Self-Healing Classification Model for Upgrade Issue Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems lack an efficient method for managing upgrade issues in computing devices, particularly in determining whether an upgrade issue is self-healable and automating the resolution process, leading to increased cognitive burden on operational specialists and prolonged issue resolution times.
Innovation Solution
A method involving a self-healing classification model that synchronizes with a production host environment, allowing for the determination of self-healable upgrade issues and automating the self-healing process, while also updating and retraining the model based on resolution reports to improve prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis and resolution of upgrade issues is performed by operational specialists, then accuracy in determining self-healable issues can be maintained, but cognitive burden increases and resolution time is prolonged
Solution Approach 1:
The system enables self-healing by automatically analyzing upgrade issues using a classification model and executing resolution actions without human intervention. The operational specialist merely initiates the process, while the system autonomously determines self-healability and performs resolutions, eliminating manual analysis time while maintaining accuracy through the trained model.
Solution Approach 2:
The patent replaces the mechanical process of manual analysis by operational specialists with an automated machine learning classification model. The model processes upgrade issues algorithmically, substituting human cognitive effort with computational analysis that operates faster and consistently, thereby reducing resolution time while preserving determination accuracy.
2Measurement precision
If comprehensive analysis of upgrade issues is performed manually, then accuracy in identifying self-healable issues is maintained, but cognitive burden on operational specialists increases
Solution Approach 1:
The classification model performs the comprehensive analysis autonomously without requiring operational specialists to manually evaluate each upgrade issue. The system self-services the analysis function, freeing specialists from cognitive burden while maintaining identification accuracy through the model's trained capabilities.
Solution Approach 2:
The patent substitutes the mechanical cognitive process of manual issue analysis with an automated machine learning system. The classification model handles the complex analysis task algorithmically, replacing human mental effort with computational processing that achieves the same accuracy without the associated cognitive burden.
3Productivity
If a self-healing classification model is implemented to automate issue resolution, then resolution time is reduced and productivity increases, but system complexity increases
Solution Approach 1:
The classification model serves multiple functions: it determines self-healability, guides resolution actions, and can be retrained with new data. This multi-functionality consolidates what would otherwise require separate systems into a single versatile component, increasing productivity while limiting the growth of overall system complexity.
Solution Approach 2:
The system incorporates feedback mechanisms where resolution outcomes are fed back into the classification model for continuous improvement. This feedback loop enables the system to self-optimize over time, maintaining high productivity while managing complexity through iterative refinement rather than requiring increasingly complex architectures.
4Measurement precision
If the self-healing classification model is continuously retrained with resolution reports, then prediction accuracy improves, but computational resources and time are consumed
Solution Approach 1:
The system performs retraining periodically or on-demand rather than continuously, balancing accuracy improvement with computational resource consumption. Resolution reports are accumulated and used to retrain the model at appropriate intervals, maintaining prediction accuracy while avoiding unnecessary continuous computational overhead.
Solution Approach 2:
The retraining process adjusts the model's parameters based on new resolution reports, improving prediction accuracy. By changing only the necessary model parameters rather than restructuring the entire system, the patent achieves accuracy improvements with minimal additional computational resources.
Data Source
AI summary
A method for managing applications includes obtaining, by a client in a customer environment, an upgrade issue report for the application, making a first determination that a resynchronization of a client self-healing classification model with the production host environment (PHE) self-healing classification model is required, wherein the PHE self-healing classification model is stored in the PHE, performing the resynchronization with the PHE self-healing classification model to obtain a synchronized client self-healing classification model, applying the synchronized client self-healing classification model to the upgrade issue report to obtain a state of the upgrade issue report, making a second determination that the state indicates a self-healable state, based on the second determination, performing a self-healing process on the application based on the upgrade issue report, and storing a resolution report based on results of the self-healing process, wherein the PHE is operatively connected to the customer environment.