be identified accurately, but the time and resources required for knowledge extraction increase
Solution Approach 1:
The patent applies partial action by creating multiple partial failure knowledge instances, each focusing on a specific failure mode rather than requiring complete knowledge extraction for all possible failures. Each instance provides sufficient information for its specific failure mode, and the collection of instances achieves comprehensive coverage over time without the initial burden of extracting all knowledge at once.
Solution Approach 2:
The patent performs preliminary action by pre-structuring each partial failure knowledge instance with standardized elements (phenomenon, candidate faulty parts, inspection means). This preliminary structuring allows for efficient storage, retrieval, and integration of knowledge instances, reducing the time required for future knowledge extraction and updates while maintaining identification accuracy.
3Reliability
If failure knowledge data is updated consistently, then diagnosis reliability is maintained, but the effort required for data maintenance becomes extremely large
Solution Approach 1:
The patent implements self-service through an automatic evaluation mechanism that assesses the relatedness between new partial failure knowledge instances and existing ones. The system automatically determines whether to integrate new instances or identify contradictions, maintaining consistency without requiring manual review of every update. This self-service approach preserves diagnosis reliability while dramatically reducing maintenance effort.
Solution Approach 2:
The patent introduces an intermediary evaluation mechanism that mediates between new and existing failure knowledge instances. This intermediary layer automatically assesses relatedness and manages integration, serving as a buffer that maintains consistency without requiring direct human intervention in every update operation, thereby improving maintenance efficiency while preserving reliability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided is a maintenance operation assistance system comprising: a failure knowledge database wherein failure knowledge data is recorded; a failure knowledge coupling unit for reconstructing partial failure knowledge data into failure knowledge data; and an inspection procedure generation unit for presenting an inspection procedure using the reconstructed failure knowledge data. The failure knowledge coupling unit evaluates and adjusts the relatedness of nodes among different instances of partial failure knowledge data, and connects the different instances of partial failure knowledge data. On the basis of the reconstructed failure knowledge data, the inspection procedure generation unit sets priorities for when presenting the inspection procedure, and presents the inspection procedure to a diagnostic interface unit in accordance with the priorities.