ML Work Item Routing for Incident Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud computing systems face challenges in efficiently managing and prioritizing incident reports, as they often rely on manual processes that are time-consuming and prone to errors, leading to suboptimal resource allocation and response times.
Innovation Solution
The implementation of machine learning techniques to automatically categorize, prioritize, and assign incident reports using predictive models based on historical data, leveraging virtual agents and chatbots to enhance the incident report management process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual processes are used to manage and prioritize incident reports, then flexibility in handling diverse incidents is maintained, but time consumption increases and response times deteriorate
Solution Approach 1:
The system enables self-service through automated incident report management where the system automatically categorizes, prioritizes, and assigns incident reports without requiring manual intervention. Machine learning models analyze incident data and perform routing decisions autonomously, reducing time consumption while maintaining operational flexibility through configurable parameters and categories that can be adjusted as needed.
Solution Approach 2:
Manual mechanical processes for incident management are replaced with automated machine learning-based systems. The patent substitutes human analysts manually reviewing and routing incident reports with ML models that automatically analyze incident data, determine priority levels, and assign appropriate resources, significantly reducing time consumption while maintaining flexibility through programmable classification rules.
2Measurement precision
If manual prioritization and assignment of incident reports is performed, then accurate understanding of each incident can be achieved, but productivity decreases and response times worsen
Solution Approach 1:
The system creates multiple copies of incident reports and distributes them to different machine learning models for analysis. Each model can independently evaluate the incident and suggest prioritization, allowing parallel processing of multiple incidents simultaneously. This copying approach enables the system to maintain accurate prioritization through model consensus while dramatically increasing productivity by processing many incidents in parallel rather than sequentially.
Solution Approach 2:
The patent implements a universal machine learning framework that can handle multiple types of incident reports across different categories and domains using the same core system. The ML models are designed to be multi-functional, capable of analyzing various incident types (technical issues, service requests, complaints) and performing multiple tasks (categorization, prioritization, resource assignment) simultaneously, thereby increasing overall productivity without sacrificing accuracy in any specific incident type.
3Reliability
If more resources are allocated to incident management, then response quality improves, but costs increase
Solution Approach 1:
The system dynamically changes parameters such as priority levels, resource allocation, and response thresholds based on real-time incident analysis. Machine learning models evaluate incident severity, historical data, and current system state to automatically adjust resource allocation parameters, ensuring high-quality response for critical incidents while minimizing resource expenditure for lower-priority issues, thereby improving reliability without proportionally increasing costs.
Solution Approach 2:
The patent applies partial action by allocating resources selectively rather than uniformly to all incidents. The machine learning system determines the appropriate level of resource投入 for each incident based on its priority and complexity, applying full resource allocation only when necessary for high-priority incidents while using minimal or automated resources for routine incidents. This approach maintains high response quality for critical issues while reducing overall costs through optimized resource distribution.
Data Source
AI summary
Systems and methods for using a mathematical model based on historical information to automatically schedule and monitor work flows are disclosed. Prediction methods that use some variables to predict unknown or future values of other variables may assist in reducing manual intervention when addressing incident reports or other task-based work items. For example, work items that are expected to conform to a supervised model built from historical customer information. Given a collection of records in a training set, each record contains a set of attributes with one of the attributes being the class. If a model can be found for the class attribute as a function of the values of the other attributes, then previously unseen records may be assigned a class as accurately as possible based on the model. A test data set is used to determine model accuracy prior to allowing general use of the model.