Predictive Ticketing for IT Systems Using Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current IT ticketing systems rely on manual processes that are inefficient due to user-provided information inaccuracies, delayed issue reporting, and inability to quickly identify root causes, leading to misclassification and delayed issue resolution.

Innovation Solution

Implement a predictive ticketing system using machine learning that extracts features from monitoring data, applies a machine learning model to identify suitable insights, and generates predictive tickets with expected future symptoms and root cause descriptions, reducing the need for user input and improving issue resolution efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If manual ticketing processes are used with user-provided information, then the system can capture user-reported issues, but the accuracy and reliability of issue descriptions deteriorate due to user errors and varying descriptions

Engineering Contradiction:

Improveaccuracy of issue descriptionsVSAvoiduser reporting process

Core Design Contradiction:

Loss of informationVSEase of operation

Solution Approach 1:

The patent replaces manual user reporting with automated machine learning-based monitoring systems that continuously collect and analyze system data. The ML model automatically generates issue descriptions and classifications based on objective system metrics, eliminating user-provided information entirely and its associated inaccuracies.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service through automated monitoring and detection mechanisms that continuously track system health without requiring user intervention. The ML model autonomously identifies issues, classifies them, and generates tickets, making the system self-diagnosing and self-reporting.

Inventive Principle:

Principle #25Self-service

2Loss of time

If user-reported issues are processed manually, then tickets can be created, but the time to detect and resolve issues deteriorates due to delayed reporting and manual investigation

Engineering Contradiction:

Improveissue detection and resolution timeVSAvoidticket processing efficiency

Core Design Contradiction:

Loss of timeVSProductivity

Solution Approach 1:

The patent implements preliminary action through continuous automated monitoring that detects issues before they manifest as user-reported problems. The ML model analyzes system metrics in real-time and proactively generates tickets, enabling preventive maintenance and early intervention before issues impact users.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system replaces manual ticket processing with automated ML-based analysis that rapidly evaluates system data, classifies issues, and generates tickets. This automated pipeline eliminates manual investigation delays and accelerates the entire ticket creation and routing process.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Difficulty of detecting and measuring

If manual playbooks are used for issue resolution, then human operators can address problems, but the ability to quickly identify root causes deteriorates due to distant investigation starting points and complex branching causes

Engineering Contradiction:

Improveroot cause identification accuracyVSAvoidinvestigation time

Core Design Contradiction:

Difficulty of detecting and measuringVSLoss of time

Solution Approach 1:

The patent replaces manual root cause analysis with automated ML-based diagnostic systems that continuously monitor system metrics and trace issue origins. The ML model analyzes correlations between multiple system parameters to automatically identify root causes, eliminating the need for human operators to manually investigate from distant symptom points.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system implements continuous feedback loops where ML models analyze system metrics, generate predictions, and refine their accuracy based on resolved issues. This feedback mechanism improves root cause identification over time by learning from historical data and actual resolution outcomes.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS10740692B2Machine-learning and deep-learning techniques for predictive ticketing in information technology systems

Publication Date: 2020.08.11 SERVICENOW INC

AI summary

A system and method for predictive ticketing in information technology (IT) systems. The method includes extracting a plurality of features from monitoring data related to an IT system, wherein the plurality of features includes at least one incident parameter, wherein the monitoring data includes machine-generated textual data; applying a machine learning model to the extracted plurality of features, wherein the machine learning model is configured to output a suitable insight for an incident represented by the at least one incident parameter, wherein the suitable insight is selected from among a plurality of historical insights; and generating a predictive ticket based on the suitable insight, wherein the predictive ticket includes a textual description of an expected future symptom in the IT system.