Requirements Knowledge Graph Verification for Consistent Technical Design
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing requirements engineering processes lack the ability to automatically verify the consistency and completeness of requirements, leading to potential manufacturing issues such as malfunctions, quality problems, and non-compliance with customer or regulatory expectations.
Innovation Solution
A computer-implemented method that generates a knowledge graph from acquired requirements, analyzes it for completeness and consistency, and proposes additional requirements to ensure consistency and completeness, using AI and large language models to guide human interaction and secure the design process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If general-purpose document or spreadsheet editors are used for requirements engineering, then ease of operation is improved, but reliability deteriorates due to inability to automatically verify consistency and completeness
Solution Approach 1:
The patent introduces a specialized requirements engineering software tool that acts as an intermediary between the user and the requirements data. This tool provides automated verification of consistency and completeness through AI-based analysis, while still allowing users to input and manage requirements in a structured manner. The tool mediates between the simplicity of general-purpose editors and the need for automated verification, delivering both ease of use and reliability.
2Reliability
If specialized requirements engineering software tools are used, then reliability is improved through automated verification, but device complexity increases
Solution Approach 1:
The requirements engineering software tool performs automated verification of consistency and completeness using AI-based analysis, enabling the system to self-verify its own data. This self-service capability reduces the need for manual verification by engineers, thereby improving reliability without proportionally increasing operational complexity. The system automatically detects inconsistencies and gaps in requirements, providing reliable verification through intelligent automation.
Solution Approach 2:
The patent replaces manual verification processes (mechanical/systematic human checking) with AI-based automated analysis. Instead of engineers manually reviewing requirements for consistency and completeness, the system uses machine learning models and natural language processing to automatically verify requirements, thereby improving reliability while managing complexity through intelligent substitution of human effort.
3Device complexity
If manual verification of requirements consistency and completeness is performed, then device complexity is reduced, but productivity deteriorates due to time-consuming processes and potential delays
Solution Approach 1:
The requirements engineering software tool performs automated verification of consistency and completeness as a preliminary action before requirements are finalized and passed to subsequent development phases. By conducting AI-based analysis early in the requirements engineering process, the system identifies and flags inconsistencies and gaps before they propagate to design and manufacturing, thereby improving productivity by preventing downstream rework and delays.
4Productivity
If automated AI-based verification is implemented, then productivity is improved by preventing manufacturing issues, but device complexity increases
Solution Approach 1:
The patent implements a feedback mechanism where the AI-based automated verification system continuously analyzes requirements for consistency and completeness, providing real-time feedback to engineers. The system detects issues, communicates them to users, and allows for correction, creating a closed-loop verification process. This feedback-driven approach improves productivity by preventing manufacturing issues while managing complexity through systematic automated analysis and user interaction.
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
A computer-implemented method of designing a technical object comprises: acquiring a plurality of requirements for the technical object under design; generating a knowledge graph from the acquired plurality of requirements; analysing the generated knowledge graph for completeness and/or consistency; based on a result of said analysing, generating and outputting a number of proposed further requirements e.g. to an operator; if the operator accepts one or more of the proposed further requirements, amending the plurality of requirements with the accepted one or more of the proposed further requirements and repeating at least the analysing step; clearing the plurality of requirements for use in subsequent design and manufacturing of the technical object under the condition that the analysing step indicates completeness and/or consistency of the plurality of requirements. Manufacturing of technical objects having malfunctions, quality issues or not meeting expectations is prevented.