Correlation Model Generation for Operations Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large-scale information and communications systems, existing operations management systems face challenges in detecting performance deterioration and identifying the source of abnormalities, especially when these abnormalities do not exhibit clear signs or are expected to occur in the future, leading to potential service disruptions and increased administrative burdens.
Innovation Solution
An operations management apparatus that generates a correlation model by deriving a correlation function between time series variations of performance information from multiple elements, allowing for the analysis of changes in these correlations to detect performance abnormalities and identify their source based on new performance data not used in the model generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If threshold-based failure detection is used, then clear failures can be detected, but performance deterioration and future failure signs cannot be detected
Solution Approach 1:
The system performs preliminary analysis by establishing correlation models between performance parameters during normal operation. These models enable early detection of performance deterioration trends before they develop into actual failures, allowing preventive actions to be taken in advance
Solution Approach 2:
The system continuously monitors performance parameters and compares actual values against predicted values from correlation models. When deviations exceed thresholds, the system generates alerts and can automatically adjust operations, creating a closed-loop feedback mechanism that maintains service reliability
2Ease of operation
If manual operations management is performed, then flexible decision-making is possible, but administrative burden increases and errors occur
Solution Approach 1:
The system performs self-diagnosis by automatically analyzing performance data, identifying anomalies, and determining failure causes using correlation models. This eliminates the need for manual examination of system logs and parameters, reducing administrative burden while maintaining operational flexibility
Solution Approach 2:
The system replaces manual analytical processes with automated computational analysis. Correlation models and algorithms automatically process performance data, identify patterns, and determine failure causes, substituting human judgment with machine-based analysis that is both efficient and scalable
3Measurement precision
If correlation models are generated from all performance data, then detection accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The system generates correlation models only for performance parameters that exhibit significant correlations, rather than processing all possible parameter combinations. This selective approach maintains detection accuracy for critical parameters while reducing overall processing time and computational resources required
Data Source
AI summary
An operations management apparatus which acquires performance information for each of a plurality of performance items from a plurality of controlled units and manages operation of the controlled units includes a correlation model generation unit which derives a correlation function between a first series of performance information that indicates time series variation about a first element and a second series of performance information that indicates time series variation about a second element, generates a correlation model between the first element and the second element based on the correlation function, and obtains the correlation model for each element pair of the performance information, and a correlation change analysis unit which analyzes a change in the correlation model based on the performance information acquired newly which has not been used for generation of the correlation model.