Operations Management System Prioritizing Anomaly Detection Order
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing operations management systems face delays in detecting faults across multiple analyzed systems due to sequential analysis, leading to delayed countermeasures when using invariant analysis methods.
Innovation Solution
An operations management system that includes correlation model storage, analysis order management, and order control mechanisms to prioritize fault detection across multiple systems based on performance value correlations and abnormality scores, allowing for real-time updates in detection order.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If one analysis apparatus carries out invariant analysis sequentially for each of multiple analyzed systems, then investment cost is reduced, but fault detection delay increases
Solution Approach 1:
The patent divides the analysis work into segments by distributing multiple analyzed systems across multiple analysis apparatuses. Each apparatus handles a subset of systems simultaneously, transforming the single-sequential-process into multiple-parallel-processes, thereby reducing total fault detection time while maintaining cost efficiency
Solution Approach 2:
The patent implements dynamic allocation of analysis tasks where the analysis apparatus can flexibly adjust which systems are analyzed in each time period based on detected abnormality scores. Systems with higher abnormality scores are prioritized for immediate analysis, creating a dynamic response mechanism that reduces detection delay for critical systems
2Device complexity
If analysis is performed sequentially for multiple systems, then device complexity is reduced, but productivity decreases
Solution Approach 1:
The patent makes each analysis apparatus universal by enabling it to analyze multiple different analyzed systems. The apparatus can dynamically assign and execute analysis tasks for any system in its managed set, allowing one apparatus to perform multiple functions across different systems, thereby improving productivity without proportionally increasing device complexity
Solution Approach 2:
The patent ensures continuous fault detection by having analysis apparatus operate continuously across multiple time periods. Instead of completing all analysis for one system before moving to the next, the apparatus continuously cycles through analyzing different systems in different time periods, maintaining uninterrupted monitoring and improving overall detection speed
Data Source
AI summary
An operations management system, including a memory configured to store program instructions and a plurality of analytical models respectively used for detection of anomaly in a plurality of targets, and a processor configured to execute the program instructions including an order controller configured to control an processing order of the detection of anomaly performed by the operation management system to be the same as a descending order of score of anomaly of the plurality of targets, and an analyzer configured to detect, in the processing order, anomaly in each of the plurality of targets.