Natural Language Requirement Refactoring with LLM Prompts

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods struggle to efficiently and accurately translate natural language requirements into a formal requirements language, requiring expertise and multiple iterations to ensure grammatically correct and unambiguous statements, which can lead to errors and inconsistencies.

Innovation Solution

Utilizing a large language model (LLM) neural network with prompt engineering techniques to identify and translate atomic statements, and refactor non-atomic statements into formal requirements language, guided by INCOSE guidelines, ensuring unambiguous and automated generation of requirements language statements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional methods are used to translate natural language requirements into formal requirements language, then expertise and multiple iterations are required, but this leads to errors and inconsistencies

Engineering Contradiction:

Improveaccuracy of requirements translationVSAvoidtime for multiple iterations

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent introduces an intermediary system consisting of a natural language processor and a formal language generator that acts as a mediator between natural language requirements and formal requirements language. This intermediary automatically translates and validates requirements, eliminating the need for manual expert iteration while maintaining high accuracy through structured processing and rule-based validation.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If manual translation of natural language requirements is performed, then grammatical correctness can be achieved, but the process requires expertise and multiple iterations

Engineering Contradiction:

Improvegrammatical correctness of requirementsVSAvoidcomplexity of translation process

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical process of manual expert translation with an automated computational system. The natural language processor and formal language generator use algorithmic processing, pattern recognition, and rule-based transformation to automatically produce grammatically correct formal requirements, eliminating manual intervention while reducing process complexity through automation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs self-validation and self-correction through automated grammatical checking and consistency verification. The formal language generator inherently ensures grammatical correctness by design, and the system automatically detects and resolves inconsistencies without requiring external expert review, making the process self-sufficient.

Inventive Principle:

Principle #25Self-service

3Ease of manufacture

If natural language requirements are translated without automated refactoring, then the process is simpler, but the output may contain ambiguous or non-atomic statements

Engineering Contradiction:

Improveease of requirements generationVSAvoidclarity of requirements statements

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent applies segmentation by breaking down complex natural language requirements into atomic statements through automated refactoring. The system identifies and separates individual requirements, ensuring each statement is independent, unambiguous, and meets formal language criteria. This segmentation maintains ease of generation while dramatically improving clarity and precision of the output.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20250285043A1Refactoring input strings

Publication Date: 2025.09.11 FORD GLOBAL TECH LLC

AI summary

A computer that includes a processor and a memory, the memory including instructions executable by the processor to determine when a natural language requirement statement includes an atomic statement using a large language model (LLM) neural network based on a first prompt statement. The first prompt statement can include a first request to label the natural language requirement statement, a description of a requirements language, and the natural language requirement statement. The first output from the LLM includes an explanation statement which indicates reasons the natural language requirement statement is atomic or not atomic. When the LLM determines that the natural language requirement statement is atomic, the LLM can translate the natural language requirement statement into the requirements language based on a second prompt statement that includes a second request to translate the natural language requirement into the requirements language and the natural language requirement statement.