Incident Remediation Prediction for Low-Severity Recurrence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems for mitigating incident recurrence are slow and resource-intensive, particularly for incidents that do not meet severity thresholds, leading to potential reoccurrence and inefficient use of resources.
Innovation Solution
A machine learning model is used to semantically match historical incident data, predicting remediation actions for incidents that were not initially reviewed by mapping similarities between incidents with and without assigned remediation actions, thereby enabling automated assignment and reduction of recurrence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional systems manually review and assign remediation actions for all incidents, then remediation actions can be accurately assigned, but the process becomes slow and resource-intensive
Solution Approach 1:
The system performs preliminary actions by pre-processing incident data and pre-computing similarity metrics before actual remediation assignment is needed. Historical incident data is pre-analyzed and stored with extracted features, enabling rapid comparison when new incidents occur without requiring full manual review
Solution Approach 2:
The patent introduces an intermediary automated processing system that acts as a mediator between incident detection and manual remediation assignment. This intermediary system uses machine learning models to predict remediation actions and filter incidents, allowing human reviewers to focus only on complex cases while routine incidents are handled automatically
2Reliability
If conventional systems manually review all incidents to assign remediation actions, then remediation actions can be properly assigned, but resource utilization becomes inefficient
Solution Approach 1:
The system enables self-service by allowing the automated processing system to independently handle incident analysis and remediation prediction without requiring human intervention for each incident. The machine learning models autonomously process incidents, extract features, and generate remediation recommendations, freeing human resources to focus on exceptional cases
Solution Approach 2:
The patent segments incident processing into different categories based on complexity and risk level. Routine incidents with clear patterns are automatically processed by the system, while complex or novel incidents are flagged for manual review. This segmentation allows resources to be allocated efficiently based on incident characteristics rather than treating all incidents uniformly
3Productivity
If severity thresholds are used to filter incidents for review, then resource consumption is reduced, but incidents without remediation actions may recur
Solution Approach 1:
The patent replaces the mechanical severity threshold filtering system with an intelligent machine learning-based prediction system. Instead of relying on fixed severity criteria that may miss subtle patterns, the system uses trained models to predict remediation actions and identify at-risk incidents based on learned patterns from historical data, improving detection accuracy while maintaining resource efficiency
Data Source
AI summary
Aspects described herein may use machine learning models to predict one or more remediation actions to mitigate reoccurrence of an incident that has become restored based upon previous incidents of an entity. Historical incident data is compiled into two incident datasets: one representative of incidents that were assigned a remediation action to mitigate reoccurrence of the incident, and a second representative of incidents that were not assigned a remediation action. A machine learning model matches relationships between data in the two datasets and outputs scores representative of similarities. Based on the scores, one or more remediation actions are mapped to an incident in the second dataset and the remediation action is performed for the incident.