Hierarchical Fault Tree Generation for Accurate Component Failure Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fault tree generation methods face challenges in accurately generating fault trees from system to device/component level due to difficulties in modeling system complexity and component configuration differences, leading to inclusion of unrelated failure factors and omissions.
Innovation Solution
A system comprising a causal relationship storage section, a system level fault tree generation section, and a device/component level fault tree generation section that stores and generates fault tree information based on component connection relationships, turbulence in information transmission, and causal relationships, ensuring accurate generation without noise or omission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If MFM information is used as original data for automatic fault tree generation, then the failure factor can be found out automatically, but it requires extensive knowledge engineering techniques and product knowledge to produce the MFM information
Solution Approach 1:
The patent uses existing FTA results from similar products as templates to generate fault trees for new products. Instead of requiring comprehensive knowledge engineering to create MFM models from scratch, the system copies and adapts proven fault tree structures from comparable products, significantly reducing the knowledge engineering burden while maintaining generation efficiency.
2Adaptability or versatility
If designers with limited experience or restrictive knowledge take charge of FTA, then the design work can be divided among multiple designers, but the failure factor cannot be found out immediately, resulting in elongated FTA time
Solution Approach 1:
The system incorporates feedback mechanisms where FTA results from experienced designers are stored and used to improve the automated generation process. The system learns from past FTA outcomes and uses this feedback to refine its automatic fault tree generation, enabling less experienced designers to achieve results comparable to experts while reducing overall execution time.
Solution Approach 2:
The patent introduces an intermediary automated system that bridges the gap between inexperienced designers and expert-level FTA results. This intermediary system uses stored FTA knowledge and similarity matching to guide less experienced designers through the analysis process, providing suggestions and structures that compensate for their limited knowledge while maintaining productivity.
3Productivity
If the configuration of components differs between past failure products and current analysis targets, then the automatically generated fault tree may contain unrelated failure factors or omissions
Solution Approach 1:
The patent applies local quality by selectively adapting fault tree structures based on specific component configurations. Instead of uniformly copying entire fault trees, the system identifies and copies only the relevant portions that match the current product's component structure, while excluding or modifying sections that correspond to different configurations. This ensures high accuracy by tailoring the generated fault tree to the specific local characteristics of each component.
Solution Approach 2:
The fault tree generation process is segmented into modular sections corresponding to different component groups. The system divides the overall fault tree into manageable segments that can be independently generated and validated. This segmentation allows for more precise matching between past and current product configurations, reducing the risk of including unrelated failure factors or omitting critical ones.
4Adaptability or versatility
If new components that have not been handled before are introduced, then new knowledge about the introduced component is required, but this may interrupt immediate finding of the failure factor and require much time for FTA
Solution Approach 1:
The system performs preliminary actions by pre-storing FTA knowledge and component failure patterns in a database before they are needed. When new components are introduced, the system can quickly retrieve and adapt existing knowledge structures rather than requiring immediate analysis. This preliminary preparation significantly reduces the time needed to handle new components while maintaining accurate failure factor identification.
Data Source
AI summary
This failure tree generation device includes: a causal relationship storage unit that stores information indicating a linkage of the causal relationship of defects of respective component parts constituting subjects to be analyzed in a manner such that the information is associated with the connection relationship of the respective component parts; a system-level failure tree generation unit that generates, for each of the component parts and on the basis of component part constitution information indicating the constitution of component parts to be analyzed and component part connection information indicating the connection relationship of the respective component parts, first element information which is information indicating disturbance having occurred in information transfer between the respective component parts and which indicates the relationship between each of the component parts and a phenomenon having occurred on the component part, and generates, on the basis of the respective items of first element information, system level failure tree information indicating the causal relationship of defects of the component parts; and an equipment/component part level failure tree generation unit that searches for the causal relationship storage unit on the basis of the component part constitution information and the component part connection information, that generates two or more items of second element information indicating information relating to a plurality of events connecting to any one first element and indicating the relationship between the respective component parts and phenomena having occurred on the respective component parts, and that generates, on the basis of the respective items of the second element information, equipment/component part level failure tree information indicating the causal relationship of defects of the component parts, by way of a hierarchical structure.