Unified Cognitive Root Cause Analysis for Multi-Domain System Maintenance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current root cause analysis systems are inefficient, take too long to identify the root cause of problems, and lack real-time capabilities, often requiring extensive documentation and being domain-specific, non-cognitive, and offline, which hampers timely and accurate problem resolution in various domains.
Innovation Solution
A computer-implemented method and apparatus that uses a historical model database, unsupervised models, linear complexity models, and real-time objective functions to identify and select the most optimal solution for system maintenance by analyzing key performance indicators, determining time-cost and resource-cost complexities, and predicting recurrence frequencies, enabling real-time root cause analysis across multiple domains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional root cause analysis methods are used, then comprehensive problem analysis can be performed, but the analysis takes too long and cannot provide real-time solutions
Solution Approach 1:
The system pre-processes historical problem data and builds a knowledge base of root cause patterns before actual problems occur. This preliminary action enables the system to quickly match current problems against pre-analyzed patterns, providing real-time root cause identification without performing comprehensive analysis from scratch.
Solution Approach 2:
The system creates simplified representations (copies) of complex historical problems and their solutions in a structured knowledge base. When a new problem occurs, the system compares it against these copied patterns to rapidly identify matching root causes, achieving both speed and accuracy.
2Measurement precision
If domain-specific root cause analysis systems are used, then accurate analysis for specific domains can be achieved, but the systems cannot be applied across multiple domains
Solution Approach 1:
The system employs a universal problem representation framework that can model problems across different domains using common structures and patterns. The knowledge base stores root cause patterns in a domain-agnostic format, allowing the same system to accurately analyze problems in IT, manufacturing, healthcare, and other domains without requiring domain-specific customization.
3Loss of information
If offline root cause analysis is performed, then thorough documentation and analysis can be completed, but the solutions are not available in real-time
Solution Approach 1:
The system maintains continuous operation by processing problems as they occur rather than batching them for offline analysis. The automated knowledge base continuously updates with new problem-solution pairs, and the matching engine continuously queries for relevant patterns, ensuring real-time root cause identification without interrupting system operations.
4Reliability
If extensive documentation is required for root cause analysis, then comprehensive problem understanding can be achieved, but the analysis process becomes complex and time-consuming
Solution Approach 1:
The system extracts only the essential features and patterns from problem documentation that are necessary for root cause identification. Rather than requiring complete exhaustive documentation, the system identifies and extracts key diagnostic indicators, automatically comparing them against patterns in the knowledge base to determine root causes, thereby simplifying the analysis process while maintaining diagnostic accuracy.
Data Source
AI summary
A computer-implemented method, system, and non-transitory computer program product for maintaining a system. A domain of the system is identified. Problem information identifying a problem in the system is captured. Key performance indicators are obtained from a historical model database for the identified domain. An unsupervised model is applied to the key performance indicators to identify historical solutions to historical problems that are similar to the problem in the system. A linear complexity model is used to identify potential solutions for the problem as an historical solution for each historical problem that has the lowest combination of time-cost complexity, resource-cost complexity, and recurrence frequency for the historical problem. A real-time objective function is used to select a solution to the problem from the potential solutions. A maintenance operation to be performed on the system to implement the solution and fix the problem is identified.