LLM Conformance Assistant for Process Mining Non-Conformance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current process mining tools are unable to automatically identify and explain the reasons for non-conformance between actual and expected process executions, requiring manual intervention that is time-consuming and knowledge-intensive.
Innovation Solution
A conformance assistant using large language models is employed to analyze process models and instances of execution, generating descriptions of non-conformance and providing recommendations for improvement, including activity name corrections and process enhancements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis is used to determine reasons for non-conformance, then accuracy of identification is improved, but time consumption and operational complexity increase
Solution Approach 1:
An AI assistant acts as an intermediary between the process mining tool and the user. The tool identifies non-conformance cases objectively, and the AI assistant analyzes event logs to generate explanatory reasons, bridging the gap between automated detection and human understanding without requiring manual intervention.
Solution Approach 2:
The system enables self-service by automatically generating explanations for non-conformance cases. The AI assistant autonomously analyzes event logs, compares them against process models, and produces human-readable reasons for deviations, eliminating the need for users to manually determine causality.
2Loss of information
If manual analysis is used to determine reasons for non-conformance, then depth of understanding is improved, but knowledge requirements and operational difficulty increase
Solution Approach 1:
The AI assistant serves as an intermediary that translates complex process mining data into understandable explanations. It processes event logs and process models to generate natural language reasons for non-conformance, making deep analytical insights accessible to users without specialized knowledge.
Solution Approach 2:
The patent replaces the mechanical process of manual analysis with an automated AI-based system. The AI assistant performs the cognitive tasks of analyzing event logs, comparing them against process models, and generating explanatory reasons, substituting human expert analysis with automated intelligent processing.
3Productivity
If automated process mining is implemented, then productivity is improved, but ability to explain non-conformance deteriorates
Solution Approach 1:
The AI assistant provides multi-functionality by combining automated case identification with explanatory analysis. It not only detects non-conformance cases efficiently but also generates human-readable explanations for the reasons behind deviations, serving multiple functions within a single integrated system.
Solution Approach 2:
The AI assistant acts as an intermediary layer between automated process mining tools and users. It receives objective non-conformance identification from the tool and supplements it with explanatory analysis, preserving both the efficiency of automation and the explanatory depth needed for understanding.
Data Source
AI summary
Systems and methods for generating a description of non-conformance using a large language model are provided. One or more prompts defining 1) instructions, 2) a textual description of a process model of a process, and 3) an instance of execution of the process are received. A description of non-conformance of the instance of execution to the process model is generated using a large language model based on the textual description of the process model and the instructions. The description of the non-conformance of the instance of execution to the process model is output.