Chronic Network Incident Detection from Repeated Reset History
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication network systems fail to detect and address chronic incidents effectively, leading to repetitive failures, outages, and degraded services due to repeated temporary corrective actions that do not resolve the root cause of the incidents.
Innovation Solution
Implement a method and system that monitors alarms, generates incident reports, and tags chronic incidents based on a history of similar repetitive resolutions, prioritizing them for targeted root cause investigation and automated resolution plans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If temporary corrective actions (reset actions) are repeatedly applied to resolve incidents, then incident resolution speed is improved, but the root cause remains unresolved leading to chronic incidents and network degradation
Solution Approach 1:
The system performs preliminary analysis of incident patterns by monitoring and storing incident data before chronic degradation occurs. The chronic incident detection mechanism proactively identifies repetitive incident patterns and triggers root cause analysis before the network service is significantly degraded, preventing the cycle of temporary fixes.
Solution Approach 2:
The system implements feedback loops where incident resolution data is continuously monitored and fed back into the detection mechanism. When the same incident pattern recurs at a network element, the system learns from previous resolutions and automatically escalates to root cause analysis, preventing repeated temporary fixes and improving long-term reliability.
2Reliability
If manual monitoring and analysis of incident patterns is performed, then chronic incidents can be detected, but system complexity and operational overhead increase
Solution Approach 1:
The system performs self-service by automatically monitoring its own incident data and detecting chronic patterns without requiring manual intervention. The chronic incident detection mechanism autonomously analyzes incident histories, identifies repetitive patterns, and triggers appropriate responses, reducing operational overhead while maintaining high detection accuracy.
Solution Approach 2:
The patent introduces an intermediary chronic incident detection mechanism that sits between the incident management system and network operations. This intermediary automatically processes incident data, identifies patterns, and provides structured information to operators, reducing the complexity of manual monitoring while improving detection reliability.
3Productivity
If automated systems perform reset actions on network elements, then resolution efficiency is improved, but automated systems may not identify appropriate actions for complex chronic incidents
Solution Approach 1:
The system dynamically adjusts the resolution approach based on incident characteristics. For simple incidents, automated reset actions are performed immediately. For chronic incidents detected through pattern recognition, the system dynamically escalates to more complex resolution processes including root cause analysis and multiple action items, ensuring appropriate responses for each incident type.
Solution Approach 2:
The patent segments the incident resolution process into different levels: automated simple resolutions for routine incidents, and structured multi-step processes for chronic incidents. This segmentation allows automated systems to handle routine cases efficiently while directing complex cases to more sophisticated handling procedures, maintaining both efficiency and appropriateness.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method comprises generating an incident report based on an alarm, obtaining a history of prior incident reports associated with the network element, the history of prior incident reports including data associated with a plurality of prior incident reports that were created in response to a plurality of prior alarms triggered at the network element, and each of the prior incident reports comprising a resolution identifier identifying a resolution of a prior incident, determining that the incident is a chronic incident when the history of prior incident reports includes at least a threshold quantity of prior incident reports comprising the resolution identifier identifying that the prior incidents at the network element were resolved based on at least one of a self-clear action or a reset action, and adding a tag to the incident report indicating that the incident report describes the chronic incident.