Anomaly Handling Support Apparatus Using Bayesian Restoration Probabilities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly handling methods prioritize handling methods based on past execution frequencies, which can lead to longer restoration times when the number of execution times is small or varies across apparatuses, resulting in suboptimal method selection.
Innovation Solution
An anomaly handling support apparatus that calculates restoration probabilities for handling methods using Bayesian statistics across multiple apparatuses or nodes of the same type, setting priorities based on these probabilities to determine the optimal handling method for anomaly resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If handling methods are prioritized based on past execution frequencies in individual apparatuses, then the system uses simple prioritization logic, but the prioritization accuracy deteriorates when execution times are small or vary across apparatuses
Solution Approach 1:
The patent merges data from multiple apparatuses of the same type into a unified dataset for calculating execution frequencies. By combining historical handling method execution data across multiple apparatuses, the system achieves more accurate frequency statistics even when individual apparatuses have limited data, thereby improving prioritization accuracy without significantly increasing system complexity
Solution Approach 2:
The patent creates a universal prioritization mechanism that works across multiple apparatuses of the same type. The handling method prioritization is not apparatus-specific but rather type-specific, allowing the system to leverage data from multiple instances while maintaining applicability to any individual apparatus, thus improving accuracy without proportionally increasing complexity
2Speed
If handling methods are prioritized based on individual apparatus execution frequencies, then the system responds quickly to local patterns, but the prioritization becomes unreliable when data volume is small
Solution Approach 1:
The patent combines execution frequency data from multiple apparatuses to establish more reliable prioritization. By aggregating data across multiple instances, the system achieves statistically significant frequency calculations that are more reliable, while still maintaining fast response times through efficient data aggregation and processing
3Measurement precision
If the system collects and processes data from multiple apparatuses, then the prioritization accuracy improves, but the data processing complexity increases
Solution Approach 1:
The patent implements a universal data processing approach where handling method execution data from multiple apparatuses of the same type are processed using a common algorithm. This universal processing mechanism improves prioritization accuracy through aggregated data while avoiding the need for complex apparatus-specific processing logic
Solution Approach 2:
The patent uses a standardized data collection and processing template that can be copied and applied across multiple apparatuses. This template-based approach allows the system to efficiently aggregate data from multiple sources without requiring custom processing logic for each apparatus, thereby improving accuracy without proportionally increasing complexity
Data Source
AI summary
In an embodiment of the present disclosure, a prior probability is used by an approach of Bayesian statistics, for all nodes stored in fault handling history data or for a target node for which a fault cause has been identified and all nodes of the same type as the target node, restoration probabilities of handling methods for an identical fault cause are calculated, priorities are set for the handling methods based on the calculated restoration probabilities, and a handling method to be presented is determined in accordance with the priorities.