IT Case Matching via Activity Flow Attributes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large or complex IT infrastructures, IT incident management is challenging due to the difficulty in identifying and leveraging similar past IT support cases for resolving recurring issues, especially with rich information structures involving collaborative processes.
Innovation Solution
An IT support management system incorporating a similarity computing module and an indexing module to identify similar past IT support cases by comparing attributes such as textual, structural, and metadata information, using techniques like cosine similarity and edit distance, to facilitate efficient case matching and knowledge reuse.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual incident management is used in large IT infrastructures, then IT personnel can address incidents, but the complexity and difficulty of managing incidents increases
Solution Approach 1:
The system enables automatic self-service by automatically matching current incidents with historical resolved incidents based on multiple attributes (title, description, activity flow, metadata), retrieving relevant information, and presenting solutions without requiring manual intervention from IT personnel for routine incidents
Solution Approach 2:
An automated incident management system acts as an intermediary between IT personnel and the complex IT infrastructure, handling incident matching, information retrieval, and resolution suggestions automatically, thereby reducing the direct cognitive load on personnel while managing infrastructure complexity
2Loss of time
If manual search for similar past cases is performed, then IT personnel can find relevant information, but time and resources are consumed
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing historical incident data, activity flows, and resolutions in a structured format, enabling rapid automatic matching and retrieval during actual incident response without requiring manual search operations
Solution Approach 2:
Manual mechanical search processes are replaced with automated computational methods including text similarity algorithms, activity flow comparison, metadata filtering, and machine learning models that automatically identify and retrieve relevant past cases based on multiple attributes simultaneously
3Loss of information
If rich information structures with collaborative processes are used, then comprehensive case information is available, but difficulty in identifying similar cases increases
Solution Approach 1:
The rich information structure is segmented into distinct comparable attributes including title, description, activity flow steps, metadata, and resolution information. Each attribute is processed and compared separately using appropriate algorithms, making the detection of similarities in complex multi-dimensional data more manageable and systematic
Solution Approach 2:
The system transforms complex collaborative process information into structured dimensional representations where each attribute (textual content, activity sequence, metadata) is encoded as a separate dimension, enabling automated comparison algorithms to efficiently detect similarities across the multi-dimensional information space
Data Source
AI summary
A particular case is matched to further cases, where the matching is based on plural attributes contained in the particular case and in the further cases, wherein one of the plural attributes relates to a flow of activities taken to address the respective case.