Dynamic SLA Proposal via ML Analyst Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current information handling systems face inefficiencies in resolving user application problems due to ineffective prioritization and assignment of problem tickets to analysts, leading to missed service level agreements and reduced user satisfaction.
Innovation Solution
A ticketing system that utilizes machine learning to categorize problems and identify subject matter expert analysts in real-time, based on historical and operational data, to intelligently assign problem tickets, thereby improving service level agreement closure times and analyst efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If problem tickets are assigned to the first available analyst, then assignment speed is improved, but service level agreement closure time increases and analyst expertise mismatch occurs
Solution Approach 1:
The system performs preliminary actions by pre-categorizing problems using machine learning models and pre-identifying suitable analysts based on their expertise profiles before tickets arrive. This allows the system to have ready-made assignment recommendations, reducing both assignment time and closure time while ensuring expertise matching.
Solution Approach 2:
The patent replaces the mechanical first-come-first-served assignment system with an intelligent system using machine learning algorithms. The ML models analyze problem characteristics and match them with analyst expertise profiles, substituting simple sequential assignment with sophisticated predictive matching that optimizes for both speed and effectiveness.
2Device complexity
If manual ticket assignment is used, then system complexity is reduced, but productivity and service level agreement compliance deteriorate
Solution Approach 1:
The system enables self-service by using machine learning models that automatically categorize problems and identify suitable analysts without human intervention. The ML algorithms independently analyze ticket data, compare it against trained models, and generate assignment recommendations, allowing the system to serve itself rather than relying on manual analyst assignment.
Solution Approach 2:
The patent changes the parameters of the assignment system from simple availability-based metrics to complex multi-dimensional parameters including problem categorization, analyst expertise levels, historical performance data, and SLA requirements. These parameter changes enable more sophisticated matching while the automated ML processing keeps the system manageable despite increased complexity.
3Reliability
If re-assignment to another analyst is performed when the first analyst cannot resolve the problem, then problem resolution capability is improved, but loss of time increases due to re-assignment overhead
Solution Approach 1:
The system performs preliminary matching by using machine learning models to identify the most suitable analyst before assignment. By pre-analyzing problem characteristics and comparing them with analyst expertise profiles, the system ensures the first assignment is likely to be correct, reducing the need for re-assignment and minimizing time loss.
Solution Approach 2:
The system implements feedback mechanisms where resolution outcomes are fed back into the machine learning models. This continuous learning process improves the accuracy of analyst-problem matching over time, increasing first-assignment success rates and reducing re-assignment frequency, thereby improving reliability while minimizing time loss.
Data Source
AI summary
Systems and methods for machine suggested dynamic real time service level agreements in operations may include a client information handling system and a ticketing information handling system having a ticketing processor subsystem. The ticketing processor subsystem may receive a problem ticket that may include a problem from the client information handling system. The ticketing processor subsystem may identify a learned problem profile, that may be based on the problem ticket, and learned analyst profiles associated with the learned problem profile. The ticketing processor subsystem may also determine a SLA estimation, that may be based on the learned problem profile and the learned analyst profiles. The ticketing processor subsystem may also send a SLA proposal that may include the SLA estimation and recommended analysts to the client information handling system.